Andresen J.T.-Linguistics and Evolution - A Developmental Approach-CUP (2013) PDF

Linguistics and Evolution
Evolutionary linguistics – an approach to language study that takes into

account our origins and development as a species – has rapidly developed
in recent years. Informed by the latest findings in evolutionary theory, this
book sets language within the context of human biology and development,
taking ideas from fields such as psychology, neurology, biology, anthropol-
ogy, genetics and cognitive science. By factoring an evolutionary and devel-
opmental perspective into the theoretical framework, the author replaces old
questions – such as “what is language?” – with new questions, such as “how
do living beings become ‘languaging’ living beings?”
Linguistics and Evolution offers readers the first rethinking of an intro-
ductory approach to linguistics since Leonard Bloomfield’s 1933 Language.
It will be of significant interest to advanced students and researchers in all
subfields of linguistics, and the related fields of biology, anthropology, cogni-
tive science and psychology.
Ju lie Tetel A n d re s e n is Associate Professor in the Department of English

at Duke University.
Linguistics and Evolution
A Developmental Approach
Julie Tetel Andresen

University Printing House, Cambridge CB2 8BS, United Kingdom
Published in the United States of America by Cambridge University Press, New York
Cambridge University Press is part of the University of Cambridge.
It furthers the University’s mission by disseminating knowledge in the pursuit of
education, learning and research at the highest international levels of excellence.
www.cambridge.org
Information on this title: www.cambridge.org/9781107650114
© Julie Tetel Andresen 2014
This publication is in copyright. Subject to statutory exception
and to the provisions of relevant collective licensing agreements,
no reproduction of any part may take place without the written
permission of Cambridge University Press.
First published 2014
Printing in the United Kingdom by TJ International Ltd. Padstow Cornwall

A catalogue record for this publication is available from the British Library
Library of Congress Cataloguing in Publication data
Tetel, Julie, 1950–
Linguistics and evolution : a developmental approach / Julie Tetel Andresen.
pages cm.
Includes bibliographical references and index.
ISBN 978-1-107-04224-7 (hardback) – ISBN 978-1-107-65011-4 (paperback)
1. Language and languages – Origin. 2. Human evolution. 3. Information
theory in biology. 4. Anthropological linguistics. I. Title.
P116.T46 2013
401–dc23 2013020831
ISBN 978-1-107-04224-7 Hardback
ISBN 978-1-107-65011-4 Paperback
Cambridge University Press has no responsibility for the persistence or accuracy of
URLs for external or third-party internet websites referred to in this publication,
and does not guarantee that any content on such websites is, or will remain,
accurate or appropriate.
Contents
List of figures page vi
Introduction: Historiography’s contribution to

theoretical linguistics 1
Part I Theoretical considerations 9

1 Language → languaging 11
2 Developmental systems theory 40
3 A twist in the cognitive turn 68
Part II A developmental systems linguistics 101

4 Evolutionary scenarios I: the standard story and the
self-reproductionist script 103
5 Evolutionary scenarios II: the emerging story and the
self-realizationist script 132
6 The ontogenic script begins 165
7 The ontogenic script continues 196
Part III What to do next 227

8 Revisit: Skinner, Chomsky and construction grammar 229
9 Revise: introductory linguistics textbooks 260
Bibliography 281
Name index 298
Subject index 303
v
Figures
1.1 Saussure’s talking heads1 page 17

2.1 Part of an epigenetic landscape2 55
2.2 Interactions underlying the epigenetic landscape2 56
3.1 The dynamics of the boundary3 90
3.2 The dynamics of a living being with a nervous system3 90
3.3 Structural coupling between living beings with nervous systems3 91
5.1 The natural drift of living beings as distances of complexity with
respect to their common origin3 140
7.1 The flow paradigm for Standard Average European 215
1
From Course in General Linguistics, by Ferdinand de Saussure, translated by Wade Baskin.
©1959 Columbia University Press. Reprinted with permission of the publisher.
2
From The Strategy of the Genes. A Discussion of Some Aspects of Theoretical Biology, by
Conrad Waddington. ©1957 Taylor and Francis. Reproduced by permission of Taylor and
Francis Books UK.
3
From The Tree of Knowledge, by Humberto R. Maturana, Ph.D. and Francisco J. Varela, Ph.D.
©1987 Humberto R. Maturana and Francisco J. Varela. Reprinted by arrangement with The
Permissions Company, Inc., on behalf of Shambhala Publications Inc. Boston, MA. www.
shambhala.com.
vi
Introduction: Historiography’s contribution to
theoretical linguistics
What is needed for a twenty-first-century linguistics is an understanding of

language that is inspired not by Descartes but by Darwin. A linguistics inspired
by Descartes is beautiful but static. A linguistics inspired by Darwin is messy
and dynamic. A linguistics inspired by Descartes assumes that communication
occurs and proceeds to explain how it occurs. A linguistics inspired by Darwin
is motivated by the whys: why communication occurs, why a group has the
particular language it has, why on any given occasion an individual says this
or that or nothing at all. A linguistics inspired by Descartes abstracts linguistic
universals away from time and space and lets them disappear into the mysteries
of the genome. A linguistics inspired by Darwin tethers itself to whole bodies
whose feet are on the ground and seeks to understand the possible relationship
between brain development genes such as ASPM and Microcephalin and the
degree of difficulty in learning a tonal language like Chinese or a non-tonal lan-
guage like English. A linguistics inspired by Descartes operates in a framework
where the terms nature and nurture function in familiar opposition, precludes
investigation into the explanatory dimensions of both evolutionary time and an
individual’s lifetime, and accommodates with difficulty micro-variables in the
human genome. A linguistics inspired by Darwin dispels the conceptual chaos
of the nature/nurture opposition and recasts explanations within the framework
of a developmental system that has evolutionary stability.
For the past fifty years, it is fair to say that large parts of linguistic theory
and practice in the United States and elsewhere in the world, along with cer-
tain branches of the cognitive sciences, have been influenced by the Cartesian-
inspired linguistics promoted by Noam Chomsky. Its methodology of analyzing
sentences in isolation is complemented by the equally context-independent
goal of discovering the abstract, universal principles that are said to under-
lie an individual’s knowledge of language. This knowledge or competence is
theorized to be an inborn complex cognitive system that secures the ability
to understand and produce an infinite number of novel sentences. How this
knowledge comes to be inborn is an evolutionary question that has been only
relatively recently addressed. How this knowledge develops is not addressed at
all, because a universal, inborn ability is not open to individual development.
1
2 Historiography’s contribution to linguistics
Although many linguists have parted ways with Chomskyan linguistics and
diverse approaches to our subject matter are flourishing, a comprehensive and
integrated articulation of the messy, feet-on-the-ground, temporally aware
understanding of language as a product of evolution, of cultural history, and of
individual development has yet to emerge.
In the past several decades, researchers in a variety of social and biological
sciences – some with an interest in language, some without – have come to
recognize the importance of context, and at every level, beginning with the
understanding that the observer is not outside of that which s/he is observing
but is, indeed, a part of it. These researchers, working in a framework broadly
known as constructivism, ground their analyses in the richness of circumstance
and the dynamics of development. When it comes to accounting for human
language behaviors, a constructivist account is not concerned to describe a
theoretical entity such as competence or the language faculty but rather to pro-
vide both an evolutionary (phylogenic) and individual (ontogenic) script for
those behaviors. Introducing linguists and language theorists to a constructivist
account of our subject matter in order to open a path to the discipline’s future
is the purpose of Linguistics and Evolution.
On constructivism
The term constructivism is seemingly everywhere these days. Constructivism
has arisen, in part, as a response to what is often called the rationalist-realist
(think Descartes) account of cognition, truth, science and the world. Theorists
involved in articulating a constructivist account of human experience are cog-
nitive scientists, epistemologists, neuroscientists, philosophers, psychologists
and historians of science, and they are interested, in particular, in understanding
and describing the processes and dynamics of cognition. The key idea is that
organisms’ experiences of the world are not prior to and independent of their
sensory, perceptual, motor and manipulative activities, but rather emerge from
or, as it is said, are “constructed by” those activities. When it comes to human
perceptual, conceptual and behavioral experiences, the role of language is seen
to be a crucial part of that which needs to be understood and described.1
1
The constructivism advocated here is not to be conflated with the concerns of social construc-
tionism, which is also known sometimes as social constructivism. Barbara Herrnstein Smith
has pointed out that social constructionism is a critical stance that has been taken by cultural
anthropologists, feminists and gender theorists, among others, in order to challenge frameworks
in which racial categories, gender biases or certain sexual behaviors are deemed to be biological,
natural or even scientifically proven. These cultural theorists typically stress the social side of
human activity, speak of certain practices as being socially constructed, and often acknowledge
deliberate political engagement (2005: 4–5). While there is much to admire in the work of social
constructionists, the set of concerns that animates their projects is different from that which
informs Linguistics and Evolution.
On constructivism 3
The related term developmental also requires a bit of explanation. In a book

otherwise congenial to the approach laid out in Linguistics and Evolution, evo-
lutionary biologist Richard Lewontin offers the following critique of what he
calls developmental explanation in his field: “The development of an individual is
explained in standard biology as an unfolding of a sequence of events already set
by a genetic program. The general schema of developmental explanation is then
to find all the genes that provide instructions for this program and to draw the net-
work of signaling connections between them” (2000: 11). Lewontin – and all of
Linguistics and Evolution – denies this view of developmental explanation, which
is a version of preformationism, the idea that the human sperm contains a micro-
scopic infant folded in a fetal position, such that this already-formed infant sup-
posedly grows larger during fetal development with the mother’s egg providing
only the nutrition for its growth. Lewontin does endorse a constructivist view of
development and asserts: (i) that environment and organism are causally linked;
(ii) that organisms “not only determine what aspects of the outside world are rele-
vant to them by peculiarities of their shape and metabolism but they actively con-
struct, in the literal sense of the word, a world around themselves”; and (iii) that
organisms are in a constant process of altering their environment (2000: 54–55,
emphasis mine). Of course, neither Lewontin nor any other constructivist thinks
that the term constructing means that organisms get to make up the world any
way they want, for they are always constrained by the histories of the particular
phylogenies to which they belong and by the features of the niches they inherit.
Turning to matters of language, the psycholinguist Stephen Levinson in Space
in Language and Cognition endorses a position he calls “partial constructiv-
ism,” in order to account for the differences in various communities around the
world with respect to spatial cognition that correlate with the differences in the
spatial categories available in those communities’ languages. This particular
topic and Levinson’s remarkable contribution to it will be investigated in the
last chapters of this book. For now, it is enough to note that Levinson’s partial
constructivism actually accords well with the (complete) constructivism of this
book, given that Levinson opposes his position to that of a (non-partial) con-
structivism that he defines as one where “language actually introduces coordin-
ate systems that might not otherwise be available at all.” This definition seems
to suggest that language could be something deposited from the outside into
an otherwise empty organism. In addition, Levinson rejects a position he calls
“activation” where “language merely favours, exercises and develops one or
another system, all of which are antecedently available in cognition” (2003:
214), which is a position also rejected in the present study. In other words, both
for Levinson and for the account presented here, the organism’s insides are as
relevant to the constructing process as that which the organism has available in
the environment (in the case of Levinson’s work, the spatial categories avail-
able in one particular language or another).
In addition, the use of the term constructivism here is generally consistent

with Annette Karmiloff-Smith’s neuroconstructivist approach as well as Adele
Goldberg’s constructionist approach to grammar. For Goldberg, the term con-
structionist has at least two meanings: (i) an approach that emphasizes the role
of grammatical constructions which are conventionalized pairings of form and
function; and (ii) an approach that emphasizes that languages are learned, i.e.,
constructed, from input together with general cognitive, pragmatic and pro-
cessing constraints.
Disciplinary chiropraxis
The discipline of linguistics has grown so large in recent decades that profes-
sional training requires specialization by subdiscipline. One such subdiscipline
is linguistic historiography, and it is usually known as a subfield that reaches
broadly into the historical record of the discipline. However, linguistic histori-
ography can also be deployed as a method of theoretical intervention when
the understanding of the discipline’s past is leveraged to open a path to the
discipline’s future. Thus, despite its seemingly backward-looking orientation,
linguistic historiography can be very much a present-tense activity. It can also
be a hands-on activity in that reading into the history of the discipline is a way
of finding the places in the old bones of linguistic theory that need adjusting.
Because, as has just been noted, the study of the subject matter of linguistics
can be undertaken from so many different perspectives – e.g., artificial intelli-
gence, biology, cognitive science, cultural anthropology, evolutionary biology,
neuroscience, philosophy, primatology, psychology, sociology – the discip-
line of linguistics is currently suffering from an embarrassment of riches. As
a result, the theoretical skeleton of twentieth-century linguistics is bearing a
larger load than it was intended to handle. It is now subject to the pressures and
misalignments due to excess weight. In short, it is out of whack and is in need
of chiropractic adjustment.
The major adjustment that is needed involves incorporating into the discip-
line an important tradition of thinking about our subject matter that has thus
far fallen outside the mainstream of disciplinary linguistics. Notably, it is a
tradition that developed in the wake of Darwin. It begins with the American
psychologist William James and continues with thinkers from a variety of
disciplines and national origins including, among others: the Russian semio-
tician V. N. Vološinov, the Russian psychologist Lev Vygotsky, the British
embryologist Conrad Waddington, the American developmental psychologist
Susan Oyama, as well as the Chilean neurobiologists Humberto Maturana
and Francisco Varela. These are researchers who all share a deep interest
in the role of language in their disciplines and have accordingly reflected
in interesting and useful ways on our subject matter. They all share three
Narrative overview 5
general tendencies: (i) an orientation toward our subject matter that can be
summarized, circularly, as languaging as an orienting behavior; (ii) an allow-
ance for the role and place of the individual that allows, at the same time, for
a layered approach to our subject matter, one that engages with a wide variety
of neurological, cultural and even ethical elements, even if the theorists were/
are not interested in exploring those layers themselves; and (iii) an appre-
ciation of how bodies (for the most part human) behave in environments
and engage in feedback loops with their environments and each other. These
theorists foreground the highly circumstanced nature of whatever it is that
they are studying, always with bodies embedded in environments and minds
fully embodied.
They are the problem setters. They are the ones whose works have opened
up new avenues of investigation in their respective fields of inquiry and are
finding increasing application in neighboring disciplines. Other problem set-
ters will appear in the following pages. The purpose is not to promote these
individual theorists as such but to suggest how their approaches to our subject
matter may be worked in an effective way for a twenty-first-century account
of language. At the very least, writing them into our discipline will help to
limber up disciplinary theory and practice. One way to make a place for them
in mainstream linguistic theory is by taking an enlarged view of what counts
as the history of our discipline. This ploy is well known to historiographers,
who are apt to say things like “social parenthood is bestowed by the children.”
Linguistic historiography is a way of choosing who we, as linguists and lan-
guage theorists, want to be now by reimagining our origins.
Narrative overview
Part I engages with theoretical considerations. Chapter 1 undertakes a review
of the twentieth-century disciplinary objects – langue and competence – in
order to argue that however useful they were to the ongoing development of
linguistics throughout the previous century, they are not now capacious enough
to accommodate the wealth of cultural, behavioral and neurological phenom-
ena concerning our subject matter that have enriched linguistic study in the
past few decades. In place of proposing another disciplinary object, the present
study offers instead the term languaging – which denotes an activity and not
a thing – and outlines how it is to be understood in its various neurological,
behavioral and cultural dimensions. Placing languaging front and center in a
linguistic theory also ushers in two new organizing questions, namely: “how
do living beings become languaging living beings?” and “how do we become
the particular languaging living beings that we do?”
Chapter 2 begins with an exposition of developmental systems theory elabo-
rated first and most cogently by Susan Oyama, whose main achievement is to
have eliminated the nature/nurture dichotomy, and ends with the search for
what Oyama would call linguistic interactants. In between, a useful illustra-
tion of developmental systems theory applied to the psychological category
emotion is outlined. It is important to recognize that elimination of the nature/
nurture dichotomy does not mean the elimination of interest in an organism’s
interior, that is, the so-called cognitive side of the languaging equation. Thus,
in Chapter 3, the work of Maturana and Varela is called upon to provide a rigor-
ous understanding of the interior and its relationship to the exterior through the
exposition of their concept of autopoiesis. This chapter also opens up a prelim-
inary examination of languaging and the brain.
Part II outlines the particulars of a developmental linguistics. In order to
account for the first new organizing question: “why is it that (typically) only
human beings become languaging living beings?”, some kind of phylogenic
story must be told. Since quite a number of linguists and language theorists
have been producing evolutionary scripts for human language/languaging, it is
necessary to review various arguments that are currently in play for explaining
how human language/ing came to be instantiated, developed and maintained in
the species. Chapter 4 outlines the more or less standard evolutionary story as it
is accepted in mainstream linguistics, while Chapter 5 draws once again on the
work of Maturana, now in collaboration with Mpodozis, in order to frame the
evolutionary work being pursued in terms of the second neo-Darwinian syn-
thesis that is currently afoot. What is important to note for the second synthesis
is the renewed attention to the importance of behavior in evolutionary theoriz-
ing and the growing conviction that evolution can no longer be understood only
in terms of changes in gene frequencies.
The path is now open to ontogenic explanations in Chapter 6. Here Linguistics
and Evolution has to make good on the challenge to explain how a living
being becomes a languaging living being without falling into explanations
that pit (a notion of necessary) nature against (a notion of contingent) nurture.
Fortunately for the present study, a number of researchers who study children
and language do not stop their story of the living being becoming a languaging
living being at age four or five or even at puberty but understand that the story
continues throughout the lifetime of the languager. Now, accounting for any
one individual’s lifetime of languaging goes beyond the scope of this book.
However, because half the interest of a developmental systems linguistics is
to account for how languaging living beings become the particular languaging
living beings that they do, Chapter 7 necessarily takes up the topic of linguistic
relativity as it pertains, for example, to the cross-cultural experiences of emo-
tion, time, space and color.
Part III addresses what to do next by answering two pertinent questions:
“how did we get to where we are now?” and “where do we go from here?”
Narrative overview 7
Chapter 8 pauses to look back and review how it happened that theoretical lin-
guistics came to the explanatory impasses that it has in the last few decades,
and it does so by showing how B. F. Skinner’s Verbal Behavior can be use-
ful to linguists, now over fifty years after it was buried by Chomsky’s 1959
review. Skinner’s rehabilitation can, among other things, help set the stage for
fully appreciating recent trends in construction grammar. Chapter 9 looks to
the future, in particular to the way linguistics is presented in standard introduc-
tory textbooks, which will often be the first encounter the next generation of
linguists will have with the discipline. Leonard Bloomfield’s Language (1933)
currently serves as the template for all introductory textbooks on the market
today. Certainly, these textbooks now include findings from, say, neurolin-
guistics, syntax and sociolinguistics that were not available eighty years ago.
Nevertheless, the time has passed for updating chapters. What is needed now is
a radical rewriting of the whole.
In sum, Linguistics and Evolution lays out an approach to linguistics that
builds into its theoretical foundations two temporal dimensions, the phylogenic
and the ontogenic. It furthermore sifts its theoretical approach through a third,
historiographic dimension in order to reexamine some of the discipline’s basic
assumptions and presuppositions. The goal here is not to present a complete
account of our subject matter but rather to indicate the corners in linguistic
theory and practice that have already been turned and to present an outline
of a theoretical reintegration that provides the basis for a new introductory
approach to the subject. Weaving the languaging-as-an-orienting-behavior line
of thinking into the discipline and making space for a constructivist account of
our subject matter are designed to facilitate a future for the field in which the
incorporation of the latest findings in the social and biological sciences will be
a matter not of concern but of course.
The disciplinary shifts outlined in the present study call for an accompanying
shift in historiographic perspective – how linguists think about the history of
their discipline. Just as Thomas Kuhn’s The Structure of Scientific Revolutions
(1962) inspired a previous generation of linguists, so Ludwik Fleck’s Genesis
and Development of a Scientific Fact (original German edition 1935) can now
provide an alternative understanding of how disciplines develop and change,
and it is one that should prove useful in the present moment. The contrast
between the two approaches is most readily apparent in their titles, with Kuhn
underlining disciplinary discontinuity in a structured framework and Fleck
emphasizing continuity in an organic framework. It is of note that Kuhn wrote
the Foreword to the 1979 English edition of Fleck’s text and acknowledges
that he first read it in 1949 or 1950. Although the main subject of Fleck’s
Genesis – the story of how the relationship of the Wasserman reaction to syph-
ilis came to be a medical fact – is not the most obvious point of departure for a
reimagined linguistics, one of Fleck’s central points is pertinent at the outset:

that any so-called new approach (Denkstil or thought style) will carry vestiges
of the historical, evolutionary developments of various elements from a pre-
vious approach and that, accordingly, a newer approach is not to be preferred
because it has now, somehow, discovered the truth but rather because it now
makes available connections to other ideas and/or disciplines that practitioners
in the field feel the need or the desire to make.
Part I
Theoretical considerations
1 Language → languaging
Darwin’s Origin of the Species (1859) made an immediate impression on phi-

lologists at the time – in fact, too much of one. A scant six years after its publi-
cation, namely in 1865, the Société de Linguistique de Paris was founded and a
year later formally banned in its bylaws discussion of the origin of language in
its transactions and debates. The London Philological Society followed suit in
1873. The reason for the ban can be found in the words of the most influential
American linguist of the nineteenth century, William Dwight Whitney, who, in
the first volume of the American Philological Association, wrote:
No theme in linguistic science is more often and more voluminously treated than [the
origin of language], and by scholars of every grade and tendency; nor any, it may be
added, with less profitable result in proportion to the labor expended; the greater part
of what is said and written upon it is mere windy talk, the assertion of subjective views
which commend themselves to no mind save the one that produces them, and which are
apt to be offered with a confidence, and defended with a tenacity, that are in inverse ratio
to their acceptableness. (1869: 84)
Despite this caustic view of the state of evolutionary explanations, Whitney
saw no need for the American association to feel bound by the French example.
He declared the topic “legitimate,” even a worthy challenge since “all [the]
fundamental doctrines [of linguistic philosophy] are involved in its solution”
(emphasis mine). Whitney’s endorsement of the topic, however, did not win the
day. The prohibitions set by the societies in Paris and London were effective,
and for the next one hundred years, philologists/linguists dropped the topic and
pursued other matters.1
By the 1970s, the need to understand the languaging network in the context
of human evolution began to be felt, and interest in the topic has increased
yearly. However, two stumbling blocks have arisen in the attempt to offer an
1
Although most contemporary commentators on this subject mention the French and British bans,
none of them mentions Whitney’s approval of the topic and his reason for that approval. It seems
that the prominent French linguist Michel Bréal shared Whitney’s sentiment and is reported to
have said that he could not imagine that “the study of language would retain much relevance and
vitality if this question [of the evolution of language] was removed from its domain” (Aarsleff
1976: 12).
11
evolutionary account of the subject matter of linguistics: (i) twentieth-century

conceptions of language were uninformed by evolutionary considerations; and
(ii) when the subject of evolution returned toward the end of the twentieth
century, many linguists did not think it their job to reconceive their approach
to their subject matter but rather to explain how a particular theory of lan-
guage – the mainstream variety being some iteration of Chomskyan-inspired
grammar – could be accounted for evolutionarily. However, no matter how
earnest the explanation, the evolutionary script for that theory is necessarily
an afterthought. The project of Linguistics and Evolution is to make evolution-
ary considerations part of the conceptual framework from the outset. In order
to make that case, this chapter begins with an account of the two twentieth-
century theoretical objects of study that linguists have inherited and then opens
onto the path of an alternative approach.
A first historiographic review: twentieth-century objects of linguistics

Although linguistic theory and practice is a collective enterprise, it is primar-
ily Ferdinand de Saussure and Noam Chomsky who have provided the vision
for the twentieth-century theoretical skeleton. These two linguists in particular
distinguish themselves by having posited theoretical objects of study – langue
by Saussure and competence by Chomsky – that have provided frameworks not
only for organizing linguistic data but also for determining which data count as
important and/or relevant. Langue and competence are theories about how to
identify and extract for analysis the most useful and interesting data from the
heterogeneous mass of linguistic phenomena that surrounds us. These objects
are designed to give linguists an idea of what they are supposed to study when
they are studying linguistics. Saussure’s and Chomsky’s linguistic theories are
theories about how and where to slice the dense data pie.
Langue and competence are also responses to the anxiety philologists/lin-
guists have felt for the past several hundred years to put their science on a
par with other sciences, in particular the so-called hard sciences. Some call
it physics envy. In the nineteenth century, the philologists were concerned to
find immutable laws that would make their discipline comparable to branches
of learning concerned with permanent and invariable relations, such as math-
ematics, chemistry and physics. These and similar sciences command a certain
respect from some because they are able to make predictions. The study of
sound changes, viewed as a mechanical process by the philologists, seemed to
offer the promise of making the (mere) study of language into a true science,
one that had predictive power. For instance, Grimm’s Law and Verner’s Law
are still known by their nineteenth-century names, and they reverberate with the
status of, say, one of Newton’s laws or the Second Law of Thermodynamics.
In the twentieth century, Saussure, who had been trained as a philologist,
thought his discipline would continue to fall short of realizing its status as an
Twentieth-century objects of linguistics 13
independent science not until it was able to make predictions but rather until it
could secure for itself a well-defined object of study. This object would need to
be unique to the linguistic perspective so that it could keep things tidy by iden-
tifying phenomena that did not seep into territory that could be claimed by a
related discipline, e.g., sociology or psychology. Saussure performed a feat of
conceptual surgery and declared the true object of linguistic science to be the
theoretical construct langue, which is a system of signs, a storehouse of sound-
images. Decades later, Chomsky reoriented linguistics around the notion of
competence, which is the mental system that underlies a person’s ability to
speak and to understand a given language. Discovering the principles of this
complex cognitive system is meant to be the subject matter of linguistics. Early
on, in Aspects of the Theory of Syntax, Chomsky acknowledged that “observed
use of language or hypothesized dispositions to respond, habits, and so on,
may provide evidence as to the nature of this mental reality” – then added a
but-clause echoing Saussure’s desire for linguistics to take its place among
the (respectable) sciences – “but surely cannot constitute the actual subject
matter of linguistics, if this is to be a serious discipline” (1965: 4). In the same
passage, Chomsky notes that his competence is similar to Saussure’s langue.
However, he rejects langue because it is “merely,” as he puts it, a systematic
inventory of items. Instead, he endorses a return to Wilhelm von Humboldt’s
conception of underlying competence as a system of generative processes.
Indeed, these two terms – langue and competence – are not completely
equivalent, given that Saussure’s langue is the (inert) property of the collectiv-
ity, whereas Chomsky’s competence is the (generative) property of the individ-
ual. Nevertheless, it makes sense to speak of them together because:
(i) langue and competence stake out the territory of what can be considered
so-called crucial data which exists in opposition to the territory welter-
ing with so-called peripheral data. The peripheral territory is identified
in Saussure’s notion of parole, i.e., the individual act of speaking, and
in Chomsky’s notion of performance, i.e., the observed use of language,
noted in the passage above, which Chomsky says shows “numerous false
starts, deviations from rules, change of plan in mid-course, and so on.”
Langue and competence are thus and by definition abstractions, in that the
phenomena they pick out from the welter are not actual, i.e., not an act of
parole, not produced in performance;
(ii) langue and competence are both intended to capture the static relations that
exist deep below the surface of linguistic phenomena. The historical ante-
cedents of those relations are not of interest, which means that a temporal
dynamic was not factored into the conceptual creation of either construct.
That is to say that neither langue nor competence was designed to play on
the evolutionary or phylogenic dimension, and constructing a satisfying evo-
lutionary rationale for either of them is proving difficult, if not impossible;
(iii) just as langue and competence are difficult to transfer to the phylogenic
axis, so they are difficult to see working out on and through individual
development, aka the ontogenic dimension. This is because the system
of either langue or competence is considered to be finished as soon as it
emerges in the individual. Saussure’s langue is said to be “deposited” (at
an unspecific age) in the brain. Chomsky’s competence comes complete
in the genome. There is simply no role for individual development, nor
any room for individual differences. Thus, langue and competence are not
simply non-developmental, they are anti-developmental;
(iv) langue and competence both function like slightly different kinds of X-ray
machines examining a corpus of linguistic data, as if to reveal the reality
of the frame supporting all the messy flesh. In this sense, then, the meta-
phor of the skeleton to describe their visions of twentieth-century linguis-
tic theory is apt. Saussure’s and Chomsky’s skeleton theories are, indeed,
quite skeletal, with Saussure’s langue offering a study in slim elegance
and Chomsky’s competence a tour de force of chic anorexia;
(v) langue and competence are the centerpieces of what Linguistics and
Evolution terms a speaker-biased linguistics. Langue is the storehouse of
elements from which speakers assemble utterances and through which they
understand them. Competence is the capacity to generate and understand an
infinite number of novel sentences. They account for speakers’ behaviors or,
rather, the potential for that behavior, and they are taken for granted. Now, the
untheorized and privileged position of the speaker is not a problem exclusive
to Saussure and Chomsky, for we can go back to Edward Sapir’s Language
(1921) and Leonard Bloomfield’s Language (1933) and we will not find a
discussion of the terms speaker and/or listener there either. The nineteenth-
century philologists were concerned to reconstruct protolanguages whose
speakers and listeners were long dead, and so the idea of defining and dis-
cussing them did not arise. Twentieth-century linguists revived speakers and
listeners, at least theoretically, but they did not bother to discuss or define
them. The taken-for-granted focus on the speaker is symptomatic of a lin-
guistics that has long and unfortunately privileged: the oral over the aural;
the auditory over the visual; the spoken over the written; the producer of
the utterance over the circumstances that contribute to the production of the
utterance. As always, what is said of the speaker holds true, in principle, for
the signer of any signed language, such as American Sign Language (ASL).
A speaker-biased linguistics, furthermore, pulls our attention away from the

consequential, non-trivial nature of listening. First, the listener is important
because without one (even one who is in the same skin), we have no reason to
open our mouths to say anything. Bottom line: we speak in order to have effects.
We speak so that certain kinds of things happen and others don’t. “Please pass
the salt.” “When in the course of human events it becomes necessary …” “I’m
feeling sad.” Even when the only reason to speak is supposedly for a listener to
understand, the listener’s understanding counts as an effect. Understanding is
not a passive or empty category. It is as much a doing as is speaking. And it can-
not be taken for granted. Second, and as a summary of the last forty-plus years
of sociolinguistics, listeners affect in the first place what speakers say and how
they say it. Third, not just any creature on earth makes an effective listener. The
capacity to listen is very human. But is it uniquely human? The question puts us
squarely in the midst of the apes-with-language debates that will be reviewed in
more detail in later chapters. For now, an intriguing passage from ape-language
researcher Savage-Rumbaugh and co-authors suggests the gist:
Washoe2 was quite good at asking for a banana, but completely inept at giving one. Indeed,
Washoe was completely inept at listening and giving anything. For Washoe, language had
only one function: making the world do what she wanted. In other words, half of what we
regard as language was missing. For Washoe showed little understanding of what was said
to her. She had been taught to speak, but she had not been taught to listen.
Listening seemed too simple. Listening it was thought, was just sitting there and
hearing what someone else said. Scientists assumed that there was nothing there to
measure. All studies of child language prior to the work with Washoe had looked at
what children said, not at what they understood. Indeed, it was deemed neither possible
nor relevant to look at what children understood. Understanding, if it was thought of at
all, was thought to come along with speaking. (1996: 50–51)
This passage clearly exemplifies what is at stake in imagining, narrowly, that
language is somehow about the free expression of thought. Such a view of lan-
guage, which does not factor in the listener from the beginning, cannot account
for – inhibits our recognition of – the significant differences between the ways
humans respond to language and the ways that non-human primates exposed to
human language, namely those enculturated apes like Washoe (aka chimpan-
beings or humanzees), respond; and, finally,
(vi) langue and competence both exist to guarantee in advance the unasked
question that hovers over all of twentieth-century linguistics:
Q1: How is it possible for two people to communicate?
A1′: They share the same langue.
A1″: They share the same competence.
And these answers are embedded within a theory of language that has a prefig-
ured answer to the following question:
Q2: What is the purpose of communication?
A2′: To make common or to transfer thoughts, feelings, ideas, etc. from a speaker to a
listener.
2
Washoe was a chimpanzee trained in ASL by Allen and Beatrice Gardner in the 1970s.
The idea that language exists for the purpose of transmitting information
from one head to another, as mentioned in the introduction, has been around
since the seventeenth century and John Locke at least. In recent decades it has
been referred to as code theorism. An oft-quoted phrase from Book III, Chapter
XI, Paragraph 5 of Locke’s 1690 An Essay Concerning Human Understanding
goes like this: “Language, being the great Conduit, whereby Men convey their
Discoveries, Reasonings, and Knowledge, from one to another, he that makes
an ill use of it … does … break or stop the Pipes whereby it is distributed to the
public use and advantage of Mankind” (1975 [1690]: 510). For Locke, language
is a “conduit” and when one “makes an ill use of it,” one “breaks” or “stops” “the
pipes whereby it is distributed.” The image of the pipe is particularly vivid.
The British linguist/philosopher Roy Harris has undertaken the most com-
plete critique of code theorism. He characterizes the code-theoretic model of
communication as one where a speaker packages or wraps a message in a code
and sends it through a channel to a listener on the other end who then unwraps
the message. Note that a code does not alter the content of a message, only
its form. Code theorism is also known as the telepathic, telementational, tele-
graphic or duplicative transmission model of communication. Harris points us
to the (in)famous and unanalyzed picture of the talking heads in Saussure’s
Cours, introduced with the spare commentary. It is an excellent example of
what Fleck would call an Ideogramme, namely a picture whose ideas and
meanings often lie below the threshold of articulation and thus directs percep-
tion unconsciously. Saussure’s goes like this:
In order to separate from the whole of speech [langage] the part that belongs to lan-
guage [langue], we must examine the individual act from which the speaking-circuit
can be reconstructed. The act requires the presence of at least two persons; that is the
minimum number necessary to complete the circuit. Suppose that two people, A and B,
are conversing with each other [see Figure 1.1]. (1959 [1916]: 11)
This sketch takes for granted that communication, in the sense of duplicative
transmission, simply occurs – seemingly unproblematically – and that it occurs
by means of the process represented between the two heads, namely by means
of the langue (the code) deposited in the two heads. The concept of the code
has long formed and deformed Western thinking about the nature of language
and the very process of life itself.3 Saussure’s sketch, which has lurked in the
3
Almost from the beginning of modern genetics Conrad Waddington was critical of the transfer of
Shannon-Weaver information theory’s notion of the code to the understanding of the relationship
between phenotype and genotype, as it took shape in the metaphor of the so-called genetic code.
He tells us that in the very first steps of the transition from genotype to phenotype information
theory can still be applied. However, it is in the next stages of the formation of the phenotype that
information theory becomes unable to deal with the situation. Attempts to rescue it have been
made but, for Waddington, they simply do not work. “In sum, information theory in the strict
sense is not a useful tool in considering the relation between the genotype and phenotype” (1975
Figure 1.1 Saussure’s talking heads
linguistic imagination for a very long time, is remarkable for what it does not
include, and those lacks tell us a lot about the assumptions that make code theo-
rism work. Note that the heads are floating in contextless space; they are cut off
from their bodies; no actual speech is represented. The picture is entirely static;
the heads are immobile; they are not affecting one another; no time elapses. The
heads are positioned as equals with respect to one another; their profiles are
symmetrical; they are twins. All idiosyncratic differences have been abstracted
away in order to arrive at an essentialized sameness between these two heads.
The overwhelming idea represented here is that of sharing and commonality:
the speakers share a common langue, they are speaking on common ground,
and they are achieving communication by imparting something common. The
picture is entirely commonsensical: communication is the “making common”
of something traditionally supposed to be thoughts, propositions, ideas, feel-
ings, intentions, any and all other mental states. In addition, in the three-line
explanation of the sketch, Saussure notes that two individuals constitute the
minimum exigible, as he said, for the speech-circuit to be established. Although
he claims two individuals as the minimum, the heads represented in the sketch
have the unspoken effect of suggesting that two is the maximum. Indeed, the
dyadic model of linguistic interaction is so entrenched in traditional linguistics
that a model incorporating more than two people seems to transgress disciplin-
ary grounds to become, say, sociology or anthropology.
Although Chomsky took issue with Saussure’s idea that langue was a social
product passively deposited in an individual’s brain, he did not question the
assumptions behind Saussure’s talking-heads model of communication. He
even took them a step farther into abstraction by eliminating the need for two
symmetrical people floating in face-to-face contextless space by theorizing
the proper object of linguistic study to be the competence embodied in the
[1968]: 213–17). For Waddington, the term genetic code is a misnomer. Nevertheless, the notion
of a genetic code has been a persistent trope in the thought style of much of modern genetics.
idealized speaker–listener. This idealized speaker–listener was thereby relieved

of the need to be represented in company with even one other person, further
reducing the idealized dyad to an idealized monad. Instead of two heads float-
ing in contextless space, in Chomsky’s writings, there are sentences floating in
contextless space. These sentences are not associated with a head. They are not
coming out of any mouths. They are not responses by an idealized speaker to
any given situation, nor are they designed to elicit any subsequent responses
from an idealized listener. They are just there.
In fact, Chomsky ratcheted the discussion up a rationalist notch by saying
that the purpose of language is:
A2″: To serve as an instrument for the free expression of thought.
For Chomsky, the free expression of thought is allied to his notion of creativ-
ity. He distinguishes “the normal creative” use of language from the “usage
of language that has true aesthetic value … what we call true creativity, as in
the work of a fine poet or novelist or an exceptional stylist.” For Chomsky, the
“normal creative” use of language is “the ordinary use of language in everyday
life, with its distinctive properties of novelty, freedom from control by external
stimuli and inner states, coherence and appropriateness to situations, and its
capacity to evoke appropriate thoughts in the listener” (1988: 138). It is pre-
cisely this use of language which is (theoretically, at least) free from control
by external stimuli and inner states that Chomsky deems to be the task of the
linguist to describe. And to what purpose is this use of language put? A rela-
tively recent answer can be found in his statement that “language use is largely
to oneself: ‘inner speech’ of adults, monologue for children” (2002: 77).
A code-theoretic linguistics short-circuits investigation of the many and var-
ied linguistic phenomena that fall outside the supposed purpose of language
conceived in terms of the transmission of information. Now, twentieth-century
alternative models of the speech event do exist. The Russian-American linguist
Roman Jakobson understood that language had more than one function, and
although he imagined a model of the speech event that is code theoretic in fla-
vor (he speaks of the code , of the addresser encoding a message, and of the
addressee decoding it), he was clearly interested in getting beyond the trad-
itional and narrow focus on the referential function of language, the supposed
transmission of information. In his presidential address at the annual meeting
of the Linguistic Society of America in 1956 (published in 1976), Jakobson
presented his model of the speech event, with its six factors and six corre-
sponding functions, a function being defined as a focus on one of the factors.
The address was also, not trivially, the moment when the term metalanguage
was introduced to the linguistic community.
Jakobson identifies the six factors in terms of an addresser who sends a
message to the addressee . This message (e.g., talking about the weather,
what you did last night) requires a context (e.g., the existence of the weather,
the events of last night) and a code (language) fully, or at least partially, common
to the addresser and addressee. Finally, there must be some kind of contact ,
a physical channel and psychological connection between the addresser and the
addressee, which enables them to enter and stay in communication. All six fac-
tors share certain aspects and are not fully separable. For instance, the moment
we articulate some message, the factor of the code comes into play, because the
addresser has already chosen one code or another. Although the factors are not
strictly separable, each factor determines a single, different function.
In Jakobson’s account, all six factors are present in every speech event,
but not all functions will be equally foregrounded. For instance, the so-called
em otive or expressive function is focused on the addresser and orients the
addressee to the speaker’s attitude toward what s/he’s speaking about. The pur-
est grammatical form of the emotive function would be the interjection. An
orientation toward the addressee which instantiates the c o native function
would find its purest grammatical expression in the vocative and imperative.
The phatic function focuses on the factor of contact and serves to establish,
to prolong, or to discontinue communication; to attract the attention of the
interlocutor, or to confirm continued attention. The phatic function could also
be called cocktail party talk. Text messages, for instance, tend to be phatic or
conative. Jakobson later rewrote his presidential address in order to stress the
po etic function, because, as he writes: “Insistence on keeping poetics apart
from linguistics is warranted only when the field of linguistics appears to be
illicitly restricted, for example, when the sentence is viewed by some linguists
as the highest analyzable construction” (1990 [1976]: 72). The poetic function
focuses on the message for its own sake. The poetic function is foregrounded
in political slogans of the type I like Ike, in memorable phrases that ring down
through the ages like veni, vidi, vici, and in advertising jingles. A clever, rela-
tively recent one is Maybe she’s born with it. Maybe it’s Maybelline, where
two-thirds of the product name appears twice before the whole product name
is pronounced. Turning to the metalingual function, Jakobson notes that it
plays a vital role in verbal development and occupies a large place in the verbal
behavior of preschool children. It is present in utterances such as “What does
that mean?” Such a question brings focus on the code.
Although Jakobson identified six functions and stressed one of them, namely
the poetic function, it is the case that today it is still usually the referential
function that is: (i) the implicit primary function in standard theories of mean-
ing, which have long supported the notion that the purpose of language was
to mirror the world; (ii) the assumed primary function in many evolutionary
accounts of language; (iii) often the only function investigated by those who
study children; and (iv) often the only function investigated by those interested
in arguing for or against linguistic relativity.
The insufficiencies of narrowing linguistic investigation to only one function

are clear because it occludes perception of very ordinary linguistic activities,
for instance, that of talking to oneself. It is nearly remarkable that no linguist
seems to have studied such a commonplace phenomenon, to wonder why or
under what conditions one talks to oneself. Chomsky asserts that language is
for so-called inner speech, but he does not seem to be interested to say when
or why. It is as if the linguists left the problem, implicitly, to the psychologists.
Indeed, Vygotsky did study the activity of children talking to themselves, and
he came to understand that this was the child’s way to reorient himself to a
problem in the face of frustration. Vygotsky’s work is, of course, important to
us here since he placed such an emphasis on the developmental dimension and
the role that language plays in transforming child thought into adult thought.
The overall point is that one-on-one, face-to-face interactions are not neces-
sarily the norm in our daily lives, as much as we might like to theorize them
as the norm in some moment(s) in phylogeny and ontogeny.
In sum, both langue and competence seek to establish the invariants among
the many variations. If we define the projects of Saussure and Chomsky to be
the study of the uniformities and permanent or steadily recurring conditions in
human speech generally, then we can see Saussure’s and Chomsky’s interests
to be continuous with those of the nineteenth-century philologists. The diffe-
rence is that in the nineteenth century, the philologists dealt piecemeal with
separate phenomena, while in the twentieth century, linguists concerned them-
selves with the system or whole pattern to which those phenomena belong.
Langue and competence are theories of the systemic or patterned relationships
among linguistic elements, of how linguistic elements hang together in a spe-
cific language. The problem of how those elements have come to hang together
either in historical and/or evolutionary time is not at issue, nor is it of interest
to investigate how the system comes to be instantiated in the development of
the individual. That is to say, both langue and competence are projections of a
philologically inspired understanding of the term language.
Philology/philologism
Philology is more traditionally known as historical and comparative linguistics,
and its purpose is to reconstruct the protolanguages from which the daugh-
ter languages are said to be derived and thus to establish language families.
Philologism, by way of contrast, is what can be called the inadvertent forget-
ting of context. What are the conditions that have led to and that maintain this
forgetting of context in modern linguistics?
William James is useful here, in particular a passage from The Principles
of Psychology entitled “The Sources of Error in Psychology.” James writes
Philology/philologism 21
that “The first of them arises from the Misleading Influence of Speech” and
continues:
Language was originally made by men who were not psychologists, and most men to-
day employ almost exclusively the vocabulary of outward things. The cardinal passions
of our life, anger, love, fear, hate, hope, and the most comprehensive divisions of our
intellectual activity, to remember, expect, think, know, dream, with the broadest genera
of aesthetic feeling, joy, sorrow, pleasure, pain, are the only facts of a subjective order
which this vocabulary deigns to note by special words. (1981 [1890]: 193–94)
It is not difficult to translate this passage into something like: the way we
speak about our subject matter was handed down to us by human beings
who were not linguists, and most of us today employ almost exclusively the
vocabulary of code theorism to describe our interactions, and this code-theo-
retic vocabulary calls the medium of the transfer or transmission of informa-
tion language.
Thus when we engage in or witness an interaction of the following type: A
asks: “Pardon me. Could you tell me the time?” and B responds: “Why, yes, it
is noon”; or if B responds: “My watch says 11:50, but I think it may be about
noon, since I’ve noticed that my watch is running a bit slow lately,” we are
inclined to say that A and B understand one another, that A and B understand
the meanings of each other’s words, that A and B get each other’s message,
that A and B speak the same language. If B were to respond to A’s question
with the words: “The rock of Gibraltar,” we would be inclined to say that B
did not understand A, that B did not understand the meaning of A’s words, that
B did not receive A’s message, that A did not speak clearly or possibly that A
and B were speaking in code, such that “Pardon me. Could you tell me the
time?” might really mean something like “When is the shipment of cocaine
coming in tonight?” and the words “The rock of Gibraltar” really mean “Seven
o’clock.” We simply have no terminology in which to describe the outward
effects of successful or unsuccessful interactions other than with a vocabulary
that secures the success or lack of success of those interactions in terms of
mental events that are inside the heads of the relevant participants and that are
activated by a medium called language. And on the face of it, the success of
those interactions does seem to conform to the way that we, as observers, see
them. Such is the logic of code theorism.
After a discussion of how the absence of a special vocabulary for subjective
facts hinders the study of all but the very coarsest of them, James continues:
Empiricist writers are very fond of emphasizing one great set of delusions which lan-
guage inflicts on the mind. Whenever we have made a word, they say, to denote a certain
group of phenomena, we are prone to suppose a substantive entity existing beyond the
phenomena, of which the word shall be the name. (1981 [1890]: 194)
Here I note briefly that the word language has enjoyed a long run as a sub-
stantive entity in linguistic circles and then quote the conclusion to James’s
remark:
But the lack of a word quite as often leads to the directly opposite error. We are then
prone to suppose that no entity can be there; and so we come to overlook phenomena
whose existence would be patent to us all, had we only grown up to hear it familiarly
recognized in speech. It is hard to focus our attention on the nameless, and so there
results in a certain vacuousness in the descriptive parts of most psychologists. (1981
[1890]: 194)
Now, of all the charges that can be laid against linguists in the last several hun-
dred years, they cannot be accused of having overlooked phenomena or left “a
certain vacuousness” in their linguistic descriptions. Linguists have devoted
themselves to formulating an extensive metalanguage to make patent the phe-
nomena we would otherwise overlook in the absence of such terminology. No,
linguists cannot be faulted now for failing to provide an appropriate metalan-
guage. It is almost as if they had taken up the challenge of the criticism James
leveled, over one hundred years ago, at his fellow psychologists.
So, the pertinent question at this point is: how does it happen that lan-
guage, of all vexatious terms, should be the one that has escaped the scrutiny
of healthy skepticism all these years? Certainly linguists have been careful
to distinguish the so-called ordinary language, a use of the term that refers
to a culturally specific communication system, e.g., the French language or
the Zulu language, from a so-called technical sense of the term, which in the
Chomskyan framework refers to something like an internal component of
the mind/brain, aka internal language or I-language. But, still, the question
remains why linguists are so inclined to hang on to thinking of language as
a thing, even a highly abstract thing. The most available answer comes from
the possibility that code theorism is so entrenched in our ways of thinking
that it keeps a substantive notion of an entity called language in place, one
that requires a definition. However, this explanation fails to explain why lin-
guists neglected to take the easy empiricist shot at eliminating the substantive
nonsense in the language lump that would have undone code theorism in the
process. So we must look for other snares, as James called the psychologists’
errors and confusions.
The next snare on the horizon is hypostasis. To hypostasize is to regard or to
make something (which is not necessarily a thing) into a thing. It is often called
reification, from the Latin root res, “thing.” Think of it as “thing-ification.”
The American linguist Leonard Bloomfield used the term hypostasis to
describe the process of abstracting a segment (a word, part of a word, or a
phrase) from a speech stream, thereby making it available as an object to be
spoken of in its own right, and then using it in a new context. Bloomfield
writes:
we can also mention speech-sounds; talking about a person who lisps, for instance,
someone may say, “I am tired of his eternal yeth, yeth.” The commonest case is hypos-
tasis, the mention of a phonetically normal speech-form, as when we say, “That is only
an if,” or “There is always a but,” or when we talk about “the word normalcy” or “the
name Smith.” We may even speak of parts of words, as I shall speak in this book of
“the suffix -ish in boyish.” Hypostasis is closely related to quotation, the repetition of
a speech. (1933: 148)
Bloomfield’s interest is in how certain segments of the speech stream may
become objectified – in grammatical terms “turned into nouns” – and he is
right to imagine that quoting is the process by which this sometimes occurs.
However, only the first of his six examples shows the process of quotation in
its most direct form. The matter is relatively straightforward: once an instance
of yeth (or any other utterance) has been uttered, it is “out there” in the world
and can be treated like any other object and referred to, i.e., quoted. The case
of if and but is only one step removed, in that those conjunctions serve as short-
hand for an entire phrase indicating some objection, and they are italicized
only because most readers still recognize them as conjunctions, not as nouns.
In any case, Bloomfield is concerned to acknowledge the process of how one
utterance can refer to (parts of) another utterance. He did not yet have the term
metalanguage at his disposal, because Jakobson had not yet introduced it.
Indeed, Jakobson folds Bloomfield’s understanding of hypostasis into his
own treatment of a message referring to the factor of the code, which is pre-
cisely the factor that corresponds to the metalingual function:
When we say, The pup is a winsome animal or The pup is whimpering, the word pup
designates a young dog, whereas in such sentences as “Pup” is a noun which means a
young dog or, more briefly, “Pup” means a young dog or “Pup” is a monosyllable, the
word pup … is used as its own designation. Any elucidating interpretation of words and
sentences – whether intralingual (circumlocutions, synonyms) or interlingual (trans-
lation) – is a message referring to the code. Such a hypostasis – as Bloomfield (1933:
148) pointed out – “is closely related to quotation, the repetition of speech,” and it plays
a vital role in the acquisition and use of language. (1990: 388)
Although it is good of Jakobson to have quoted Bloomfield on quoting, it would
have been better for him to discuss how the metalingual function differs from
quotation. “I am tired of his eternal yeth, yeth” works as an example of one
utterance quoting another, but when Bloomfield himself speaks of words, “as
I shall speak in this book of ‘the suffix -ish in boyish’,” he is far from quoting
anything or “merely” offering a “repetition of speech.” Instead, Bloomfield’s
and other linguists’ (metalinguistic) descriptions function as little platforms
that cantilever them outside any given utterance so that they can gain perspec-
tive on the utterance. Both quotes and metalinguistic descriptions are ways to
handle language talking about language, but their functions are quite different,
with the former looping language into language, and the latter trying to find a
way out of the language loop.
We find ourselves in a bit of a conceptual and descriptive pretzel. This is

tricky terrain, no doubt about it. Perhaps Saussure was right in thinking that
linguistics is different from other sciences; however, even in that event, he
misdiagnosed the problem. That is, it is not the case, as Saussure thought,
that linguistics distinguishes itself from other disciplines such as, say, astron-
omy (whose objects of study are stars and planets) or zoology (whose object
of study is animals) in that it does not have a predetermined object of study
(and, accordingly, will fail to get its footing as a science in the absence of one).
Rather, it may well be the case that linguistics distinguishes itself by the fact
that the objects under analysis (some segment of language) and the eventual
descriptions of those objects (another segment of language) have similar struc-
tures. Linguists have no recourse but to name names, to represent language in
terms of language, to speak and write and even think of it in terms of itself.
In other sciences, the usual method of description does not bear a point-to-
point correspondence with the thing described. The descriptions of physical or
chemical events, for instance, do not resemble the events themselves. In con-
trast, a written report of something spoken and the original issuance of the
words do have an acoustic-graphic equivalence. However, they do not have a
functional equivalence. That is, the conditions surrounding the issuance of the
words (issued somewhere or another for some reason or another) are necessar-
ily different from the conditions surrounding the representation of the quotation
of the issuance of the words (the marks situated on a printed page put there for
some reason or another). In other words, it is all too easy to abstract words, sen-
tences, questions, answers, greetings, etc. from their contexts and then to forget
that the words, sentences, questions, answers, greetings on the page strongly
resemble but are not the original spoken (or written or signed) words, sentences,
questions, answers, greetings, etc. To make matters worse, only one type of
so-called language object can be conveyed with a high degree of efficiency by
marks on a page, namely propositional language, i.e., Jakobson’s referential
function, so our graphic conventions allow us to too easily forget not only the
context of an utterance but also the vocal contours, along with the facial expres-
sions, the gestures and the bodily dispositions of which the utterance is a part.
The philosopher W. V. Quine noted decades ago in Mathematical Logic (1940)
that: “A quotation is not a description, but a hieroglyph; it designates its object
not by describing it in terms of other objects, but by picturing it.”
Upon further reflection, however, the problem in linguistics does not seem
to be unique but is rather one that arises in the social sciences in general, since
these are the disciplines where the observer – the linguist, the sociologist, the
political scientist, etc. – functions within and as a part of the phenomenon
observed: language/ing, a social system, a political system, etc. So, it may
well be the case that there is no such thing as a description or a critique of a
political system that is not itself political. However, it may be that the problem
of hypostasis in the discipline of linguistics is particularly bedeviling because

of the airtight relationship between, for instance, the marks on this page that
you the reader read as “It is four o’clock” and the sound sequence “It is four
o’clock” spoken by an individual in response to the question “What time is it?”
and then interpreted by the questioner who might, in turn, respond by saying
something like “Oh, is it that late already?” or “Thank you.”
This forgetting of context by linguists has led to a general tendency to hypos-
tasize the entire process of that which we are studying. When marks situated on
the printed page take on a life of their own and become objects deemed worthy
of analysis in and of themselves, and when collections of such analyses pile
up, then we get a sense that there really and truly is some kind of larger object
of which these marks are a part. We decide to call that larger object language,
since that is one of those marks on our pages that we have to account for any-
way, and since it conveniently already means that larger object in which and
through which the collections of analyses have meaning. The insufficiencies of
the writing system and a methodology heavily dependent on written records
have encouraged this forgetting.
Not all approaches to our subject matter have inevitably forgotten context. In
a work published not too many years after Saussure’s Cours entitled Marxism
and the Philosophy of Language (1973 [1929]), the Soviet-era semiotician V.
N. Vološinov offered a telling critique of linguists’ tendencies to hypostasize
their subject matter. Just as the American psychologist B. F. Skinner would
later do in Verbal Behavior (1957) and the British philosopher J. L. Austin in
How To Do Things With Words (1962), Vološinov argued for a linguistics that
puts linguistic phenomena firmly in their contexts and makes the verbal inter-
action the basis of analysis. To make his case, he analyzed the two major cur-
rents in linguistic thought of his day and thereby presciently critiqued a large
swath of twentieth-century linguistics. Vološinov described the two major cur-
rents, here severely summarized, as:
Abstract objectivism: a view of our subject matter as a stable system of forms that
appears ready-made to an individual such that individual acts of speaking are fortuitous
refractions and variations or plain and simple distortions of the forms of language that
are necessarily the same from one individual to the next; and
Individualistic subjectivism: a view of our subject matter as an activity, an unceasing
process of creation whose creative force follows the laws of individual psychology.
The most recognizable purveyor of abstract objectivism for us is, of course,
Saussure, who repackaged seventeenth- and eighteenth-century code theorism
for the twentieth century. The founder of individualistic subjectivism is the
philosopher-philologist-diplomat-educator Wilhelm von Humboldt, and we
recognize him as one of Chomsky’s inspirations. Individualistic subjectivism’s
emphasis on language as an activity rather than as a thing is promising but
undercut by the idea that it is somehow a personal possession. So now one can
complain, with 20/20 hindsight, that Chomsky took the worst of both worlds:
he allied abstract objectivism’s hypostasized notion of our subject matter as a
thing with individualistic subjectivism’s notion of it as a personal possession.
Double whammy.
How was this double whammy possible? Or, rather, why did Vološinov’s
cogent and compelling critique of the disciplinary assumptions and method-
ology of linguistics have no impact? What are the possibilities? It might have
been that, given that it was written in Russian, it was a bit out of the way. But let
us remember that Slavic thought in general has had an enormous impact on lin-
guistics, beginning with the Polish linguist Baudouin de Courtenay, who coined
both the terms phoneme and morpheme, the former in conjunction with Mikolaj
Kruszewski; and it continues through the work of Jakobson, who founded the
Prague School of Linguistics in 1928 and came to the United States in 1941. Or
perhaps Vološinov’s book was published at a bad time, both in Soviet and world
history, and did not have the impact it deserved. However, that does not explain
why it made no waves in linguistic circles after Jakobson introduced the text to
English-speaking audiences in the 1970s. Or perhaps the word Marxism in the
title limited its popularity. However, it must be noted that the Danish linguist
Otto Jespersen with his innocuously titled Language: Its Nature, Development
and Origin met the same fate as Vološinov’s in that it did not succeed in launch-
ing a “new attitude with regard to the study of living speech” (1922: 97) that
would begin with an investigation of the child. Neither, for that matter, did the
Prague School, which also emphasized the structural analysis of language in the
process of development, namely children’s language.
So, what happened? I diagnose the problem – predictably – historiograph-
ically. That is, when twentieth-century linguists determined that their own
history was continuous with that of the nineteenth-century philologists, the
disciplinary image was set, and no critique, whether from inside or outside the
discipline, was going to undo it. We come to the third and most important rea-
son – in addition to an uncritical acceptance of code theorism and to a deeply
entrenched tendency toward hypostasizing – why linguistics is in the ossified
state that it is currently in, and that is:
Philological Regression, aka Philologism. When, in the early decades of the twentieth
century, the focus of linguistic activity shifted from historical to structural perspectives
and the nineteenth-century philologists were retroactively dubbed linguists, the bond
between them was secure, and the die was cast as to what would count as linguistic
methodology and purpose.
Nineteenth-century philologists were motivated to reconstruct protolan-
guages as a representation of what the original language must have been in
order to have given rise to the daughter languages said to have derived from
it. Written records of the daughter languages, when available, are valuable
guides, and the older the record, the potentially more valuable. The oldest
written records of an Indo-European language, for instance, are of Sanskrit

and date back, according to some sources, to 1800 BCE and, according to other
sources, to 1000 BCE. Alphabetic Greek comes in second with written records
dating as far back as the eighth century BCE (although Mycenaean Greek,
written in syllabic script, has examples dating as far back as, perhaps, 1400
BCE), followed by Latin, which makes its first written appearances around
the sixth century BCE, then comes Gothic as a representative of the Germanic
languages in the fourth century of our era, and so on. When we take into con-
sideration a now-extinct branch of Indo-European known as Anatolian, then
we can find written records of a language known as Hittite that date back to the
seventeenth century BCE. The written records were often religious texts, e.g.,
the most archaic Sanskrit being that of the Vedas, or political histories, in the
case of Hittite. The exact contents did not matter to the philologists, however,
since they were interested in the records only as monuments of the languages
they were studying, and they could study them precisely as if those monuments
had been created for that kind of study.
The point is that these written records are extremely useful, and they per-
fectly well serve the purposes of scholars who have nothing to do with the lan-
guages they study except study them, namely philologists. However, it seems
entirely possible that twentieth-century linguists could have appreciated the
successes of the philologists in achieving their reconstructive aims and, at the
same time, have imagined a relationship with their subject matter that is other
than the passive understanding of a set of marks on a parchment or on a tab-
let, the conditions of whose production were long lost in the distant recesses
of time. Although there are certainly differences between nineteenth-century
historical and comparative philology and the tradition of twentieth-century lin-
guistics from Saussure to Chomsky and including Bloomfield, the twentieth-
century linguists transferred the nineteenth-century philological understanding
of ancient languages to the twentieth-century study of living languages. No
moment can be identified when this transfer occurred; it had already hap-
pened by the time introductory twentieth-century linguistics textbooks identify
nineteenth-century philology as part of the history of what they are doing. No
authors consciously made the transfer. In fact, linguists at the time were quite
sure they were breaking from philological tradition.
Nevertheless, here we are at the beginning of the twenty-first century, and
linguistics still suffers from being seen – in some quarters, including our own –
as a science of sentences, or in Vološinov’s more trenchant description, a sci-
ence taking as its ultimate realium “the isolated, finished monologic utterance,
divorced from its verbal and actual context and standing open not to any pos-
sible sort of active response but to passive understanding on the part of a phil-
ologist.” We can now see that the twentieth-century decision to rename the
philologists as linguists was not a progressive act to dispel the odor of fustiness
hanging over the nineteenth-century word. Rather, it was a regressive act, one
that acknowledged that a speaker-biased linguistics where the so-called native
speaker produces living monuments, severed from contexts, to be recorded
and analyzed by the linguist was, indeed, methodologically continuous with a
reconstruction-oriented philology where ancient written monuments, severed
from contexts, were deciphered and analyzed by the philologist.
When we de-monumentalize our subject matter and alter our relationship
to it by putting ourselves inside of its dynamics, we find no object language,
not even a highly abstract one, such as langue or competence. Rather, we find
ourselves immersed in a never-ending flow of linguistic interactions. Maturana
and Varela refer to that flow as languaging, a term which solves the dual prob-
lem of de-substantializing our subject matter and of giving metalinguistic
presence to the experience. The individual experience of languaging can be
imagined, preliminarily, as a strand in a temporal braid whose other strands
include embodied motion-emotion and which is further embraided with what
is usually called context, that is, with the languaging experiences of other cur-
rent languaging living beings, as well as with what is usually called culture,
that is, with the historical products of former languaging living beings. The
experience of languaging is undifferentiated, unless we are inclined, in a fur-
ther movement of languaging, to mark distinctions within the activity, so that
we may and often do recognize discrete activities such as responding, stating,
greeting, eavesdropping, rehearsing, instant messaging, gossiping, cramming
for a test, writing a linguistics book, while we are in the midst of other activ-
ities such as eating lunch, standing in line at the bus stop, taking a walk.
As we go about our daily lives, interacting with friends, family and col-
leagues, we tend not to notice what might be called routine interruptions in
the flow of our languaging together, aka misunderstandings. As sociolinguist
William Labov has noted, “there is a tendency for members of the speech
community to underestimate the amount of misunderstanding in their daily
lives. The many mishearings and malcomprehensions that occur in the course
of the day are quickly forgotten.” Labov goes on to observe that linguists
tend to participate in this same bias and thus overestimate “the efficiency of
phonemes, morphemes and syntactic rules that are the subject matter of the
discipline” (1987: 329). A further critique of the psychological status of the
linguist’s phonemes comes from another quarter, namely the distributed lan-
guage movement. Robert Port argues that speakers do not know their language
using a phonological code. Instead, “they get along very well with a far richer
and more concrete representation of speech for storing concrete and abstract
linguistic fragments and chunks” (2010: 45). The upshot is that “despite the
frequent insistence that language is basically a code,” Port’s research demon-
strates that “individual speakers solve … representational problems independ-
ently of each other, and thus differently.” The linguist’s code, which is abstract
and low-dimensional and which exists only as a set of statistical regularities

at the level of the community, dramatically underrepresents the rich, concrete
and detailed resources language users bring to their various languaging tasks
(2010: 52–53, my use of languaging, not Port’s). For instance, a person’s recall
for words read from a list is better when the same voice is reading the words
than when different voices are reading them.
Although both Labov and the distributed language movement help to desta-
bilize a code-theoretic account of our subject matter, we are interested in par-
ticular in Labov’s observation concerning routine communicative interruptions
in order to contrast them with those of a greater, and therefore more noticeable,
nature. That is to say that one of the most obvious distinctions we can make
within languaging is to acknowledge that the languaging behaviors of other
languaging living beings are different enough from our own to prevent a flow
of mutual interactions. In cases where the flow is interrupted, we may attempt
phatic strategies of the type suggested by Jakobson either by asking “Do you
hear me?” or by speaking louder: “I said, ‘Bring me the soap, no, not the towel,
the soap!’” When these strategies do not reveal a problem with the channel, as
Jakobson might say, such that speaking even louder will not produce the desired
results, we may attempt a metalingual strategy by asking “Do you understand
what I mean?” When this does nothing to establish a flow, we have perceived
something more than a misunderstanding and have come upon something more
like a border, one that is either uncrossable or, at least, very difficult to cross.
We tend to want to name these borders. Indeed, languaging behaviors, which
are always historically produced and which, as such, are always embraided with
languaging behaviors recognized as Other, tend to be recognized by names that
are already given, e.g., English, Spanish, Spanglish and so on.
The genius of nineteenth-century philology and twentieth-century linguistics
has been to work precisely at the borderland between languages, at the place
where distinctions are most acutely felt, and the resulting description of that
borderland is an invaluable collective work of art and craft. Whereas the nine-
teenth-century philologists knew they were reconstructing languages that were
not actually used the way they had reconstructed them, the twentieth-century
linguists, in applying the philological method to living languages, came to merge
their professional experience of exploring the borderland with the experience of
languaging living beings. However, either languaging living beings lead rela-
tively borderless existences as monolinguals, except when they are in a foreign
language class, in a foreign country or in an encounter with someone with a dif-
ferent languaging history, or they constantly cross borders on their way from one
language to another as bilinguals, multilinguals or language learners. In all cases,
the languaging living being’s experience is not of the borderland. The merger of
linguists’ descriptions with the experience of the language user did not happen
all at once, of course, but rather over the course of the twentieth century.
Stopping to explore the borderland is a fine thing to do as a linguist, but

not the only thing to do as a linguist. Now, making the description of the bor-
derland part of the experience of the languaging living being is a downright
counterproductive thing to do as a linguist, because although such efforts prod-
uce beautiful machinery, those beautiful machines – langue, competence – are
self-contained and operate on their own outside of any context. It is as if they
just aren’t plugged in, i.e., they don’t connect to any wish, desire, need, reason
for why any individual would be inclined to respond, state, greet, eavesdrop,
rehearse, instant message, gossip, write a linguistics textbook, etc. A devel-
opmental systems linguistics is not, by way of contrast, constrained by being
solely responsible to the information about our subject matter that is contained
in our grammar books. For one, our grammars distort the experience of the
languaging living being by assuming that the borderland metalanguage cap-
tures the essence of what is needed to know about that experience. For another,
for languaging living beings living their lives through one or more languages,
the grammar books also leave too much out of the account – lack the requisite
depth – we need in order to understand our subject matter in the fullness of the
biological, evolutionary, neurological, psychological and sociological research
now available to us.
Some readers may find this terminological shift to languaging straightfor-
ward, even a bit ho-hum. They might say, “No linguist is really deluded into
thinking that language is a so-called thing.” They might invoke Sapir to make
the point that over eighty years ago already, when he saw his Nootka inter-
preter writing his language, he understood that there was nothing between the
flow of phonetic elements in his informant’s head and his transcription of “the
intention of the actual rumble of speech” (1921). Others may be grateful. “Very
nice,” they may say, “I’ve been thinking along these lines but never bothered to
formulate a new term, and it so happens that I have plenty of openings in my
discursive space to incorporate it now.” Still others may find it unbalancing.
They may experience some emotioning upon confronting the term and express
their emotioning, perhaps, in the words: “How am I going to operate without
a technical definition of language, for heaven’s sake?!” Or they might find it
a kind of fudge, “This so-called languaging business is getting us nowhere.
We’re playing terminological games.” Or even, “There is something there in
language that is real. I feel it!”
This is not to deny that there is some such thing called language, but to use
it to refer to anything other than a proper noun, e.g., the Romanian language, is
counterproductive. It is also to notice that the job at hand: (i) is not necessarily
to provide a more or less adequate understanding of certain salient terms, e.g.,
language, that we have inherited; and (ii) is to understand the network of ideas
in which these terms find their place and then to reconceptualize that network,
if need be, which effort necessarily requires some new languaging in order to
Languaging as an orienting behavior 31
bring such a reconceptualization into existence. My intervention in the history

of the discipline at this particular moment in time can be seen as a strategic
tweak that has as an intended consequence the elimination of the universalist
essence clinging to the language lump that often bedevils linguists’ interac-
tions, for instance, with biologists, who by comparison operate quite specific-
ally (seemingly) without essences of any kind.
Languaging as an orienting behavior

Most linguists are willing to readily acknowledge that what they call lan-
guage has no existence, that it is never a so-called finished fact and that struc-
tural descriptions are idealizations and distortions of always-moving targets.
Nevertheless, we linguists are in a very old habit of describing our subject
matter as a thing made, and our grammars and our dictionaries further conspire
to make us think of our subject matter as an accomplished fact. We induct our
students into the profession with textbooks that reinforce the old habit. We
fight against our old habit, and we regularly remind one another – warn one
another! – not to reify our subject matter. Creolist Salikoko Mufwene notes,
almost as an aside, that: “a language is an abstraction which linguists should
not overly reify” (2001: 17). It would be great to alter “not overly reify” to “not
reify at all,” but the term languaging, like every other term that ever was or will
be, is as much a hypostasized fiction as the term language.
Nevertheless, the shift from noun to verb, from language as an abstraction
of some sort to languaging as an activity helps to facilitate a welcome shift in
point of view. The term languaging focuses our attention away from descrip-
tions produced by an observer (the linguist) and toward processes experienced
by those engaged in the processes in statu nascendi (languaging living beings).
A focus on languaging furthermore encourages us out of the explanatory frame-
work of code theorism.4 Interactions among languaging living beings can no
longer be described as transferring messages or information; there are no more
inputs or outputs, just as there are no codes, no encoding, no decoding, no
more speakers and no more listeners. The term languaging steers us toward a
framework in which languaging behavior is understood as an orienting behav-
ior. As languaging living beings, we experience continuous orientation with
respect to the cognitive domains of other languaging living beings, and that
orientation will occur separately for each of us, according to our own particular
histories of interactions. Given the relatively brief history of interactions that
we (you the reader and I) have established thus far in this first chapter, it is not
4
Because my effort is to leave the theoretical presuppositions of structuralism behind, I do not
find using the Saussurean term langage, “speech,” to be a serviceable alternative to languaging.
Saussure himself dismissed langage as an object of study because “we cannot put it into any
category of human facts, for we cannot discover its unity” (1959 [1916]: 9).
necessarily to be expected that you will be oriented enough within the inter-
pretive framework suggested here for this paragraph to make perfect sense to
you at this point. In his “A Short Essay on Languaging,” the American cultural
anthropologist Alton Becker puts the matter thusly:
This history of our particular interactions, oral or written, builds for each of us a domain
of discourse, a constantly changing – drifting – domain of discourse in which we live
and have an identity. It is obvious, I think, that my memory is not your memory, and it
is out of my memory (what I have called my own prior texts) that I speak and under-
stand your speaking. It is within this domain of discourse that my shifting orientations
take place. It is my world. It was not, however, created by me alone but rather by my
recurrent, multiple interactions with others. We share worlds, say Maturana and Varela,
because we specify them together in our actions, in our languaging together. (1991:
230)
The central point of languaging is in the last sentence of this passage: to the
extent that we share cognitive domains, we have specified them together in our
languaging together.
Most significantly, the term languaging accords better with the Darwin-
inspired tradition of thinking about our subject matter that begins, more or
less, with William James, who was the first to consolidate this particular
understanding of languaging. James does not use the term languaging, by the
way; he speaks instead of the “word” in “its steering function” (1981 [1907]:
84) and believes that large tracts of human speech are nothing but signs of
direction in thought. The minor tradition achieves its clearest formulation and
widest elaboration in the early work of Maturana and Varela together and the
later work of Maturana and Varela separately, specifically as they develop
their conception of autopoiesis. But now we are a bit ahead of ourselves. (See
Chapter 3.)
As linguists and language theorists consider incorporating the term languag-
ing into a twenty-first-century linguistics, it is useful to note that new terms
are not always welcomed as apt, useful or necessary. The American cultural
anthropologist Clifford Geertz, for one, was not impressed with the move to
transform well-known concepts into new parts of speech. In an essay enti-
tled “The Growth of Culture and the Evolution of Mind,” Geertz takes issue
not with language/ing but with mind/minding: “One of the most frequently
suggested methods for rehabilitating mind as a useful scientific concept is to
transform it into a verb or participle … But this ‘cure’ involves falling in with
the school bench story that a ‘noun is a word that names a person, place, or
thing,’ which was not true in the first place” (1962: 715). He objects to the shift
for three reasons. First, it opens the door to other new forms, many of which
are awkward-sounding. Second, it does not necessarily secure the conceptual
effects that it is supposed to. And, third, the term mind serves a purpose that is
not achieved with the new term minding. He writes that:
mind is a term denoting a class of skills, propensities, capacities, tendencies, habits; it

refers in Dewey’s phrase … to an “active and eager background which lies in wait and
engages whatever comes its way.” And, as such, it is neither an action nor a thing, but
an organized system of dispositions which finds its manifestation in some actions and
some things. (1962: 716)
In other words, the term mind continues to have a place.

It is worth considering Geertz’s objections. The third objection is the easiest
to deal with. The term language already serves a very useful purpose in des-
ignating a conventionally recognized repertoire of languaging activities, e.g.,
the English language. The first objection may have some weight – or, rather, it
may have had some in 1962. Geoffrey Nunberg’s fun essay from 1997 entitled
“Verbed Off” tells another story. Nunberg observes that a language like English
easily shifts nouns to verbs and verbs to nouns, often without anyone noticing.
For instance, the verb to telephone was added only four years after Alexander
Graham Bell patented his new invention, and no one subsequently batted an eye
at to radio, to FedEx, to fax, to keyboard, etc. Very recently, and in the other
direction, people are now talking about fellowshipping, scrapbooking and even
sweatering, although people have been knitting sweaters for thousands of years.
As Nunberg puts it: “But maybe [turning nouns into verbs] is no less than our
active lifestyles deserve. That’s what modern life feels like, just one damn thing
after another – breakfasting, carpooling, officing, keyboarding, Stairmastering,
journaling, bedrooming. Why shouldn’t we wordsmith new ways to talk about
it?” (2001 [1997]: 153). Since languaging is woven through all our activities all
day long, we may as well language about it as such. Verbing the noun language
thus puts linguistics in sync with the English of the moment and hence with cur-
rent (English-speaking) languagers’ experiences.
But what of Geertz’s second objection? Following James, it seems best to
secure the conceptual effects of leaving one epistemological framework and
entering another with new terminology. The last fifty years of linguistic theor-
izing about language steers our thinking toward a dress covering the body of
the mind. In this framework, all the dresses, i.e., languages, in the world may
look different, but they are all understood to be cut from the same cloth, i.e.,
they are products of and covers for a universal mind and language as a system
that is everywhere the same. More precisely, they are projections of the human
mind which is, indeed, The Human Mind, common to all. One of the more
recent pronouncements by Chomsky is:
Going back to language, what you have is a system that is, as far as we know, essentially
uniform. Maybe there was some speciation at one point but only one species survived,
namely us; there seems to be no variation in the species. True, we find Williams’ syn-
drome and Specific Language Impairment. But that’s not variation in the species in any
meaningful sense: those are deviations from the fixed system that occur now and then,
but the basic system seems to be uniform. (2002: 147)
For Chomsky, Williams’ syndrome and Specific Language Impairments (SLIs)

are not seen as possibilities in human variation generally speaking, but rather
as not counting as variation “in any meaningful sense.”
By way of contrast, in Linguistics and Evolution, languaging is understood:
(i) to scaffold the cognition of languaging living beings and to be an activity

through which certain cognitive and conceptual developments are induced.
In other words, the term languaging suggests deep-down cognitive as well
as pervasive cultural effects that the term language is not currently theo-
rized to have. One of those deep-down cognitive effects of languaging
pertains to the ways different frames of reference – intrinsic, relative and
absolute – organize the spatial domain and thus literally orient languagers
to particular spatial relations on the horizontal axis. Levinson’s Space in
Language and Cognition (2003) demonstrates convincingly how deeply
different orienting frames of reference penetrate non-linguistic cognition
and affect significant aspects of cultural life;
(ii) to span the troublesome distinction between the terms language and
speech that some theorists maintain. This embrace of the language–speech
continuum expands the brain part of the brain–culture orientation of point
(i), above, and highlights the importance of the basic sensory, perceptual
and motor abilities that support languaging activities but are not speech
specific. Their functioning comes crucially into view when they mal-
function. For instance, individuals who are impaired in identifying and
producing rapidly successive stimuli – elements occurring within tens of
milliseconds embedded in ongoing speech, which is a critical time frame
in terms of phonetic contrasts – will have various components of their
languaging behaviors disrupted; and these disruptions extend to all their
languaging behaviors, including reading and writing (Tallal and Benasich
2002). Similarly, people with developmental stuttering have problems
with motor tasks other than speech (Packman et al. 2007). One upshot of
this general point is that the term Specific Language Impairment is a mis-
nomer, since the deficits of most SLI children are not specific to language.
More recently, the term language learning impairment (LLI) has been
gaining currency to reflect that more global visual, auditory and motor
problems can disrupt language learning (Tallal and Benasich 2002);
(iii) to allow, precisely, for its multimodality. This means that, even in the
processing of speech, languaging is not solely an auditory activity. Two
classic studies bring attention to the importance of vision in the ordinary
linguistic processing of speech. McGurk and MacDonald (1976) show the
degree to which visual information can affect auditory perception, while
the work of Tanenhaus et al. (1995, 2000) stresses the role of vision in the
processing of syntax;
(iv) to include, crucially, gesture as a dimension of its multimodality. The dra-

matic case of IW comes into play here. At age nineteen, IW lost all sense
of touch and proprioception below the neck. That meant that he lost all
motor control that depends on bodily feedback. He slowly relearned to
walk and eat and do everything else, not by his lost proprioception and
spatial position sense, but rather by using cognition and vision. Of inter-
est is that his speech and gesture (in the service of languaging) were not
lost, while instrumental gesture was. That is, he cannot pick up a brick if
he cannot see his hands. However, he can and does gesture while speak-
ing, even if he cannot see his hands (Goldin-Meadow 2003: 242; McNeill
2005: 234–45). As for people who cannot see at all, namely the congeni-
tally blind, they also naturally gesture, even when they know they are
speaking with another blind person (Iverson et al. 2001). Furthermore,
memory loss interrupts speech and gesture jointly; it is not that “gesture
is a ‘gap-filler’ when speech fails” (McNeill 2012: 13). Following in the
wake of McNeill’s and Goldin-Meadow’s groundbreaking work on co-
speech gestures, researchers have found growing evidence of the import-
ance of gestures in speech processing and comprehension (Morsella and
Krauss 2004; Dick et al. 2009; Holle et al. 2010);
(v) to ease open the flattened space of the view of language held by those who
imagine it to be a mapping between meaning and sound or, in updated
Chomskyan terminology, between the sensory-motor and conceptual-
intentional systems. This point furthers (i) and (ii), above, and is made
in order to give linguists and neurobiologists the sense of a more four-
dimensional behavior than the one given by the Chomskyans who pre-
sume language to be a two-dimensional, static, one-to-one set of lines
drawn between a pre-given sensory-motor system and a pre-given concep-
tual-intentional system. The investigation of the ways that these dynamic
systems intersect, interact and affect one another is a significant part of the
project of a developmental linguistics. Michael Tomasello recognized the
insufficiencies of this flattened space when he requested, with some exas-
peration, “Could we please lose the mapping metaphor please?” (2001).
In losing the mapping metaphor, the linguist opens onto the understand-
ing of the spatial – that is to say, imagistic – dimension of languaging that
McNeill’s language-gesture studies so clearly bring to the fore;
(vi) to orient the linguist and language/ing theorist to the over-richness of con-
texts, as J. L. Austin has taught us, in which all languaging occurs as well
as to the underspecification of utterances in the sense that they can never
fully convey all that is meant. As far as the underspecification of utterances
is concerned, what is taken to be meant always exceeds what is said. This
is sometimes called the principle of ineffability. As for the over-richness
of contexts, this would mean at the most mundane level that, for any
given speech event, while it might be true that the sun is shining, it is four
o’clock and three chairs are present, none of these (or any of the almost
infinite other aspects of the context of a speech event) may rise into pertin-
ence. Nevertheless, those contextual resources are present and available;
(vii) to underscore the idea, given the over-richness of contexts and the under-
specification of utterances, that each languaging living being will respond
differently to these conditions and will thus be oriented differently in his
or her cognitive domain in any given languaging situation. This is so des-
pite the fact that languaging living beings with long histories of interac-
tions often manage to coordinate their behaviors very well. This is also so
despite the fact that certain languages may tend to reliably induce certain
cognitive and conceptual developments rather than others. The point here
is that the brain organization that is created by and supports languaging
activities (as well as other higher cognitive functions) may vary widely
across individuals and in “idiosyncratic patterns that are as unique as
their finger prints” (Elman et al. 1999: 248). This is to note that all vari-
ation is meaningful, in the sense of “counts”;
(viii) to foreground the intersubjective nature of the activity, such that languag-
ing is not a personal possession; and
(ix) to be extendable to the term languaging living being, which is meant to
replace the misleading terms speaker and listener and to move us out of
the code-theoretic framework in which they operate. The term languaging
living being is maximally flexible in that it does not specify a particular
role, disposition or activity of the languaging living being at any given
time, e.g., if a languaging living being is eavesdropping, then it can be said
that the languaging living being’s living is embraided with the activity of
eavesdropping. If the term languaging living being is deemed to be too
long, then the term languager may be used, although it has less of a ring.
The term languaging living being facilitates the incorporation of under-

standings that have come from discourse analysis and that disrupt long-held
assumptions and expectations regarding participation in linguistic interac-
tions. Take the sociologist Erving Goffman’s work on what he termed footing,
which refers to a change in the alignment taken among the participants in an
exchange expressed in the way that the production or reception of an utterance
is managed. Mining this idea, sociolinguist Barbara Johnstone notes that “con-
ventional ways of talking about discourse roles sometimes collapse several
footings and make them difficult to disentangle” (2002: 121). For example, it is
all too easy to collapse the roles of the principal (the person or group who has
decided what to say and is responsible for its having been said), the author (the
person who planned the actual words) and the animator (the person who wrote
down or spoke the words) into one. When a speechwriter writes a speech for a
politician, the politician is the principal, but the speechwriter is the author. If
the speech is read aloud by a spokesperson, the spokesperson is the animator
but neither the author nor the principal. “Because, in the Western tradition,”
Johnstone writes, “we imagine that the people who write novels are typically
principal, author, and animator, it is easy for us to miss seeing how all novelists
appropriate and rework ideas and words that are already circulating” (2002:
122). The term languaging living being is neutral and stands in contrast to the
way that the term speaker is heavily loaded to consolidate the intersection of
the three roles.
By uncoupling the study of our subject matter from a focus on the speaker,
previously overlooked phenomena can come into view. For instance, although
it is widely believed that children cannot learn a language by merely overhear-
ing it, newer studies suggest that adults who had regularly heard their heritage
language during childhood but spoke only the majority language were able to
acquire a more native-like accent when they came to learn the language later in
life than those adults who had had no exposure to the target language as children
(Au et al. 2002). Another study demonstrates how second-born children at age
21 months and 24 months have more advanced pronoun production than first-
born children at those ages, although their general language development is not
different. The idea here is that “children may discover the relations between pro-
nouns and speech roles more easily from overheard speech and be more likely to
make correct generalizations about the meaning of pronouns” (Oshima-Takane
et al. 1996: 623). These findings bring into question how much of a language
can be acquired with little or no corrective feedback, and they open up a space
for investigating how pervasive languaging is in our lives, that is, how much
of our lives are lived in a linguistic bath, even when we are not involved in the
active production of an utterance or even active participants in a speech event.
Overhearing is as much a linguistic activity as any other, a very particular one
whose characteristics and consequences have only begun to be explored.
An orientation toward languaging also helps us reinterpret not only the mas-
sive neuro- and psychological literature that now exists but also the equally
large amount of research on children and language that has been produced.
Instead of thinking of language in terms of taking information from the out-
side world in, languaging opens up not only a developmental space within the
individual but also an intersubjective one where vocal and non-vocal activity
can be seen as continuous. A decade and a half ago, Steven Pinker could con-
fidently state that “grammar development does not depend on overt practice,
because actually saying something aloud, as opposed to listening to what other
people say, does not provide the child with information about the language he
or she is trying to learn” (1994: 280). Now, however, there is plenty of research
to suggest that practice is important; children can and do use their own produc-
tions as the basis for further development (Nelson 2007).
Goldstein and West (1999) similarly challenge the traditional view of articu-
latory development as being a matter of vocal development that relies on cog-
nitive maturation. They offer instead a model of vocal development that relies
on both proprioceptive and social feedback. Going even further beyond the lis-
ten-to-learn model, Goldstein, King and West (2003) show how babbling both
regulates and is regulated by social interaction. They thereby expand on the
traditional view that foregrounds imitative learning when they demonstrate the
(perhaps surprisingly) robust influence of non-auditory social stimulation on
infant babbling. For too long, linguists have overlooked not only the effect of
the total social environment on aspiring languagers but also aspiring languag-
ers’ influences on their environments which, in turn, affect their development.
Decades before Goldstein et al., Condon and Sander published an intriguing
study that revealed a complex interaction system in which the organization
of the neonate’s full-body motor behavior “is entrained by and synchronized
with the organized speech behavior of adults in his environment.” They showed
how the infant “moves in precise, shared rhythm with the organization of the
speech structure of his culture [and] participates developmentally through
complex, sociobiological entrainment processes in millions of repetitions of
linguistic forms long before he later uses them in speaking and communicat-
ing.” The idea is that the richness and complexity of language behavior devel-
ops on the basis of an organized response of the infant to the organized pattern
of adult speech, making infant motor organization the operational format for
later speech. The study made passing reference to the insufficiencies of the
then “contemporary alternative of transformational grammar” and concluded
by saying that “the ‘bond’ between human beings should be studied as the
expression of a participation within shared organizational forms rather than as
something limited to isolated entities sending discrete messages” (Condon and
Sander 1974: 99–101). Studies such as these underscore how a more complete
understanding of ontogeny becomes possible when the total body of the infant
and the total social environment are taken into account.
Finally, and continuing the lexicographical activism of Linguistics and
Evolution, the term philology needs to be destigmatized. The first pass at a res-
cue is to say that a philologist is an investigator of languaging behaviors whose
sociological environment is lost. Another is to say that philologists work on the
borderlands between languages, e.g., writing grammars or pursuing descrip-
tions in a structuralist framework. This is so even if they are working on only
one language, e.g., Russian, which sometimes leads to what can be called cog-
nitive philologism, where the description of the borderland is brought inside to
the mind/brain, aka modularism. Clearly, the term philologism is pejorative.
However, because writing grammars has long been a respectable and necessary
activity, then the rehabilitated term philology, both historical and structuralist,
could be used to characterize the line of disciplinary activity that comprises
much of the last two thousand years of grammar writing as well as most of
twentieth-century general linguistics writing to the extent that its theoretical
orientation and methodological approach is continuous with that of nineteenth-
century philology. As such, the term philology indicates the activity of analyz-
ing language in terms of properties and features that permits the reconstruction
of language lineages and facilitates the crossing of language borders.
An examination of our subject matter in terms of languaging relaxes some of
the rigidity that the term language has imposed on linguistic theory and prac-
tice. With this relaxing, new spaces open up and new phenomena come into
view. It does not take much – perhaps only a few centimeters, in some cases –
for an expansion of scope to have a large-scale consequence. For instance,
in The Origin of Speech (2008), psychologist Peter MacNeilage increases
his investigation of the origin of the human ability to organize the movement
sequences of speech to include the mandible. This is a part of the wider articu-
latory apparatus otherwise ignored in phonetics and phonology, which typic-
ally concentrate on tongue movements. However, by bringing the mandible
into view, MacNeilage is able to produce a convincing account of the instanti-
ation and development of syllable structure both phylogenically and ontogen-
ically through studying the mandibular oscillations that were originally (and
still are) used in ingestion. His work will be taken up again in later chapters.
For the moment, the key point is to recognize that a twenty-first-century lin-
guistics is one that is oriented away from a conceptual framework in which
the term language has meaning and toward one in which languaging is both
lively and engaging. Instead of asking: “what is language?” or “what are the
properties of our capacity for language?” or “what do we know when we know
a language?”, the questions become: “how do living beings become languag-
ing living beings?” and “how do we become the particular languaging living
beings that we do?”
2 Developmental systems theory
The shift from language to languaging requires a concomitant shift from a

framework where the terms nature and nurture have currency to one where
they have been eliminated. The first step in eliminating the nature/nurture
dichotomy is to confront the idea of preformationism. This is the theory that
every germ cell contains the organism of its kind fully formed and complete
in all its parts and that development consists merely in an increase in size
from microscopic proportions to those of the adult. Although this theory is now
widely discredited, the developmental psychologist Susan Oyama argues con-
vincingly that preformationist thinking is alive and well and has shape-shifted
into our ideas about information, the modern source of form. She opens her
groundbreaking book The Ontogeny of Information with the passage:
Those who have argued over the origin of ideas and of biological beings have usually
agreed that form in some sense preexists its appearance in minds and bodies, though
they have disputed the method and time of its imposition. Most solutions to the puzzle
of how form arises, therefore, including the most recent biological dogma, have incor-
porated this assumption. To the extent to which this is true, they are of limited value
in answering questions about origins and development. Whether it is God, a vitalistic
force or the gene as nature’s agent that is the source of the design of living things and
that initiates and directs the unfolding of the design thus matters little to the structure of
the argument. Nor are the problems inherent in such a notion lessened by the use of a
succession of metaphors, such as genetic plans, knowledge and programs, to serve these
cognitive and intentional functions. (1985: 1)
Her point is that information is seen to reside in molecules, cells, tissues and/
or the so-called environment. It is considered to be often latent but causally
potent and allows the molecules, cells, tissues and/or the environment to rec-
ognize, select and instruct each other, to construct each other and themselves,
to regulate, control, induce, direct and determine events of all kinds. The pre-
formationist attitude resides in the assumption that information exists before
its utilization or expression. Shedding preformationist thinking means elimin-
ating the idea that form – or its modern agent, information – exists before the
interactions in which it appears and must be transmitted to the organism either
through the genes or by the environment. In order to get beyond preformationist
40
Developmental systems theory 41
thinking Oyama recommends that we abandon the old opposition of nature

and nurture conceived as alternative sources of causal power, where nature is
push from the genes and nurture is pull from the environment. Oyama offers
developmental systems theory (DST) as a way to reformulate an understanding
of development without collapsing back into the old oppositions. One of the
achievements of DST is to have eliminated the nature/nurture dichotomy.
Eliminating the nature/nurture dichotomy and incorporating DST into a
twenty-first-century linguistics is easier said than done, given the long history
of nature/nurture thinking and its entrenchment in the discipline. The math-
ematician Francis Galton, a cousin of Charles Darwin, got the nature/nurture
ball rolling in its present form when he asserted that human personalities were
born, not made by experience. According to Steven Pinker, Galton borrowed
the alliteration from Shakespeare, who lifted it from the Elizabethan school-
master Richard Mulcaster, who wrote in 1582 that “Nature makes the boy
toward, nurture sees him forward.” Pinker titled his 2002 article unhopefully
and unhelpfully “Why the Nature/Nurture Debate Won’t Go Away.” He might
as well have titled it “Why Some People Won’t Let the Nature/Nurture Debate
Go Away,” since he seems committed to keeping it in place. His job is to give
the debate, as he sees it, proper nuance.
A representative discussion of the terms in the context of current linguistic
theory can be found in Jackendoff:
The term “innate” … requires comment. For a first approximation, it means “present
at birth.” However, it is customarily used more broadly to denote a characteristic that
appears automatically in the course of an organism’s development, whether before or
after birth. For instance the number and organization of human teeth, which develop
after birth, can be said to be innate. (2002: 72)
This is the very kind of description that Oyama’s work seeks to change. She
would say that the number and organization of human teeth are not innate;
they are developmental results. By the same token, whatever is “present at
birth” is also a developmental result of nine months of ontogeny. Jackendoff
continues:
The term is normally contrasted with “acquired” characteristics, which are due to the
influence of the environment. It is now widely understood that most characteristics
of organisms result from an interaction of innate and environmental influences. The
strength of one’s muscles depends on exercise and nutrition. But the fact that humans
develop muscles in the places they have them is innate. (2002: 72)
Again, for Oyama, acknowledging that some trait or character is the result
of both nature and nurture does nothing to alter the conceptual confusions
behind these terms. The confusion continues when, later, Jackendoff invites
us to “consider something as simple as sneezing, which I assume we can agree
is innate” (2002: 91). Assigning the label innate to the activity of sneezing is
more than useless; it is counterproductive in that it short-circuits the under-

standing of that activity that can be told in terms of the allergens in the
environment and the particular organism’s history of interactions with those
allergens. In other words, it does not distinguish between the output, namely
sneezing, which may be observable behavior of every human being on earth,
and the input, i.e., the trigger, which may be widely variable from person to
person and place to place.
To add to the confusion, the nature/nurture dichotomy hooks onto the vast
expanse of phylogenic explanations. Eliminating it requires a theoretical oper-
ation akin to a root canal, and this operation draws us into the long and inter-
twined history of the discursive interactions between philosophy and science.
In order to undo the science/philosophy tangle that will help us dissolve the
nature/nurture knot, this chapter begins with Maturana’s analysis of the differ-
ing operational coherences that obtain to create the differing disciplines of sci-
ence and philosophy, and it ends with the Australian philosopher Paul Griffiths’
analysis of the psychological category emotion which leads him to eliminate it
as a useful scientific category. In between, Oyama’s work on the nature/nurture
dichotomy not only takes center stage but also takes us back to the concerns of
the British embryologist Conrad Waddington, who recognized the importance
of development in understanding the phenotype. All four researchers – a neuro-
biologist, a philosopher, a developmental psychologist and an embryologist –
share a deep concern for how we language about the phenomena that interest
us. They also possess fine linguistic compasses that will help lead those of us
who care to follow them out of some dense conceptual undergrowth.
Maturana’s “Scientific and Philosophical Theories”

Maturana addresses the relationship between science and philosophy in an
unpublished article entitled “Scientific and Philosophical Theories.”1 Maturana
begins by describing what a theory is:
A theory is an explanatory system that interconnects many otherwise apparently unre-
lated phenomena (experiences), which is proposed as a domain of coherent explana-
tions that are woven together with some conceptual thread that defines the nature of
its internal connectivity and the extent of its generative applicability in the domain of
human actions. As such a theory is valid for those who accept both the criterion of val-
idation of the explanations that it entails, and the criterion of internal connectivity that
makes it a fully coherent conceptual system.
Thus, an explanatory system is recognized as a theory when it fulfills
two criteria: (i) the criterion of evaluation; and (ii) the criterion of internal
connectivity.
1
Provided by Cristina Magro
Maturana’s “Scientific and Philosophical Theories” 43
Internal connectivity is often spoken of in terms of predictive power, that

is, when a theoretical statement or a generalization about a phenomenon can
extend to explain cases that were not originally under review. For example,
the phenomenon of so-called island constraints in syntax has been discussed
by linguists for the last forty years. These constraints refer to certain syntactic
constructions that cannot be extracted from sentences, making them islands to
unbounded dependency relations, e.g., a question, relative clause, or topicaliza-
tion. An example would be the object of a subject clause, as in That she knew x
bothered him, in which x is an island because it cannot be made the subject of
a question: *Who did that she knew bother him? In generative analyses, these
constraints are understood in purely syntactic terms. However, in construction
grammar, island phenomena are seen to exist across sentences and across inter-
locutors, strongly raising the possibility that constraints on islands are related
to discourse, specifically to which topics are foregrounded or backgrounded.
Linguist Adele Goldberg, for one, notes that elements that are foregrounded in
the discourse are candidates for unbounded dependencies, while elements that
are backgrounded are not. She thus formulates the following generalization:
backgrounded constructions are islands (BCI) (2006: 135). Without rehearsing
her arguments here, she is able to conclude that:
It is clear that the BCI account predicts a wide variety of facts in a straightforward way:
the fact that subjects, definite relative clauses, full noun complements, the ditransi-
tive recipient argument, presupposed adjuncts, complements of manner of speaking
verbs, and factive verbs are all islands is predicted. The fact that unbounded dependen-
cies are sometimes allowed from presentational relative clauses, (reduced) indefinite
relative clauses, non-presupposed adjuncts, and verbs of saying is also expected. In
addition, the fact that direct questions and exclamative ah! is sensitive to islands is
also expected on an account that relies on the discourse properties of the construction
involved. (2006: 149)
In other words, an impressive array of syntactic phenomena is predicted, in the
sense that it is explainable by extension of a generalization of a discursive prin-
ciple. This particular group of syntactic phenomena has not previously been
understood to be related in just this way. Thus the BCI fulfills the criterion of
internal connectivity. The topic of construction grammar will be taken up in
more detail in Chapter 8.
The criterion of evaluation is met when the theory secures uptake, as J. L.
Austin might say, in a group of people who are qualified to pass judgment on
the explanatory system, e.g., when syntacticians and discourse analysts accept
the BCI. Similarly, chemists are the ones to pass muster on a chemical theory,
philosophers are the ones to judge a philosophical theory, and so on. This is a
shorthand way of saying that chemists are chemists when their work conforms
to the standards and procedures that make the discipline of chemistry what it
is. This may mean, among other things, that they have learned an established
body of data, recognize certain phenomena, follow certain methods in produ-

cing experiments and keep records in certain ways. The same could be said of
a philosophical theory, namely that a philosopher is a philosopher – engages
in doing philosophy – when that philosopher’s work is accepted by a group of
philosophers whose own work conforms to the standards that make the discip-
line of philosophy what it is. The criterion of validation in philosophy may be
of many different kinds. That is, there is no one procedure for doing philoso-
phy. The only requirement is that the procedure be internally logically consist-
ent. The philosopher John Searle has argued that it is “precisely the absence of
such things as laboratory methods to fall back on [that] forces the philosopher
to even greater degrees of clarity, rigor, and precision” (1998: 159). It could be
said that in science, the gravest mistake may be to skew the results of an experi-
ment in order to prove a hypothesis, while in philosophy, the gravest mistake
may be to generate a theory that is self-refuting.
The difference between (this version of) science and (this version of) phil-
osophy lies in their respective criteria of internal connectivity. A scientific
theory generates an explanation of a phenomenon. The internal conceptual
and operational thread that weaves together a scientific theory as a system of
explanation requires that the phenomenon under observation be preserved as
the phenomenon to be explained, even if, during the course of the observation,
the scientist is obliged to change and/or abandon every hypothesis, principle,
procedure, etc. along the way. In fact, this condition might be considered the
fun of science: a scientist’s passion to explain a particular phenomenon – e.g.,
island constructions – may and hopefully will take him or her into unexpected
places. Although the journey may take the scientist far and wide, the scientific
pursuit is, in a way, held together by the recognition of a phenomenon to be
explained and not the conceptual structure that the scientist started out with.
In Maturana’s view, the situation obtaining in philosophy is quite different. A
philosophical theory provides an understanding of some facet of human experi-
ence. The internal conceptual and operational thread that weaves together the
system of explanations of a philosophical theory concerns the preservation of
certain beliefs, principles, values or desired results that the philosopher holds
as intrinsically valid. Although the philosopher’s line of argumentation may
take many different paths and although the philosopher’s source materials may
draw from many different disciplines and kinds of knowledge, his or her pur-
suit is held together by striving to preserve a certain set of implicit or explicit
basic premises that he or she accepts a priori.
A philosopher who wishes to preserve a certain basic principle or truth,
e.g., the notion of an objective reality, must generate statements that would
not deny the principles, contradict the values or lead away from the desired
result of providing a theory of objective reality. Such is the project of Searle
in Mind, Language and Society where he states that the following positions
Maturana’s “Scientific and Philosophical Theories” 45
(among others) are true and that the attacks on them (say, from postmodernists,
among others) are mistaken: “there is a real world that exists independently
of us, independently of our experiences, our thoughts, our language” and “we
have direct perceptual access to that world through our senses, especially touch
and vision” (1998: 10). He cites among mind-independent phenomena such
things as “hydrogen atoms, tectonic plates, viruses, trees, and galaxies” (1998:
13–14). The inner connectivity of Searle’s argument derives from his move to
make the activities of philosophy and of science continuous one with the other.
For Searle, philosophy is a kind of pre-science whose knowledge is not-yet-
systematized and not-yet-secure in the way of the knowledge that we recognize
as so-called regular science. “As soon as we are confident that we really have
knowledge and understanding in some domain,” Searle says, “we stop calling
it ‘philosophy’ and start calling it ‘science’” (1998: 158).
This claim of the continuity between philosophy and science is crucial in
Searle’s argument. It allows him to inject into the heads of the physicists,
geologists, biologists, botanists and astronomers – i.e., those scientists who
investigate the supposedly “mind-independent phenomena” – the principle that
Searle understands to be a function of those sciences, namely the identification
of various parts of what is often called objective reality. However, it seems
quite otherwise to me. Scientists as scientists are not interested in conserving
or refuting a principle such as objective truth. They can and do proceed in
investigating whatever phenomena they are investigating without having any
particular interest in whether or not these phenomena are objectively real or
prove something called objective reality.
Certain scientists, such as neurobiologists, are interested in explaining the
phenomenon of human cognition, that is, how we humans perceive our world.
Their investigation, however, is neither predicated on nor devoted to either
affirming or disconfirming any particular theory of human cognition, e.g., that
we have direct access to the objective world through our senses, especially
vision and touch. When a neuroscientist wishes to reflect on the results of his
or her research with the aim of determining or establishing certain principles
or values pertaining to some facet of human experience, he or she is operating
in (Maturana’s version of) the realm of philosophy.
At different moments in our daily lives, all of us operate sometimes as sci-
entists and sometimes as philosophers. We operate as scientists when we wish
to explain a phenomenon – phenomenon : Charlotte said she would meet
me at the coffee shop at 3:00 P.M. She did not show up. Why? h y pothesis :
Charlotte is blowing me off. She doesn’t want to be my friend anymore – and
we use evidence either to confirm or to revise our hypothesis. e v idence : At
3:30 P.M. Charlotte calls from a pay phone to say that she has run out of gas
and her cell phone battery is dead so she could not get in contact earlier. We
may wish to validate her explanation by asking the friend who was with her at
the time whether what she said happened really happened. validation : John
said that Charlotte was very upset when she ran out of gas and got to a pay
phone as soon as possible to call and apologize.
We are philosophers when we reflect on the phenomenon and wish to pre-
serve a certain belief. belief: Charlotte is disorganized and self-centered.
Beliefs have consequences. consequence: I would do well not to bother to be
her friend anymore, since she doesn’t care enough about me to make an effort to
show up at our meetings. We can hold to that belief and its consequence, despite
Charlotte’s explanation of the missed meetings and John’s corroboration of it, if
we can generate the following explanation. explanation: She knew she was low
on gas, because I noticed it the day before when I was with her, and I mentioned
it to her. This is not the first time she has ignored what I have to say. In addition,
this is the third time she has stood me up, and she always has a different excuse.
Or we could reflect on the phenomenon and wish to preserve another belief.
belief: Charlotte is a good person, and I value her as a friend. consequence:
I will continue as her friend. We can hold to that belief and its consequence,
despite the fact that she did not plan this afternoon’s meeting well and for the
third time, if we can generate the following explanation. explanation: On the
many occasions when she does show up when she says she will, she always has
wonderful stories to tell, and I love her sense of humor.
The phenomenon of Charlotte’s being late requires a scientific explanation.
How we understand her lateness requires a philosophical one. Our philosoph-
ical understanding may involve, say, our overall understanding of human
beings, either that Charlotte’s behavior is just typical. People are unreliable;
or that No one’s perfect. I can accept certain faults in my friends, e.g., a lit-
tle bit of unpredictability, but not others, e.g., lying. Thus, if further evidence
turned up that Charlotte (and John) had made up the story of running out of
gas, the consequence would be that she (and John) would be off the friend list.
If that consequence is undesirable, then an adjustment of a part of the belief
system may ensue: I can accept a little lying but not stealing. Or perhaps a
heart-to-heart conversation is in order, along with a readjustment of the entire
relationship.
This ordinary human interaction illustrates the operational coherences of both
scientific and philosophical theories. Scientific theories can and do change, just
as philosophical theories can and do change. Scientific theories change when
new evidence concerning a phenomenon or related phenomena turns up that
requires an adjusted explanation of the original phenomenon. Philosophical
theories change when the beliefs that generate them have consequences that,
for whatever reason, the holder of the philosophical theory no longer finds
desirable. A person will hold to a belief in objective reality if the consequence
of abandoning it is undesirable, e.g., it would require the abandonment of a lot
of other beliefs this person holds dear. This person will abandon her belief in
A parade of linguistic phenomena 47
an objective reality if the consequence of doing so is desirable. For instance,

abandoning a belief in objective reality may improve her ability, however indir-
ectly, to explain the phenomenon that is of interest to her scientifically and that
she wishes to preserve: phenomenon : I notice that hearing/seeing/reading
(certain kinds of) sequences of sounds/signs/graphs has powerful effects on
people. How can I explain this?
To state the intermingling of linguistic science and philosophy another way,
we can say that, as languaging living beings, we are already in language when
we undertake our explanations. An explanation, according to Maturana, “is an
answer to a question that proposes a reformulation of the experience which the
question demands to be explained, as an answer (explanatory answer).” If we
accept such an explanation of an explanation, then we see that it is impossible
to reformulate our linguistic experience outside of our linguistic experience.
Hence the importance of metalanguage. And hence the importance of the his-
toriographic perspective, which allows me as a disciplinary linguist – that is,
as an observer of our observations – to observe a wide array of the disciplinary
observations of our observations.
A parade of linguistic phenomena to be explained

Let’s then review the phenomena that have captured the interest of linguists
over the last two hundred years and the explanatory principles and procedures
they have proposed to undertake the examination of those phenomena:
• Nineteenth-century philologists wanted to explain the phenomenon of the
remarkable (phonetic, lexical, grammatical) similarities that exist among
certain groups of present-day languages. They did so by positing that these
languages all descended from a common source. They reconstructed their
models of the source language according to the principle known as the regu-
larity of sound change.
• Early twentieth-century structuralists wanted to explain the phenomenon of
the regular structures (phonetic, morphological, grammatical) that the philolo-
gists were able to reconstruct for source languages. They did so by positing
that structure as such characterizes any state (be it contemporary or in the dis-
tant past) of a language and, accordingly, made a strict methodological dis-
tinction between the explanatory terms of synchronic linguistics and those
of diachronic linguistics. American structuralists were interested to write the
complete description of a language at any given time but did not require the
grammars to be a complete description of any particular speaker’s behavior.
• Late twentieth-century generativists, following Chomsky, wanted to explain
what they thought of as higher-level structural phenomena such as para-
phrase and sentential ambiguity. They began by positing a dual-level gram-
mar. An important procedure for examining the relationship between the
two levels was native-speaker intuition about grammatical and ungrammat-

ical sentences. Knowledge of the relationship between these two levels was
deemed to be too complicated for a child to learn so quickly and so well, and
thus this knowledge was theorized to be part of the biological (pre-given,
inborn, innate, genetic) endowment of human beings. More recent generativ-
ists (minimalists) now want to most efficiently map the way language links
two presupposed constructs, namely the conceptual-intentional system and
the sensory-motor system.
Notice that the phenomena that drew the attention of the structuralists and
generativists were produced by the activity of philologic-linguistic theorizing –
call them epiphenomena – and so any presuppositions that informed the study
of the original phenomenon had no reason to surface for reexamination. It was
not until another phenomenon was (re)advanced for examination – how to
explain how what we claim to be the case about our biological endowment with
regard to the languages we speak came to be the way we claim it to be – that
the philosophical system holding philology-linguistics together was brought
into great clarity. Now, the original phenomena that have come under review in
philology-linguistics were never framed in terms of evolutionary explanations,
and so it should not be too surprising that they have been difficult to account
for in evolutionary terms, as we will see in Chapter 4. However, for the most
part, some current linguists have been reluctant to abandon the hypotheses,
principles and procedures that were developed to reconstruct protolanguages
and to describe langue and/or competence. Instead, they have committed them-
selves to affirming certain truths, e.g., that recursion is the essence of language,
and so they have moved into the realm of philosophy.2 To invert Searle’s idea
that philosophy is pre-scientific knowledge not-yet-systematized, one could
say that (one version of) philosophy is a solidification of one moment in the
development of scientific knowledge.
Eliminating the nature/nurture dichotomy once and for all

Oyama’s work is devoted to the twin and intertwined efforts aimed at: (i) expos-
ing the incoherences of the nature/nurture dichotomy; and (ii) demonstrating
2
A recent article in Language notes that one needs to distinguish the conception of the term
recursion in the current Minimalist Program from the conception of the term in an earlier stage
of generative grammar. In the latter, the idea was that recursion referred to the ability for one
phrase to reoccur inside another phrase of the same type, while in the former, more recent ver-
sion, the idea is that recursion refers to the general ability to build phrases that contain phrases as
subparts (whether or not any particular language shows evidence of embedding or not) (Nevins
et al. 2009: 367, n.11). What is of interest to me here is the fact that the term is retained, while
the concept becomes so watered-down that it ends up predicting – at least, in the interpretation
of Nevins et al. – that no language will limit itself to two-word sentences. The terminological
slippage suggests a desire to retain a value rather than to explain a phenomenon.
Eliminating the nature/nurture dichotomy 49
that it is possible to have an evolutionary perspective without the concept of

genetic programming or any of its surrogates (messages, instructions, blue-
prints, algorithms, encoding, hardwiring, etc.). Her overall goal is to integrate
evolutionary and developmental studies, and it is inspired, to a certain degree,
by the work of Conrad Waddington, whose critique of the term genetic code
is footnoted in Chapter 1. As we will see more specifically in Chapter 5, both
Oyama’s and Waddington’s work anticipate what is currently being called the
second Darwinian synthesis, an expanded view of the (first) neo-Darwinian
synthesis, aka the modern synthesis, that recognizes inheritance to involve
much more than the genes.
Certainly the genes appear to link evolution and development in two ways.
They are the material link, the one around which the neo-Darwinian synthesis
solidified. They are also a conceptual link which was forged, as Oyama notes,
“by the adoption of Francis Cricks’ Central Dogma of the one-way flow of
information (from genes to proteins, never from proteins in toward the genes,
Crick, 1957).” The one-way flow of information produced what is sometimes
called the “nucleocytoplasmic gap,” the result of which was the evolutionary
and developmental disciplinary split, with evolutionists traditionally working
on the nucleus (genes), while developmentalists have been mostly involved in
studying the cytoplasm (environment).
The one-way flow of information has, furthermore, come to have the qual-
ity of an unquestionable truth. Over the years, biologists and interested others
seem to have forgotten that a term such as genetic program, with genes under-
stood to be the prime movers, is metaphorical and based on a hypothesis about
specific molecular interactions, one that is, as Oyama points out, “open to
empirical support or refutation” (1989: 7). Crick himself, already nearly forty
years ago, recognized that there might be a problem for his dogma when he
wrote an update to his 1957 paper. In “Central Dogma of Molecular Biology”
his final line reads: “… the discovery of just one type of present day cell which
could carry out any of the three unknown transfers [protein → protein; pro-
tein → RNA; protein → DNA] would shake the whole intellectual basis of
molecular biology, and it is for this reason that the central dogma is as import-
ant today as when it was first proposed” (1970: 563). Indeed, the hypothesis
is now under revision, motivated in large part by the discovery by Stanley
Prusiner in the 1980s of prions (proteinaceous infectious particles) that involve
protein-to-protein transfers (Jablonka and Lamb 2005: 124–25).
Oyama observes that there are two general strategies people have used in
arguing for a genetic program. The first strategy can be called conventional
dualism, which attributes some features of a phenotype, i.e., an organism, to
nature (traits said to be innate or genetic) and some features to nurture (traits
said to be acquired or environmental). The problem with the gene/environ-
ment partitioning is that developmental processes and products do not sort
themselves out in this way. To make matters worse, the various criteria for
making the distinction between nature and nurture are not consistent, with
nature often exemplified by:
• presence at birth;
• appearance without obvious learning;
• longitudinal stability;
• reliable timing;
• susceptibility to perturbation;
• adaptiveness;
• phylogenic relationships;
• distribution in a population.
The problem here is that the first five examples involve very distinct devel-
opmental issues, while the last three are not developmental questions at all.
Lumping them as instances of the same thing, namely genetic nature, “guaran-
tees that conceptual chaos … will be veiled by a common vocabulary” (1989: 9).
Conventional nature/nurture dualism for Oyama thus fails because it rests on an
incoherent mix of ideas; there is no consistent way of distinguishing features that
are programmed from features that are not.
Tilting the nature/nurture slash-mark so that it becomes a seesaw along
which we arrange traits as being more on the nature end of things or more
on the nurture end of things does nothing to solve the conceptual problems.
It simply continues to confound developmental and evolutionary issues. The
nature end of the spectrum is traditionally secured to the genes and/or to the
notion of a genetic program. It thus draws its authority from its association
with evolution whose standard definition is “changes in gene frequencies.”
This definition, in turn, assumes a notion of genetic control over (some parts
of) ontogeny, i.e., development. Nurture is brought into the picture to specify
those (other) parts of ontogeny not controlled by the genes. With genes in com-
petition with nurture for an explanation of development, one has succeeded
only in conflating two domains: that of evolutionary explanations and that of
developmental explanation.
In addition to conventional dualism, there is a second strategy for maintain-
ing the concept of a genetic program, and Oyama calls this genetic imperial-
ism. For the genetic imperialist, genes determine the range of possibilities; they
set limits on development. Here all developmental possibilities are already in
the genes and anticipated by the genes. The environment merely “selects” the
outcome. The environment is relegated to a backdrop while internal forces
within the organism – referred to as nature, genes, or the biological endow-
ment – are accorded causal primacy. Although it is sometimes argued that the
environment is left out of the causal loop in an attempt to isolate individual
organisms and study them as solitary agents, what seems to be motivating that
isolation is the desire to contain ontogenic variety within genetic boundaries.

The problem with this imperialistic version, Oyama says, “is that it is vacuous:
A genotype has just those developmental possibilities that it has. Used this
way, the program no longer has empirical content, but is more like a symbol of
ultimate faith” (1989: 11).
In place of the central dogma and the two strategies involving genetic pro-
grams, Oyama proposes what she calls constructivist interactionism, in which
the ontogeny of an organism constitutes a perpetual coming-into-being. For
Oyama, nature and nurture are no longer to be understood as inner and outer
forces competing to project their designs on organismal form. Rather, she sees
them as product and process. As such, nature does not preexist ontogeny as
an always-already-there-but-dormant essence waiting to be “awakened.” This
is Chomsky’s metaphor of the slumbering essence in reference to the “rich
conceptual framework already in place” and the “rich system of assumptions
about sound structure and the structure of more complex utterances” that the
child brings to the task of learning language and that are “awakened by experi-
ence” (1988: 34). For Oyama, nature is an emergent product of multiple inner
and outer causal factors. It is not transmitted; it is constructed. Nature is not
genotypic (a genetic program) but phenotypic (a constructed result). Nurture
is the process that produces or constructs the organism. Nurture contributes
to typical characters as well as atypical characters; it is as necessary to uni-
versal characters as to variable ones; it is “as basic to stable characters as to
labile ones. Nature and nurture are therefore not alternative sources of form
and causal power. Rather, nature is the product of the process of the develop-
mental interactions we call nurture” (1989: 5).
Eliminating the metaphor of the genetic code and the nature/nurture dichot-
omy has at least two consequences. First, some readers may find the thought of
doing without some notion of genetic code disorienting. It may raise the fear
that the developmental process is somehow suddenly unregulated and unre-
liable. From a developmental systems perspective, stability is found not in a
genetic code but rather in the stability of the developmental system itself; and
as we know, the developmental system is only relatively reliable, for things can
and do go wrong all the time. But, generally, the system is reliable, with some
processes remaining stable despite considerable variation in their constituents
and with some outcomes remaining stable despite variation in their process
(known as canalization see below). In other words, the developmental system
has the capacity to reduce the effect of perturbing influences on the pheno-
type. Second, bridging the nucleocytoplasmic gap requires rethinking what
counts as the so-called environment. When DNA is the center of the universe,
everything else is just the surround, but when we adopt the developmental sys-
tems perspective, the nucleus is as much the environment of the cytoplasm as
the other way around. Furthermore, throughout a developmental process, the
environment of an organism constantly changes. The point is that the environ-

ment is no longer “the (mere) outside.” We have disrupted the chain of associ-
ations where nature is inside, form, essence, basic, innate, biology, hardware,
and fixed, and where nurture is outside, content, accident, derived, acquired,
culture, software, and variable.
In shifting the terms, Oyama takes the emphasis off trying to determine,
say, a human nature, and she directs our attention to describing the process of
nurture, aka ontogeny, aka the developmental system, which comprises a het-
erogeneous and relatively reliable mix of dynamic, interacting influences. The
developmental system is not a reading off of a preexisting code. It is, rather,
a complex of interacting influences, some inside the organism’s skin, some
external to it, and including its ecological niche in all its spatial and temporal
aspects. Oyama argues that in real-world developmental systems subjectivity
and objectivity are distributed across the traditional boundaries of organism
and world:
To see the organism-niche complex as both source and product of its own develop-
ment is to acknowledge the role of activity in life process – not genes organizing inert
raw material into being, and not environments shaping or selecting passive bodies and
minds, but organisms assimilating, seeking and manipulating their world, even as they
accommodate and respond to them. (2000: 95)
The idea that behavior is a consequence of the organism’s ongoing interactions

with its environment reappears in Maturana and Varela’s notion of autopoiesis
and structural coupling (Chapter 3) and in Jablonka and Lamb’s addition of
three evolutionary dimensions (epigenetic, behavioral, symbolic) to the usual
one (genetic) (Chapter 5).
Oyama’s concept of the developmental system has considerable importance
for an understanding of evolution. First, just as the ontogeny of an organism
emerges from this kind of rich constructivist interaction, so does the phylogeny
of a species. Because an environmental niche itself experiences transforma-
tions that are reciprocal to the transformations experienced by the organism(s)
embedded in it, organisms not only pass down genetic material to their off-
spring but environments as well. As Oyama describes it, features of the ecolog-
ical niche are often passed down, either because they are in some way tied to
the organism’s or its conspecifics’ activities or because they are stable features
of the general environment. Another way of saying this is in terms of nature
and nurture: we inherit genes and social environments (these are the elements
of our nurture); nervous systems and table manners are constructed results (the
nature we come to be). Or, more aphoristically, ontogenic means are inherited,
phenotypes are constructed.
Second, for Oyama, evolution is change in the constitution and distribution
of development systems, thus making ontogeny the very basis of evolution.
Transgenerational stability and change, which characterize evolution, depend

on the degrees of reliability of developmental processes and on a large array of
means for repeated ontogenic constructions. What passes from one generation
to the next is an entire developmental system. Heredity is not an explanation of
this process but rather a statement of that which must be explained.
At this point, Lamarck’s name comes into play. One of Oyama’s sources of
inspiration, namely Waddington, notes that Lamarck is the “only major figure
in the history of biology whose name has become a term of abuse” (1975:
38) for (being interpreted as) proposing the idea that acquired characteris-
tics can be heritable. (Lamarck was also the first to use the word biology, in
1802, and perhaps the first to wholeheartedly embrace the notion of evolution.)
Oyama, following Waddington, is willing to acknowledge that while the par-
ticular transformations an organism undergoes throughout its ontogeny are not
heritable, to some degree the transformations its environment undergoes are,
along with the possibility that transformations in the developmental system
itself are, as well.
Waddington introduced at least three terms to the biological community,
genetic assimilation, canalization and chreod. The first two have received
uptake, while the third does not seem to have.3 Waddington, who worked on
Drosophila, noticed some curious experimental results. A crossvein in the wing
of a fly sometimes failed to develop if the fly was subjected to heat shock as a
pupa. By selecting and propagating flies that developed a crossveinless condi-
tion in response to heat shock, Waddington developed a population in which
most individuals were crossveinless when treated with heat. But after further
selection, a considerable portion of the population was crossveinless even
without heat shock, and the crossveinless condition was heritable. Waddington
called this phenomenon genetic assimilation, which an introductory evolution-
ary biology textbook defines as the circumstance through which a characteris-
tic that initially develops in response to the environment becomes genetically
determined (Futuyma 1998: 441). It is noteworthy that this theory of inherit-
ance has a Mendelian (as opposed to a Lamarckian) interpretation.
Waddington furthermore noticed that in certain cases, considerable variation
in gene product might yield little or no variation in the phenotype. He referred to
this phenomenon as canalization, where certain developmental processes seem
to have a so-called zone of canalization within which the phenotype varies little,
while the borders of the zone mark the thresholds at which change in the amount
of gene product yields a change in phenotype. For Waddington, the concept of
3
Waddington defines the term chreod (Greek χρή necessity and όδόζ path) as “the most general
description of the kind of biological process which has been referred to as ‘goal directed’ … it is
preferable to use words (such as chreod) which do not lay such stress on the final state but draw
attention to the whole time-trajectory.” It is a “canalized trajectory which acts as an attractor for
nearby trajectories” (1975: 223).
canalization is not linked to adaptation or to natural selection. Different kinds

and degrees of canalization may be adaptive or maladaptive, depending on the
process, the organism and the ecological niche occupied. At the same time,
adaptive traits may or may not show canalization. Thus, Waddington cautioned,
it is wrong for the evolutionary biologist to imagine that the production of adap-
tive responses will become progressively more canalized.
Waddington was interested in pathways of change along the dimension of
time, that is, the stabilization of those pathways. In order to resist the idea that
all stabilization of phenotypic form is secured by the so-called information
in the genes (which does not include the dimension of time), he pictured the
stability in terms of a set of branching valleys in a multidimensional space
that includes a time dimension. The development of an individual phenotype
proceeds along a valley bottom. Variations of genotype, or of epigenetic envir-
onment, may push the course of development away from the valley floor, up
the neighboring hillside. However, there will be a tendency for the process to
find its way back, not to the point from which it was displaced but to some later
position on the canalized pathway of change (or chreod) from which it was
diverted. If it is pushed over a watershed, it may find itself running down to the
bottom of another valley. Figures 2.1 and 2.2 show how Waddington illustrates
his points (yes, every thought style has its directed perception).
The canalization of the phenotype explains how, in a given species, e.g.,
Drosophila, all individuals will look remarkably alike when, in fact, their geno-
types are remarkably different. This uniformity, a result of the canalization of
development, conceals the heterogeneity of the genotypes and of the epigenetic
environments. Canalization refers to the process whereby development is buffered
against both genetic and environmental variation. If one peg (gene) is knocked
out, processes that adjust the tension on the guy-ropes from other pegs could leave
the landscape essentially unchanged. In other words, there is a lot of structural and
functional redundancy in the genome and the developmental process.
With Waddington and his vision of the epigenetic landscape, we have the
beginnings of the idea of the developmental system. The point is that any under-
standing of the emergence of phylogenic form must recognize the importance
of organism–environment interactions. As we will see in Chapter 5, Maturana
and Varela make the encounter of the living being with its environment an
important, if not the most important, locus of ontogenic activity. In a way, we
could say that Maturana and Varela, along with Oyama and others interested
in pursuing a second neo-Darwinian synthesis, have restored that perspective.
According to Waddington, Darwin used the term survival in the sense of “an
organism living for a long period” and the term fittest to mean “most able to
carry out the ordinary transactions of life.” Under the influence of the modern
synthesis, survival was replaced by the term reproduction and fittest by the most
effective in contributing gametes to the next generation. Thus consideration of
Figure 2.1 Part of an epigenetic landscape. The developing organism is the

ball rolling down a path indicated by the contours of the landscape. A suitable
landscape will take the ball to the same destination from a wide range of
starting points. The developmental landscape is “held down” underneath
by individual genes, and changes in the genes would affect the shape of the
landscape. The developmental system is simultaneously the product of the
genes and the determinant of their significance.
the ability to carry out the ordinary business of life disappeared from evolu-
tionary discussions and was replaced by the conception of reproductive effi-
ciency. For Waddington:
This in effect reduces Darwinism to a tautology, and leaves for quite separate discus-
sion – which is very rarely provided – why animals should have evolved all sorts of
highly adaptive structures to do unlikely things instead of simply being reduced to
bags of eggs and sperm like certain parasitic worms. These comments apply to the for-
mal mathematical theory of neo-Darwinism. The most valuable modern evolutionary
thought is not to be found in this form of theory but in more generally phrased discus-
sion. (1975: 227, emphasis mine)
All this is to say that for Waddington, and then for Oyama, “the evolution of
organisms must really be regarded as the evolution of developmental systems”
(1975: 7). What Oyama has contributed to this “more generally phrased dis-
cussion” is the truly invaluable service of carefully untangling the nature/nur-
ture conceptual mess – which terminology lingers in Waddington’s discourse,
although he does his best not to get torqued up within it – and she offers us a
fresh road along which a linguistic historiographer such as I can continue her
linguistic theorizing.
Pertinently, in the shift from nature/nurture theorizing to descriptions of phe-
nomena pursued through a constructivist interactionism, the need to identify
Figure 2.2 Interactions underlying the epigenetic landscape. The pegs in

the ground represent genes; the strings leading from them are the chemical
tendencies the genes produce. The modeling of the epigenetic landscape,
which slopes down from above one’s head towards the distance, is controlled
by the pull of the numerous guy-ropes which are anchored to the genes.
some trait or behavior as a result of (genetic) nature or (environmental) nurture

vacates the stage and makes room for identification of the relevant so-called
interactants that are responsible for contributing to the construction of the trait
or character that one is investigating. In terms of the development system as a
whole that Oyama is interested to describe, she notes that aspects of the system
that do not vary are equally as important as those that do vary, e.g., gravity,
which usually does not vary but in its absence, bone and muscle may atrophy
because the pituitary produces insufficient growth hormone. In short, gravity is
an interactant in normal organism development. Other interactants include the
genome, cell structure, intracellular chemicals, extracellular environment, par-
ental reproductive system, self-stimulation by the organism itself, the imme-
diate physical environment, conspecifics and members of other species with
which important interactions take place (1989: 27).
Applied Oyama: Paul Griffiths’ What Emotions Really Are

There are many ways of being a philosopher and many ways of identifying
what constitutes doing philosophy. Rather than using Maturana’s description
of the operational coherences of his sense of philosophy to define philosophy,
Applied Oyama: Griffiths’ What Emotions Really Are 57
we could use an institutional definition and say that a philosopher is anyone

who has an appointment in a department of philosophy. Or we could depend on
self-identification and recognize a philosopher as anyone calling him/herself a
philosopher. Or we could organize the activity of philosophizing around a set
of particular concerns, e.g., the search for first principles, the unification of
knowledge, the making of distinctions (presumably useful), or the clarification
of confusions that may arise in the way that terms in everyday language help or
hinder scientific understanding of the phenomena they purport to identify.
The philosopher Paul Griffiths has taken on the latter concern with respect to
the term emotion and performed a service not only to those in the philosophical
community devoted to the philosophy of emotion, but also to anyone interested
in the study of emotion, namely psychologists, neuroscientists and linguists
alike. Griffiths does not attempt to preserve a certain belief, principle or value.
If anything, he attempts to eliminate the category of emotion from scientific
study, since the category as it exists in everyday English does nothing to illu-
minate what is going on in people who are said to be experiencing emotion.
Griffiths makes his argument, in part, by applying the developmental systems
approach found in Oyama, and thus his work is valuable to the developmental
linguistics being sketched out here.
Griffiths begins by noting that the category of superlunary objects, that is,
objects outside the orbit of the moon, had an important role in Aristotelian sci-
ence. No one doubts that these objects really do exist. In fact, there are plenty
of objects outside the orbit of the moon. However, the category superlunary
does nothing to help us understand them. The distinction between sublunary
and superlunary is completely arbitrary and has subsequently been eliminated
from planetary investigation. So it is, Griffiths argues, with the category of
emotion which in everyday English unites three different classes of psycho-
logical states that do not belong together: (i) so-called affect programs which
involve short-term salient cases of responses such as fear, anger, etc.; (ii)
socially constructed emotions that can be seen either to facilitate social interac-
tions or to manipulate them; and (iii) the less well-understood higher cognitive
motivational complexes – e.g., embarrassment, loyalty, revenge, envy, guilt,
jealousy, love – that are perceived to incur a kind of cost for having them by
producing disruptions to what otherwise might be long-term, so-called rational
plans. Because these three classes of states are neither processed by the same
psychological/physical mechanisms nor serve the same functions, Griffiths
argues that the category of emotion should be discarded for the purposes of
scientific explanation and induction. He is not advocating that the category be
eliminated from vernacular English, although people may choose to do so in
the future. For example, the term superlunary has been dropped as a category
from modern languages, although instances of objects outside the orbit of the
moon still exist.
There is space only to suggest the richness of Griffiths’ account, but not to
do it true justice, in this brief overview of the three classes of psychological
states that are included in the vernacular English use of the term emotion.
Emotion as an affect program

The study of affect programs is highly associated with the work of psychologist
Paul Ekman, whose experiments involving facial expression link his studies to
Darwin’s approach to emotion. The central idea of affect program theory is that
certain so-called emotional responses can be classed together because they share
the properties of being complex, coordinated and automatic. Ekman has iden-
tified six species-typical human affect programs: surprise, anger, fear, disgust,
sadness/grief and happiness/joy. A person experiencing one of these responses
will exhibit a complex and coordinated pattern of facial, musculoskeletal and
vocal changes as well as changes in the endocrinal system which affect hor-
mone levels and in the autonomic nervous system. Because these responses
occur independently of deliberate evaluation, i.e., so-called reason, they can be
seen to operate automatically and hence their inclusion in a category that trad-
itional folk psychology distinguishes from reason, namely emotion.
An important aspect of Ekman’s, and then Griffiths’, description of affect
programs is that it nicely separates the input and output sides of the story.
The outputs of the affect programs are stereotyped and pancultural (= occur-
ring in all or most cultures), while no such claims are made about the inputs.
That is, affect programs are not to be seen as innate because what triggers
the stereotyped response in an individual will reflect that particular individ-
ual’s learning history. In other words, the triggers require a learning history;
they are acquired. Furthermore, while affect programs can be seen to occur in
individuals in all (or most) cultures, they are not necessarily what one would
call the same in all individuals. Griffiths, as well as all biologists, for that
matter, distinguishes between traits that are polymorphic (= exist in several
different forms in the same species) and monomorphic (= exist in the same
form in every individual). For instance, eye color is polymorphic in humans,
while leg number is monomorphic. Griffiths points out that none of Ekman’s
experiments were designed to show that affect programs are monomorphic in
humans (1997: 62–63). Because we belong to the same species, humans will
have similar types of responses to what we perceive to be dangerous or pleas-
ant. However, what we perceive to be dangerous or pleasant and how intensely
we experience the specified affects will depend on our individual histories,
including our genetic histories.
The automatic quality of the affect programs merits their identification,
according to Griffiths, as examples of the Fodorean sense of modules (see
Chapter 3), in that their systems operation is mandatory, opaque to our central
cognitive processes and informationally encapsulated. For instance, we do

not take into account conscious beliefs about the harmlessness of earthworms
when the system is deciding upon a response. “Informational encapsulation,”
Griffiths writes, “is one of the most important features of modular systems,
since it captures what is meant by their ‘separation’ from other processes, such
as those leading to longer-term, planned action” (1997: 93). I do not object to
this use of modularity, because it narrows the focus of what can be called anger,
fear, etc. and has immediate payoffs in clarifying what a scientist is studying
when s/he is studying a particular state. Such a narrowing would have given a
handle to the neuroscientist R. L. Dolan in his struggle with the unwieldy cat-
egory emotion in “Emotion, Cognition, Behavior.” In this article, Dolan begins
by claiming that emotions are less encapsulated than other psychological states
and ends by acknowledging that a distinction must be made between automatic
and stereotyped response mechanisms and other kinds of processes that have
also been traditionally called emotional (2002: 1194). His point is to show how
much cognition is thoroughly embodied – a point with which I am in sym-
pathy. At the same time, when Dolan seems to be using the term emotion to
cover all of an individual’s cognitive experience, he inadvertently underscores
Griffiths’ point about how unhelpful the use of the everyday term emotion is in
pursuing the argument, since vernacular emotion concepts simply do not cor-
respond to what a neuroscientist will find meaningful to study.
Just as humans can be seen to have similar kinds of responses to similarly
perceived situations processed through similar mechanisms, so similarities
can also be seen with organisms to whom we are related in phylogeny, e.g.,
primates, mammals and vertebrates. Given this, Griffiths finds no difficulty
ascribing the response fear to, say, a chimp or a dog or any other creature
whose psychology is (generally) homologous with ours, but he would not
extend it to, say, a squid or an octopus. How far a traditional emotion term can
be extended is not always clear, because its use relies on the coincidence of
a whole set of features, including physiological changes. Thus, for instance,
“people may agree,” Griffiths notes, “that an octopus discharges its defensive
cloud of black ink ‘when it is afraid,’ but the same speakers back off if asked
whether the octopus is ‘really afraid’ or if it ‘experiences emotion.’ The ten-
dency to disavow these applications … is even stronger in the case of insects
and … artificial intelligences” (1997: 233). Griffiths presents a convincing case
to organize and understand affect programs as we do other biological phenom-
ena, as natural kinds or species that have a historical lineage.
Socially constructed emotions

The study of socially constructed emotions has been pursued, Griffiths tells
us, by social constructionists whose most interesting insights are crystallized
in the view that there are emotional responses whose existence depends on the
existence of cultural models of normal emotional response. These emotions are
to be understood as transitory social roles that are interpreted as passions rather
than as actions. A social role is a characteristic pattern of behavior found in a
particular society that serves either the individual who plays the role, or the
society as a whole, or both. Emotions as transitory social roles are embodied
in social practices in such a way that they produce patterns of reinforcement.
They are distinguished from pretenses because of the lack of conscious access
to their causes. Nevertheless, the responses conform to local cultural norms,
even while the experiencer and the greater society accept the responses as nat-
ural and involuntary.
Q: Why is John acting so goofy? A: Because he is falling in love. The explan-
ation is perfectly acceptable in American culture, although acting goofy while
falling in love is a behavioral norm found in only some cultures. Nevertheless,
falling-in-love behavior is deemed to be natural, but it actually licenses certain
behaviors, namely goofy ones. Q: Why was Mary being so snippy at the party?
A: Because she was angry at having received the bad news earlier in the day
that she had been denied tenure. Now, it is not the case that Mary was not feel-
ing something in the wake of this bad news. It is rather that the identification of
an angry state induced by a socially justifiable cause serves to excuse her boor-
ish behavior. Oh, I see. Yes, now I understand. The understanding comes from
the fact that everyone is in agreement that emotions are psychological states
that one cannot control. Mary’s behavior is interpreted to conform to that norm
of behavior without needing to make conformity a goal.
The social role model of social constructionism explains not only how a
cultural understanding of emotion eases social interactions but also how those
interactions can be manipulated. Part of being a functioning member of a cul-
ture is knowing when to emote and how to emote. There are times, then, when
an emotion may be produced as strategic behavior, such that the individual who
is supposedly under the control of a particular emotion is able to disclaim any
behavior s/he exhibits. The individual thus knowing what emotional responses
are expected in certain situations produces them in order to achieve certain
rewards. Griffiths notes that the reward may be as great as being excused for
an act of violence or as minimal as having some sort of guideline for behavior
in an otherwise confusing situation (1997: 155). The intriguing example he
offers is the relatively recent so-called Rambo behavior syndrome exhibited in
Western males, which is not unlike the syndrome found among men in a num-
ber of southeast Asian societies known as amok. The man running amok is not
pretending to be in a frenzy, but he would not be in the frenzy if the behavior of
running amok were not present in his culture as a plausible response to certain
unbearable social pressures. Similarly, in Western culture:
men who believe that none of their options allows them self-respect exhibit a rather
stereotypical pattern of behavior, probably derived from contemporary action films.
They shoot at a large group of people, not necessarily people associated with their mis-
fortunes, before being shot or shooting themselves. They purport to be “out of control”
and are treated as such by society, yet their behavior is under the fairly precise control of
a recently developed model of how one might behave in such a situation. (1997: 141)
This is not the place to speculate on suicide bombers or students who shoot
scores of classmates, but an account of their behaviors will require some similar
explanation. Some social constructionists would argue that the idea that these
emotions are disclaimed actions means that these emotions are mythical. That
is not the point here. Rather, the idea that some emotional responses develop
because the surrounding culture reinforces them supports what Griffiths calls
heterogeneous construction, namely the idea that the psychological phenotype
is constructed through the interaction of traditional so-called biological fac-
tors, traditional so-called cultural factors, and factors that are hard to classify
in terms of that dichotomy.
Griffiths argues that, although social constructionists have overshot their
mark by claiming that the whole domain of emotional phenomena are social
constructions, they have nevertheless contributed greatly to the discussion of
these phenomena in two respects: first, they have drawn attention to the fact
that “the everyday ‘folk psychology’ of emotion is not purely descriptive. It
contains prescriptions of how people should behave, and collective pretenses
about how people do behave” (1997: 10–11); and second, they point to the
importance of cultural models of emotion in the heterogeneous construction of
the psychological phenotype, as just described.
The utility of Oyama’s developmental systems theory to the understanding of
emotion should be clear at this point. For one, as Griffiths points out, Oyama’s
developmental systems approach makes it possible to discern the common ori-
gins of two emotional responses without having to factor the responses into
identical (evolved) and different (cultural) elements. That is, it “can hypothe-
size common origins on the basis of all sorts of abstract resemblances … Two
emotional responses in different cultures might be homologous in the same
distant way that the hammer and anvil bones in the mammalian ear are homo-
logues of bones in the reptilian jaw.” For another, Oyama’s developmental sys-
tems approach separates questions about the historical source of a trait from
questions about its development. If we acknowledge the point that humans
belong to a lineage of primates that have always had culture since they formed
what observers observe as the order of primates as such, then there is every
reason to understand that culture is an important resource in the construction
of the psychological phenotype. The fact that “the development of a trait has a
complex dependence on culture,” Griffiths reminds us,
does not show that an evolutionary perspective cannot throw light on its origin. The
developmental systems theorists are free to concentrate on the question of how the emo-
tional phenotype is constructed. They do not have to concern themselves with whether
the emotion in question is biological, in which case they must assimilate its construc-
tion process to that of a highly entrenched morphological feature, or cultural, in which
case they must treat biological factors as an uninteresting background. (1997: 135)
Biology and culture are, thus, both necessary to understanding emotional phe-
notypes. To focus for a moment on cultural variation, it is clear that cultural
variation in emotion exists. The beauty of developmental systems theory is that
it sets us up to understand both that every living being who becomes a languag-
ing living being shares a common ancestry and that every languaging living
being becomes a different languaging living being due to the incorporation of
variations existing in the cultural resources.
To summarize: affect programs are brief, highly stereotyped reactions that
have only limited involvement with the cognitive processes which are involved
in regulating long-term action. They can be distinguished from the focus of
attention of the social constructionists, who have been inspired by the apparent
variation in so-called emotional phenomena across cultures. As an example of
the latter, Griffiths uses Japanese amae, i.e., “a deeply gratifying sense of child-
like dependency on a person or institution” – an emotion which is not easy to
identify with any emotion experienced in Anglo-Saxon societies – to exemplify
an emotion that might seem to be wholly “socially constructed” (1997: 137).
Higher cognitive motivational complexes

Now we want to consider Griffiths’ third class of responses that are labeled in
vernacular English as emotions, for example envy, guilt, jealousy, shame, love,
that: (a) seem much more integrated into planned, long-term actions than affect
programs; and (b) do not seem to be either figments of the experiencer’s imagin-
ations or instances of what might be seen as completely unique emotions found
only in English-speaking cultures. The point is that the understanding of higher
cognitive motivational complexes such as envy, guilt, jealousy, amae and so
forth seem to require a different descriptive framework than either the terms of
the affect program allow or what the social constructionists have offered. It is
precisely the developmental system’s framework that calls for descriptions in
terms of heterogeneous construction that provides the understanding.
Griffiths’ account avoids the nature/nurture pitfalls of the usual terms of
cross-cultural descriptions of emotion terms. He does not need to posit either
universal or uniform emotional responses (fixed nature) in the world’s cul-
tures, whereby a local culture produces additional or supplementary elements
(variable nurture) that accompany a universal emotion, or to deny any com-
mon elements (nature) among emotional responses in different cultures and to
declare that they are “all nurture.” It is perfectly reasonable to assume and, in
fact, seems to be the case, that an emotion, such as love, will be more prom-
inent and differently understood in one culture than in another and that other
emotions, such as amae – even if an emotion sharing cognitive and physio-
logical elements with it can be found in Anglo-Saxon society – would appear
very different because love/amae would not find the other social factors with
which they usually interact. The idea of heterogeneous construction alerts
researchers to look for variation across cultures in common emotions posited
by evolutionary theorists and suggests a search for cross-cultural resemblances
in the highly variable emotions described by social constructionists.
Griffiths’ case in point is the strong connection made between guilt and
responsibility in European culture, one that does not exist everywhere. Other
cultures may have emotions that resemble guilt but lack the strong connection
to responsibility. The feeling of guilt in such cases may arise simply because
the people who experience it have been involved in an event that is considered
bad. Griffiths wants to avoid the claim that these two emotions are not the
same emotion as guilt. From his point of view, the resemblance in the two
emotions is as interesting as the difference. Both emotions may be “a response
to situations in which a social norm has been violated,” and they both may
“share important developmental resources.” What is needed, in comparing the
two different developmental systems that produce these similar responses, is a
study of the age at which the emotional phenotypes of these cultures diverge
from one another and the inputs that produce this divergence. If, for instance,
“experiences of punishment play a key role in the ontogeny of guilt,” then
an examination of the occasions on which punishment is handed out might
determine how and why the emotional phenotypes are different (1997: 157).
A developmental systems approach does not, in a word, polarize; it does not
either collapse all the interesting differences into instances of the so-called
same phenomenon or efface all the interesting similarities in order to register
only so-called surface differences.
The vernacular concept of emotion in English, as Griffiths sees it, is “an
attempt to mark out a category of psychological states which produce behavior
not integrated into long-term, planned action.” As such, the category makes
a certain kind of sense. An emotion like loyalty, for instance, leads people to
honor relationships at great personal cost. Revenge can be even more disas-
trous for the person bent on it. Love is no picnic, either. In other words, these
emotions can be seen to interrupt long-term plans. They allow people to “adopt
strategies that require them to be committed to act against their own interests
should certain circumstances arise.” This goes a long way to explaining why
emotions have a traditionally negative image (1997: 243). Affect programs,
similarly, are sources of motivation not integrated into the higher concepts
of beliefs and desires and also seem irruptive, that is, they interrupt what we
recognize as cool means-ends reasoning. Disclaimed action emotions easily

fall in the category because they aim to take advantage of the special status
that emotions are accorded because when one is under their sway one is not
responsible for one’s actions (1997: 245). The classic contrast between emo-
tion and reason, first mentioned in Chapter 1, is one of the oldest dichotomies
in Western thought. Its venerability is, however, no good reason to keep it as a
category in scientific discourse.
Beginning the search for linguistic interactants

The goal of the present work is to integrate current work in developmental and
evolutionary biology, neuroscience, philosophy, sociology, cultural anthro-
pology and, indeed, linguistics, in order to answer the question: “how does a
living being become a languaging living being?” The answer will require, as
Waddington might say, a more generally phrased discussion than the one that
is currently being held in traditional, formalist linguistics, and such a discus-
sion will require recognition of the genetic, epigenetic, behavioral and cultural
domains that contribute to the living being’s languaging behaviors.
In anticipation of the presentation in Chapters 4 through 7, Oyama can be
immediately applied here, and linguists no longer have to argue whether humans
either have a push-from-the-genes ability to learn language or have language
pulled out of them from the environment (or deposited into them from the
environment). These statements preempt investigation of the developmental
process by which the living being becomes a languaging living being, as well
as the ability to account for why only some living beings become languaging
living beings. Now, this is not to deny that languaging living beings have a bio-
logical basis for our languaging but rather to ask: what does it mean for some
behavior to have a biological basis? What could it mean for some behavior not
to have a biological basis? The proper question for languaging, biologically
speaking, is not whether or not there is some kind of biological basis for it, but
rather what kind of biological basis and at what level of specificity?
There is no time like the present to begin the search for linguistic interactants.
One obvious place to look is in the genome, and the gene most often invoked in
discussions of human languaging is the FOXP2 gene found in a region of 50 to
100 genes on chromosome 7. It was identified in 2001 by geneticist Anthony
Monaco’s group at Oxford University in collaboration with cognitive neurosci-
entist Faraneh Vargha-Khadem at the Institute of Child Health in London along
with Cecilia Lai, Simon Fisher and Jane Hurst (Lai et al. 2001). They showed
that mutations in this gene cause a wide range of languaging disabilities based
on a study of a family known as KE, a number of whose members exhibit
severe disruptions to grammar and to speech production, i.e., they have trouble
controlling the fine movements in the lower half of their face.
Beginning the search for linguistic interactants 65
A team of researchers at the Max Planck Institute for Evolutionary

Anthropology in Leipzig, Germany, sequenced the FOXP2 genes of a chim-
panzee, gorilla, orangutan, rhesus macaque and a mouse and then compared
them with the human sequence. They discovered that there have been only
three changes in the gene’s protein sequence of 715 amino acids in the past
70 million years, i.e., since the last common ancestor of humans and mice.
Two of those changes have occurred in the last 6 million years, namely after
the human and chimp lineages split. The team estimated that the FOXP2 gene
became what we might call fixed in human populations – i.e., all humans could
be determined to harbor the last amino acid substitution – between 120,00 and
200,000 years ago.
Note that this gene is not the gene for speech production or grammar. The
FOXP2 gene codes for a type of regulatory protein, known as a transcription
factor, which is involved in modulating the expression of other genes, i.e., it
participates in whether other genes are turned on or off. The protein produced
by FOXP2 belongs to a subclass of transcription factors known as forkhead
proteins, and many members of the forkhead family are known to be key regu-
lators of embryogenesis (Lai et al. 2001: 521). Of note is the fact that FOXP2
seems to be important in regulating key pathways in the developing lung, heart
and gut, such that calling it a “language gene” makes no more sense than call-
ing it a “lung gene” (Fisher 2006: 288). Many forkheads are, furthermore,
critical for the normal patterning of the central nervous system, and there is
evidence to suggest that FOXP2 has effects in subcortical structures that par-
ticipate in the human reiterative ability in domains as different as syntax and
dancing (Lieberman 2005: 231). In other words, the FOXP2 regulatory gene
participates in a complex suite of developmental processes that affect many
human activities mediated by the central nervous system, including speech and
grammar.
In sum, the FOXP2 gene seems a likely candidate for a linguistic interactant.
Surely other genes will emerge as equally relevant. An intriguing hypothesis
has recently been advanced by philosopher-psychologist-linguists Dan Dediu
and D. Robert Ladd at the University of Edinburgh. They have begun investi-
gation of a relationship between the distribution of tone languages in the world,
e.g., Chinese and the languages of sub-Saharan Africa, and the presence or
absence of certain alleles (an alternative form of a gene that may occur at a
given locus) on two brain development genes, ASPM and Microcephalin, both
of which have what are called derived haplogroups (combinations of alleles
in close proximity on chromosomes, which are inherited together) that show
signs for positive selection and geographic distribution. They notice that those
areas of the world where the new alleles (D-variants) are relatively rare are also
areas where tone languages are common, and they are able to rule out either
historical or geographic factors to explain the negative correlation between
tone and the population frequency of ASPM-D and MCPH-D. They acknow-
ledge, of course, that “the effects of ASPM-D and MCPH-D on brain structure
and functioning remain largely hypothetical, but it is entirely plausible that
they influence the cognitive capacities involved in processing phonological
structures and thereby lead to linguistic biases [with respect to language acqui-
sition and/or processing]” (2007: 10945). It is the bias that is interesting, for
Dediu and Ladd do not doubt that all normal children learn the language of the
community in which they are raised. Rather, Dediu and Ladd are interested to
trace the possibility that cognitive biases in a population of language learners
could influence the direction of language change such that “extremely small
biases at the individual level can be amplified by th[e] process of cultural trans-
mission and become manifest at the population level” (2007: 10944). Whether
or not Dediu and Ladd’s hypothesis will survive further scrutiny, a develop-
mental systems approach incorporates without difficulty the possibility that a
genetic variation in a population could bias some aspects of language learning
that would be reflected in the community norms of the adult grammar and that
then would be stabilized both genetically and linguistically.
Another place to look for interactants is in a languaging community. Here
candidates at hand would be any and all of the grammatical constructions that
are explored in construction grammar and that have the requisite amount of
specificity to function as an interactant in a developmental system. That is,
they are not preexisting abstract forms waiting to be filled with content, but
rather item-specific constructions that can be learned on an item-by-item basis
and then extended (or not) to novel cases, depending on the disposition of the
child encountering them and then using them. To anticipate, once again, what
will be said about construction grammar in Chapter 8, mention can be made
here of English serial verb constructions as particularly good candidates for
linguistic interactants: the VVingPP (“Took-off-Screaming”) Construction, the
GoVPing, “don’t go sticking your nose” Construction, and The GoVPbare, “go
tell it to the mountain” Construction. They are particularly good candidates
because English does not allow serial verbs in general and because, while all
three are superficially similar, each must be described on its own terms because
they each have different syntactic, semantic and pragmatic constraints. As such,
all three must be learned separately (Goldberg 2006: 52–54).
Gesturing practices figure here as well, and differences in gesturing prac-
tices can and do correlate with differences in spoken language abilities. For
instance, it is well known that children from high socioeconomic status (SES)
families arrive at school with larger vocabularies, on average, than children
from low-SES families. A recent study has taken an interesting tack on iden-
tifying precursors for this inequality in vocabulary size and has found a posi-
tive correlation between the amount of caretaker–child gesturing interactions
when the child is 14 months old and the size of the child’s vocabulary skills at
Beginning the search for linguistic interactants 67
54 months of age (Rowe and Goldin-Meadow 2009). Since gesturing in chil-

dren typically begins around 10 months, this means that the socioeconomic
differences become evident only four months after the onset of child gesture
production. The question is why. Rowe and Goldin-Meadow answer: “Even
before they produce their own gestures, children comprehend the gestures that
others produce. Because children from high-SES backgrounds are exposed to
a broader range of meanings in gesture than children from lower-SES back-
grounds, they themselves are likely to produce more meanings in gesture,
which then promotes the development of more vocabulary words in speech”
(2009: 953). Certainly, the range of vocabulary used by caregivers is a major
factor in the vocabularies of children entering school. However, the larger
vocabulary use by high-SES parents cannot be divorced from the possibility
that the greater amount of high-SES parent–child gesturing interactions may
play a direct role in word learning by giving the child an opportunity to prac-
tice generating particular meanings by hand, at a time when those meanings are
difficult to produce by mouth. While this study suggests a possible remedy for
the SES inequalities, the point of it here is to highlight the idea that linguistic
interactants come in all shapes and sizes and can be found in many places.
A linguistic interactant is meant to be a messy clump of elements – neuro-
logical, psychological, physical, sociological – that does not know anything
about disciplinary boundary lines. Thus, it would be satisfying to find, for
instance, neurological and behavioral activities correlated with serial verb con-
structions in English, which it does not seem at present anyone has tried to
identify. Clumps of elements would count as linguistic interactants if they:
• served an orienting function in a languaging living being’s cognitive domain
or supported such an orienting function;
• were intertwined with action; and
• were bound to effects on the environment.
This discussion suggests only the barest genetic, linguistic and sociocultural
elements that can be identified as being crucial to the emergence and develop-
ment of languaging activities both in phylogeny and ontogeny. The future is
sure to produce a more complete assortment, along with more dynamic neuro-
logical descriptions of how languaging activities are knit into the brain and
correlate with behavior and cultural variables.
3 A twist in the cognitive turn
Adopting a developmental systems approach in linguistics does not mean that

everything gets flattened and that we are to dispense with descriptions of either
the interior or the exterior. Now, development systems theory does compli-
cate ideas of what and where the boundaries of the interior and exterior are.
For instance, a recent study showed that six-month-old babies of women who
drank a lot of carrot juice during the last three months of pregnancy preferred
cereal made with carrot juice to that made with water. Babies whose moth-
ers had drunk water showed no such preference (Jablonka and Lamb 2005:
162–66). This finding supports other studies that account for the robustness of
local/ethnic cuisines by the diet the mother eats when pregnant, which affects
the taste preferences of her offspring. Similarly, hearing develops around the
fourth month of gestation, and newborns at four days and even earlier can
be shown to be attentive to and discriminate between utterances in the native
language (the ones they have been hearing for the previous five months) as
opposed to a foreign language (Karmiloff and Karmiloff-Smith 2001: 43–46).
These studies and others suggest that there is no neat way to divide biology
(the traditional interior) from culture (the traditional exterior). They under-
score the idea that while the moment of birth is an important one, it is not the
point at which the effects of the outside world can be said to begin. Yet these
studies do not eliminate the need to understand and describe the interiors and
exteriors.
Although the present study wants to press the point that our brains and our
bodies have been significantly shaped by human cultural history and that what-
ever we call mind is not contained within a skull and some skin, it is still
nevertheless necessary to have a rich and rigorous way to talk about our inte-
riors. Maturana and Varela’s notion of autopoiesis provides such a way. Not
incidentally, for Maturana and Varela, a significant way we humans maintain
our autopoiesis is through our languaging together. The purpose of this chap-
ter is to make accessible the thought style in which the term autopoiesis has
meaning.
68
Opening on to the interior 69
Opening on to the interior

The usual ways into our interiors are through disciplines such as psychology,
cognitive psychology, cognitive science, neuropsychology, neurobiology and
cognitive neuroscience. Any introductory statement that could be made about
the relationship among these disciplines – e.g., that, in general terms, the psy-
cho- end of things examines the mental processes involved in human prob-
lem solving, perception, memory and language, while the neuro- end of things
delves into brain structures and architecture, i.e., the stuff that makes the psy-
cho- end of things possible – requires immediate revisions and complications.
Neurobiology does not limit itself to the study of the brain but takes the entire
nervous system within its purview, which includes, in addition to the brain, the
spinal cord, the nerves, the ganglia, and receptor organs such as the eyes. And
the cognitive part of the various equations puts at issue whether human cogni-
tion is central to the project, since cognitive scientists are interested to study
cognitive or cognitive-like activity wherever it is found, and this includes, of
course, artificial intelligence.
Attention to artificial intelligence by cognitive scientists is not surprising
given that cognitive science is a multidisciplinary approach to the study of
cognition and intelligence that emerged in the late 1950s when researchers in
psychology, linguistics and computer science worked together to develop an
alternative to behaviorist approaches to mind and language. It is also not sur-
prising given this history that a number of cognitive scientists tend to view the
human mind as a kind of computer and to speak of mental processes in terms
of computations and algorithms. This directed perception not only comes
equipped with an understanding of the mind/brain as the hardware and the
cultural inputs as the software, it also makes ‘uploading’ a universalist model
of language borrowed from generativist linguists extremely easy.1 However,
given the rise of constructivist accounts of cognition in recent decades, a grow-
ing number of cognitive scientists are moving away from the dualism implied
in the computer metaphor to embrace a more dynamic and feedback-looped
account of the relationship between the exterior of the world, with all its diver-
sity, and the interior of the nervous system. A constructivist approach, as has
been said, foregrounds the intersubjective nature of human cognition, summa-
rized in such statements as “thinking is a supremely social activity which can-
not by any means be completely localized within the individual” (Fleck 1979
[1935]: 98) or: “All human thinking gets discursified; we exchange ideas, we
lend and borrow verifications, get them from one another by means of social
1
See Evans and Levinson (2009) for a critique of Universal Grammar and its negative effects on
cognitive science. The topic of language diversity will be taken up more fully in Chapter 7.
intercourse” (James 1981 [1907]: 97). In other words, constructivist accounts

emphasize the social knowing that occurs in the dynamics of languaging as an
orienting behavior.
Where is language/ing located in the individual?

Language philosopher Talbot Taylor nicely sets up our look inside a languag-
ing living being by asking a useful question: “I have legs so that I can walk,
but is walking in my legs?” This surely rhetorical question directs attention to
the idea that although certain anatomical parts such as legs are necessary to
engage in particular activities such as walking, an understanding of that activ-
ity cannot be reduced to an examination of the anatomical parts necessary for
its engagement. A complete understanding of walking will require, in addition
to an examination of the hip socket, the foot bones, the knee joint and the thigh
and calf muscles, a discussion of a complex sequence of movements that takes
account of the effects of an erect spine, contralateral arm movements, balance
and gravity.
The comparison with walking is not made at random. Walking, like lan-
guaging, involves motor control, and evolutionary biologist Philip Lieberman
argues that motor control was the pre-adaptive basis of human syntactic ability.
The kind of creativity and flexibility as well as the constraints that are found
in the production and reception of syntactic structures are the kinds, he notes,
that are found in the activity of walking, which “involves positioning your
foot to achieve heel strike at the precise moment of contact with a floorboard
or the irregular surface of a country path. Your heel must flex at the proper
moment; and that moment is different depending on whether you are walking
or running, your posture, and whether the path is stony or smooth. There is a
‘syntax’ of walking” (2006: 13). The implication is that this pedestrian syntax
evolutionarily predates linguistic syntax. Lieberman also points out that: (i)
the left-hemisphere cortical area known as Broca’s (traditionally Brodmann’s
area 44–45, damage to which can result in expressive aphasia) mediates more
than speech production; it is involved in manual motor control and the recog-
nition of gestures; and (ii) “bilateral cortical activity during language tasks is
the norm in neurologically intact subjects,” meaning that not only Broca’s area
and the left-hemisphere cortical area known as Wernicke’s (traditionally BA
41–42, damage to which can result in receptive aphasia) are in play but also
their right-hemisphere homologues, as well as subcortical structures such as
the basal ganglia (2005: 228).
Although MacNeilage takes issue with Lieberman’s positing that the neural
basis of speech is found almost exclusively in the basal ganglia (2008: 326–27),
he supports the notion that, in order to understand the evolution of vocal-action
specialization, the postural configurations of the body must be taken into
Where is language/ing located in the individual? 71
account. As opposed to many researchers who associate the left-hemispheric

specialization of Broca’s and Wernicke’s areas with handedness, he associates
it with footedness. In his analysis, the postural demands of rapid, one-handed
reaching for, say, food while clinging vertically in an arboreal habitat, as our
ancestral primates apparently did, had several consequences: (i) a specializa-
tion of the right side of the body, mediated by the left hemisphere, for pos-
tural support; and (ii) a left-handed preference for reaching that enhanced the
right-hemisphere specialization for spatial cognition. To state the hemispheric
specializations another way, the left hemisphere mediates “whole-body action
control under routine circumstances” (2008: 211), while the right hemisphere
mediates the “apprehension of the world in situations with survival-related risk”
(2008: 215), i.e., emergencies. In MacNeilage’s argument, this asymmetry also
correlates with right-hemisphere dominance for negative emotionality and left-
hemisphere dominance for approach behavior. For language, in particular, it is
the idea of whole-body action control under routine circumstances that links it
to the left hemisphere, “as one would expect” (2008: 211).
Now, although Taylor’s question is not directly designed to address the pos-
tural configurations of our primate ancestors or contemporary human walking,
Taylor – and presumably Lieberman and MacNeilage – means for us to realize
that while an intact Broca’s area and an intact Wernicke’s area are typically
necessary to what we recognize as the normal functioning of languaging in an
individual, languaging as such is not in those areas – or in any other area of
the brain. Taylor would be likely to say that languaging is a full-bodied and, in
fact, multi-bodied activity whose understanding will require an investigation of
many areas of the brain, cortical as well as subcortical, and of other body parts,
including the ears, mouth and head and, indeed, the entire e-motioning body;
consideration of how languaging living beings interact with one another (hence
inclusion of the adjective multi-bodied in the description); and an understand-
ing of languaging in its ongoing intra-individual historical dynamic.
The upshot of a good century and a half of neurolinguistic research confirms
that, while various linguistic abilities can be seen to be localized and correlated
to specific kinds of cellular architecture, it is also the case that: (i) strict local-
ization does not hold; patients with similar lesions may have very different
disabilities, while similar disabilities may result from lesions in widely dif-
ferent areas; expressive and receptive aphasias may occur even when Broca’s
and Wernicke’s areas are unaffected; there are even cases of patients who have
had Broca’s area surgically removed in both hemispheres and who were still
able to speak; (ii) these operational localizations are to be viewed not as isola-
tions of faculties or functions but rather as areas where certain modalities of
interactions converge. Lesions in and around BA 44–45 may produce the loss
of the ability to produce fluent speech and gesture, but we do not want to say
that the functional deficiency occurred because Broca’s area is where speech
production is. Rather, we want to say that the functional deficiency occurred
because the lesions(s) impeded the convergence of activities necessary for the
synthesis of a particular conduct; (iii) recently, a new proposal has been put
forward to rename Broca’s area in order to better identify a unified conver-
gence zone for the synthesis of speech, called the LIFG (or lateral inferior
frontal gyrus). Interestingly, it includes BA 6, which is the lower part of the
premotor cortex. The idea is that the LIFG is a (somewhat expanded) area
where a series of related but anatomically distinct areas can be shown to sub-
serve more than one function in the language domain. Let us note that it is not
restricted solely to language functioning. For instance, the traditional Broca’s
area is activated when subjects are asked to search for a target hidden within a
complex geometric pattern, among other non-linguistic tasks (Hagoort 2005);
and (iv) when contemporary researchers start looking beyond the traditional
language centers for neurological activity supporting languaging, they are find-
ing it. This is also not to deny the importance of the traditional Broca’s area (or
whatever neuroscientists will tell us to call that area in the future) and of those
areas immediately surrounding it.
A recent study by Fedorenko and colleagues claims to have evidence that
there are functionally localized parts of the brain dedicated to language and
nothing else. They believe that many of the brain-imaging studies showing
overlap between language and non-language tasks may be due to flaws in
methodology, that is, in how functional magnetic resonance imaging (fMRI)
data is gathered and analyzed, which does not traditionally take account of the
fine-grained anatomical differences between brains. Their study focused on
giving subjects language tasks and comparing brain-imaging results with those
same subjects performing tasks involving arithmetic, working memory, cog-
nitive control and music, namely, functions “most commonly argued to share
neural machinery with language” (Fedorenko et al. 2011). They discovered
that there were brain regions supporting the language tasks that showed no sig-
nificant activation for any of the other tasks. While this finding is surely inter-
esting, it leaves open the following questions: (i) what conception of language/
ing are the researchers working with; does it include attention to gesture, for
instance? (ii) what activities not commonly argued to share neural machinery
with language might be involved in supporting language tasks? and (iii) could
this functional localization in the adult brain be a developmental result? That
is, in the developing languaging living being could certain kinds of language/
ing processing involve support from more brain regions, such as those recog-
nized to be involved in processing non-linguistic tasks in adults?
Whatever further brain-imaging studies will reveal, it is nevertheless the case
that, while languaging is not co-extensive with the whole of human cognition,
one cannot turn too many corners in the brain without encountering a struc-
ture and/or processing pathway that does not affect and has not been affected
The brain in twentieth-century linguistics 73
by the history of languaging in the species. For instance, an intact putamen

(a subcortical motor area associated with the basal ganglia) is necessary for
bi- and multilinguals to be able to stay on track in one language or another.
A related subcortical structure, the caudate, receives input from the prefrontal
premotor, temporal and parietal cortex and connects reciprocally to the cortex
via the thalamus. It has been shown to be sensitive to language change across
word pairs, suggesting that it, too, is critical for language control. Left caud-
ate responses, in particular, are highest “when there is a change in meaning
(semantically unrelated words in a pair) or a change in language (different
languages in a pair)” (Friederici 2006: 992).
In addition, neurobiologists have come to realize that all creatures belong-
ing to a certain lineage, e.g., mammals, do not have the same basic brain at
the microscopic level. It has recently been reported that only apes (and this
category includes humans) have a type of neuron, the Von Economo neuron
or VEN, that other primates do not. It is also the case that humans have many
more and larger VENs than other apes (Balter 2007). This finding is consistent
with the idea to be pursued in this chapter that cognition is not about “taking an
outside world in.” Rather, we humans live our cognitive domain, just as other
apes live their cognitive domains, a fact that implies differences at the neuro-
logical level. The cognitive domain of humans is furthermore so thoroughly
languaging-constituted that to be human and to be languagers is inseparable.
A second historiographic review: the brain in twentieth-century linguistics

Linguists have long recognized that a thorough understanding of our subject
matter entails recognition of a social component, an individual component
involving some part(s) of the body, as well as a temporal dimension. At the
very beginning of modern general linguistics Saussure acknowledges that:
The study of speech is then twofold: its basic part – having as its object langue, which is
purely social and independent of the individual – is exclusively psychological; its sec-
ondary part – which has as its object the individual side of speech, i.e. parole, including
phonation – is psychophysical.
Doubtless the two objects are closely connected, each depending on the other: langue
is necessary if parole is to be intelligible and produce all its effects; but parole is neces-
sary for the establishment of langue, and historically its actuality always comes first.
How would a speaker take it upon himself to associate an idea with a word-image if he
had not first come across the association in an act of parole? Moreover, we learn our
mother tongue by listening to others; only after countless experiences is it deposited in
our brain. Finally, parole is what causes langue to evolve. (1959 [1916]: 18)
Saussure understands both the historical and ontogenic primacy of parole
but focuses his theoretical attention on langue, which, he states, “is neces-
sary if parole is to be intelligible and produce all its effects.” We return to
the langue/parole problem in this chapter in order to register what the effect
of dichotomizing the study of our subject matter has been on linguists’ abil-
ity to describe a speaker’s/languager’s interior states and processes. In the
above passage, we learn that the “exclusively psychological” domain of
langue is both disembodied, given that it is “purely social and independent
of the individual,” and inert, given that “only after countless experiences is
it deposited in our brain.” This occurs “by listening to others” and not, say,
by interacting with them in consequential ways. In other words, we see a
degree zero of psychological theorizing: langue has little to do with general
psychology beyond the comment that an individual takes it upon himself “to
associate an idea with a word-image” only after coming across the associ-
ation in an act of parole; and langue has nothing at all to do with develop-
mental psychology since “it” gets “deposited” somehow, sometime “after
countless experiences.” However, langue does need some sort of location,
and Saussure locates it, predictably enough, in the brain, specifically “in the
third left frontal convolution,” which discovery Saussure duly attributes to
Broca (1959 [1916]: 10).
Bloomfield, who was not interested in identifying a theoretical construct
such as langue, takes a bolder psychological step by allying his brand of struc-
turalism with an actual school of psychology, namely American behaviorism.
This psychological perspective goes hand in hand with a bolder neurological
step. In the chapter “The Use of Language” from Language, Bloomfield distin-
guishes himself from earlier linguists who relied on so-called mentalistic, i.e.,
non-physical, factors such as spirit or will or mind to explain human conduct.
Bloomfield espouses instead the so-called materialistic theory that brings the
entire nervous system into play. He writes:
The part of the human body responsible for this delicate and variable adjustment [of
human conduct], is the nervous system. The nervous system is a very complex con-
ducting mechanism, which makes it possible for a change in one part of the body, (a
stimulus, say, in the eye) to result in a change in some other part (a response, say, of
reaching with the arm, or of moving the vocal chord and tongue). Further, it is clear that
the nervous system is changed, for a time or even permanently, by this very process of
conduction: our responses depend very largely upon our earlier dealings with the same
or similar stimuli. (1933: 33)
Bloomfield, too, mentions Broca (1933: 36), but he is reluctant to locate lan-
guage in specific brain centers because for him language is not a thing, e.g.,
grammar, but a complex of activities:
The error of seeking correlations between anatomically defined parts of the nervous
system and socially defined activities appears clearly when we see some physiologists
looking for a “visual word-center” which is to control reading and writing: one might as
well look for a specific brain-center for telegraphy or automobile-driving or the use of
any modern invention. Physiologically, language is not a unit of function, but consists
of a great many activities, whose union into a single far-reaching complex of habits
results from repeated stimulations during the individual’s early life. (1933: 36–37)
Lurking in this passage is the germ of what will later become the main criti-
cism against behaviorist accounts of language, namely that behaviorists such as
Bloomfield, who viewed the brain as a general-purpose learning organ, could
not give an adequate account of the way individuals learn language as opposed
to other activities, e.g., telegraphy or automobile driving. A view of the brain as
a specialized organ with built-in language mechanisms would soon thereafter
be proposed by the cognitivists.
Although Bloomfield’s psychology is not much more profound than
Saussure’s, Bloomfield has a wider vision of the so-called location of the
internal aspects of language/ing than does Saussure because he includes the
entire nervous system. And his statement that “Physiologically, language is
not a unit of function, but consists of a great many activities” united “into a
single far-reaching complex of habits” resonates with those of us who have a
predilection for the concept of languaging. However, Bloomfield falls in line
with the general disciplinary interests of his day in that he wants to provide a
structuralist description of a (more or less) adult grammar from the point of
view of the outside observer. Although attention to the temporal dimension is
a hallmark of behaviorist theory and practice, since behavior can be described
only in terms of histories of interactions, Bloomfield is not himself interested
in studying or describing how children become languagers. The so-called
acquisition process is only gestured at in the above passages in the phrases
“our responses depend very largely upon our earlier dealings with the same or
similar stimuli” and language is a “complex of habits that results from repeated
stimulations during the individual’s early life.” Nor is Bloomfield interested in
positing or describing the mental processes that might mediate between a lin-
guistic stimulus and a response.
With Chomsky, the cognitive turn in the discipline is effected, and the
interior takes center stage. However, that stage narrows significantly, since
Chomsky, in reaction against behaviorist accounts and anything to do with
actual stimuli, eliminates reference to the nervous system, along with inter-
est in the effects of time, phylogenic or ontogenic, on any given individual
or population. And – oddly enough or symptomatically enough – the (neuro-
logical) locations traditionally associated with language are displaced. To be
sure, Chomsky is more than willing to talk about the brain, usually in the col-
location mind/brain, but he tends to mention only one specific location in the
cortex, although not one traditionally associated with languaging behaviors.
That is to say that Chomsky often invokes the visual system (eye-brain/visual
cortex) – not for any role it might play in (his conception of) language, but
rather to make an analogy between its principles and those of his hypothesized
language faculty:
Th[e] principles [of the visual system] might be regarded as in some sense compar-
able to the principles of the language faculty. They are, of course, entirely different
principles. The language faculty does not include the rigidity principle or the principles
that govern apparent motion, and the visual faculty does not include the principles of
binding theory, case theory, structure dependence, and so on. The two systems operate
in quite different ways, not surprisingly.
What is known about other cognitive domains suggests that the same is true else-
where, though so little is known that one cannot be sure. It seems that the mind is modu-
lar, to use a technical term, consisting of separate systems with their own properties. Of
course, the systems interact; we can describe what we see, hear, smell, taste, imagine,
etc. – sometimes. There are thus central systems of some kind, but about these little is
understood. (1988: 161)
An interesting feature of this passage is Chomsky’s opinion of neurological

work, evident in his remark that “so little is known” about cognitive domains
other than the visual faculty or the language faculty. This deferral of neuro-
science has been characteristic of his work for over forty years since his 1959
review of B. F. Skinner’s Verbal Behavior and continues through to Nature and
Language (2002).
The key term in this passage is modular, which refers to a unit of process-
ing that is relatively autonomous from other processing modules. In common
usage, something that is modular is composed of standardized units that can
be used variably and flexibly. Modularity, as a psychological theory, is most
often currently associated with evolutionary psychology, in particular the work
of John Tooby and Leda Cosmides. The theoretical foundations of modularity
theory date from the early 1980s, in particular the work of the philosopher/lin-
guist Jerry Fodor, notably his The Modularity of Mind (1983). For those who
subscribe to a modular view of the mind, a module – a mental module – quali-
fies for the status of a module when it: (i) is domain specific; that is, it only
processes information about certain stimuli. Its information-processing proce-
dures are activated only by information about a particular aspect of the world
and remain unresponsive to information about other aspects of the world; (ii)
develops independently of any explicit instructions in the problem domain
with which it is specialized to deal. It is not that it develops in the absence of
any input; rather the inputs that trigger a development always contain far less
information than is present in the fully developed mechanism (another aspect
of so-called poverty of the stimulus – see Chapter 8). This means that the mod-
ule comes equipped with innate knowledge about the problem domain with
which it is specialized to deal and an innate set of procedures for applying that
knowledge to solve problems in its specialized domain; (iii) is informationally
encapsulated; a system is informationally encapsulated if there is information
unavailable to that system but which is available to the mind for other pur-
poses. A module thus does not have access to the full range of information
available in an organism’s brain. It tends to have access only to the outputs of
other psychological mechanisms and not to the information employed by those
mechanisms in generating their outputs; (iv) possesses proprietary algorithms;
proprietary algorithms treat the same information differently from other cog-
nitive subsystems; and (v) shows evidence that it was an independent target of
natural selection.
A flavor of some of these conditions is apparent in the above passage when
Chomsky notes that the mind consists of separate systems with their own prop-
erties. For Chomsky, not only is the language faculty a modular entity in the
mind/brain, it is itself composed of modules, e.g., the syntax module would
be one that accounts for how the syntactic features of sentences are processed
and produced. The operations merge and move in the Minimalist Program, for
instance, would presumably be considered by some linguists to be proper-
ties of the syntax module, and these operations would be considered to: (i) be
responsive only to syntactic information; (ii) be innate; (iii) provide outputs to
logical form and phonetic form; and (iv) be exclusively syntactic in character,
i.e., they are not reflections or effects of general properties of human cognition.
For Chomsky, the properties of the language faculty are, by definition, exclu-
sive to itself.
The rationale sometimes invoked to argue for the existence of mental mod-
ules is that they can be selectively impaired. Disruptions to so-called modular
aspects of the language faculty, for instance, can and often do occur, e.g., syn-
tactic processing and production can be impaired when an individual suffers
lesions or a stroke to localized areas of the brain, while phonology is spared.
However, one problem with this scenario for the postulation of a language fac-
ulty module is that while a person with a syntactic deficit may very well not
experience a phonological deficit, it is not the case that that individual only
experiences syntactic difficulties. S/he very well might experience muscular
palsy and/or other cognitive deficits. The question then becomes what other
kinds of neurological and psychological processing is syntactic processing and
production as such encapsulated from in the mind/brain? Syntactic process-
ing and production may well be removed, functionally and anatomically in the
adult, from that which permits the proper sequencing of syllables, but it may
also be completely bound to that which is needed to put one foot in front of the
other. Another problem with this scenario of encapsulation is that functional and
anatomical distances in languaging processing in the adult tell us nothing about
functional and anatomical interdependences in the developing languager.
Furthermore, from a developmental point of view, incapacities or abnor-
malities that appear early in childhood can be the occasion for compensatory
strategies and substantial neural reorganization. Children who have LLIs and
Williams’ syndrome are sometimes said to have, in the former case, dis-
rupted language but intact cognition and, in the latter case, intact language
and face-recognition abilities alongside other serious impairments. However,
these cases, far from proving the existence of a language module or a face-
recognition module, show instead that children with Williams’ syndrome,
for instance, process language and faces very differently than normal sub-
jects (Karmiloff-Smith 1998), while children with LLIs typically have other
cognitive and perceptual deficits (Woodward and Cowie 2004: 326). The
point here is that abnormal groups have overall atypical brain anatomy, brain
chemistry and temporal brain processes, all of which undercut a dissociation
between language and general cognitive abilities. In addition, although blind
children tend to eventually have the standard phonology of their speaking
community, they arrive at that phonology through a path that is different from
the sighted child’s. This casts doubt on Chomsky’s statement that: “Blind
children suffer serious deprivation of experience, but their language faculty
develops in a normal way” (1988: 39). Similarly, an early hearing deficit or
development delay that may subsequently not be detectable in adulthood may
nevertheless be at the origin of a life-long grammatical disorder (Karmiloff-
Smith 1998: 392).
Evolutionary psychologists tend to engage with Fodor’s point (v), above,
namely that modules can be shown to be independent targets of natural selec-
tion. They have proposed that the human mind is a vast collection of mental
modules from cheater detection, to mate selection, to predator avoidance, and
so on. It is worth noting that Fodor argues against the massive modularity theo-
rists in evolutionary psychology and has come out to defend the position that
only one module exists, namely the language module (2000). So far, the only
modules that have been successfully argued are Ekman’s affect programs, as
seen in Griffith’s work on emotion (Chapter 2). Even here, however, the mod-
ules do not come equipped with foreknowledge of what stimuli will trigger
them. What will cause a fear response or an anger response in an individual or
a group of individuals will depend on the history of interactions those individ-
uals have with particular features of their environment.
In opposition to Fodor, then, a DST-inspired linguistics envisions languag-
ing not as a mosaic trait, but rather as a connected one. A mosaic trait is one
that might qualify as a module in some sense, in that it can evolve independ-
ently of any other trait and with little effect on the rest of the organism, e.g.,
skin color. For instance, there is evidence that Europeans lightened up quite
recently, perhaps only 6,000 to 12,000 years ago, based on an analysis of gene
SLC24A5 (Gibbons 2007). However, this change seems to be related to no
further physiological effects. Having two lungs, by way of contrast, “is a con-
nected trait: you can’t change this trait without changing a great deal else in the
organism, because lung number is influenced by the genes and developmental
mechanisms that govern the bilateral symmetry of the organism” (Woodward

and Cowie 2004: 330). Woodward and Cowie go on to state that many human
cognitive abilities “may be connected rather than mosaic traits” (2004: 330),
a position consistent with a DST-inspired linguistics. In fact, a developmental
systems linguistics would say that our languaging is one of the fundamental
ways that human cognitive abilities are so connected.
Linguistics and Evolution, furthermore, embraces the idea that particular
areas of the (cortical) brain that specialize in processing certain tasks are not
isolated, autonomous areas, but rather areas where certain processes converge
to synthesize a particular behavior or set of behaviors. Now, denying the iso-
lation of brain functioning does not mean denying that brain regions may
become domain dominant. How particular cortical areas come to specialize
in processing the tasks that they do is a developmental question. How these
areas might become relatively stable in an individual and reliably reproduced
across populations and down lineages is an evolutionary question. Both will be
answered by investigating the kinds of tasks that individuals attempt to accom-
plish under particular conditions. Nevertheless, stability, when it comes to the
brain, is a relative notion: the brain is far more plastic than the modular theory
allows, and while everything is not connected to everything else, many brain
areas can and do communicate, so to speak, with one another. A case in point is
the McGurk and MacDonald article “Hearing Lips and Seeing Voices” (1976),
in which they relate the results of their experiments which are famous for set-
ting up a conflict between simultaneous aural and visual inputs that force a
resolution on the perceptual level: a video of a mouth pronouncing a velar stop
syllable [ga] with the audio saying a bilabial stop syllable [ba] will produce
the perception of an alveolar stop [da]. The authors note that knowing the trick
does not alter the perceptual effect, nor does it lessen with repeated viewing
and hearing.
Even more surprising, perhaps, is the similar kind of perceptual resolution
based on higher-level information that has been shown to have a gender bias. In
a set of experiments exploring the perception of the boundary between alveo-
lar fricative [s] and palatal fricative [ʃ], which are largely distinguished by the
lower auditory frequency of the latter, Strand (1999) found that perception
of [s] or [ʃ] depends on the listener’s expectations of the size of the speaker’s
vocal tract. Borrowing the methodology of her study from that of McGurk
and MacDonald, Strand produced a lip-synched video of a man and a woman
uttering one of ten synthetic steps from [s] to [ʃ] embedded in the carrier words
sod and shod and reported that participants assigned a different phonological
category to the same acoustic signal based on the speaker’s gender. Given an
ambiguous auditory signal, the listener would hear a male speaker say [s] and
a female speaker say [ʃ]. The explanation is that the listener’s perception is
affected by the expectation that a man would have a longer vocal tract, thus
leading the listener to believe that the stimulus lay in the higher frequency
range for that speaker, while a female would have a shorter vocal tract, thus
leading the listener to believe that the stimulus lay in the lower frequency range
of the [s/ʃ] continuum. Interestingly, physical characteristics that are culturally
conditioned, such as hairstyles that are considered more or less prototypically
feminine or masculine, can exaggerate this perceptual effect.
Whereas the McGurk and Strand experiments show the degree to which
visual information can affect auditory perception, the work of Tanenhaus et al.
(1995) stresses the role of vision in the processing of syntax. In this set of
experiments, participants were faced with two displays, each consisting of one
or two apples, two towels, a napkin and a box. In one display there was only
one apple, and it was on a towel, while the other towel was empty; in the other
display, there were two apples, one on a towel and one on a napkin, with the
second towel also empty. For each display, the participants were given two sets
of commands: Put the apple on the towel in the box, and Put the apple that’s
on the towel in the box. The first command is ambiguous, in that the prepos-
itional phrase on the towel could either modify the noun phrase the apple, thus
specifying the location of the object to be picked up, or denote the destination,
that is, the location where the apple is to be put. As the participants reacted to
the commands, their eye movements were recorded with head-mounted eye-
tracking systems. It was found that in both the one-referent and two-referent
contexts, the sequence and timing of eye movements relative to the nouns in
the speech stream did not differ for the ambiguous and unambiguous instruc-
tions. This led Tanenhaus et al. to conclude that in natural contexts, “people
seek to establish reference with respect to their behavioral goals during the
earliest moments of linguistic processing. Moreover, referentially relevant
nonlinguistic information immediately affects the manner in which the linguis-
tic input is initially structured” (1995: 1634). As they note, their conclusion
runs counter to the modular model of syntactic processing in which supposedly
only syntactic information is available to be processed in order to understand
the sentence.
In other words, there is evidence to show cross-modal communication
between and among our sensory apparatuses. Although the extent of this
cross-modal communication is still a matter of investigation, it is clear that
our auditory and visual areas exchange lots of information regarding speech
perception and production. This very much puts at issue the notion of informa-
tion encapsulation. As Woodward and Cowie have observed, “informational
encapsulation often seems to be task-relative. Whether mechanism or brain
region A is influenced in its internal processing by information or processing
in mechanism or region B may vary depending on the tasks that A and B are
engaged in” (2004: 327). Indeed, McIntosh and Lobaugh (2003) and Stephan
et al. (2003) demonstrate that it is not the nature of the incoming stimuli that
determines how or where an individual processes that information. It is rather

the nature of what the individual is to do with the stimulus that is determina-
tive – a rather nice demonstration of the neurological effects of pragmatic
interpretation.
All of these points and studies are consistent with the neurological premises
of languaging as a (non-hearing-centric, non-speaker-centric) orienting behav-
ior. They challenge the conventional understanding of the roles of hearing and
seeing in a speaker-oriented linguistics. They reinforce the phenomenological
understanding that perception is not something that happens to us; it is some-
thing we do, and our doing is fundamentally affected by the particulars of our
individual experiences in the (physical, gendered and cultural) environments
in which we language.
Let us now ask the frank question: just how much deeper into psychology or
neurobiology did Chomsky go than either Saussure or Bloomfield? Chomsky
went inside, all right, but he did not get his hands dirty with messy mental
processes. Instead, he proposed neat algorithms and subroutines, i.e., soft-
ware, that can perform its work independently of the physical platform on
which it is implemented, i.e., hardware/wetware. In so doing, not only does
the Chomskyan program perform its work only on inputs and outputs within
the system but independently of any feedback to a final output, it also: (i) pre-
cludes attention to neurobiology; and (ii) makes an end run around psych-
ology altogether. There is very little that is psychological about the modularity
hypothesis of the mind/brain and of the language faculty. Rather, they are
extremely convenient models of the insides that justify the descriptions that
linguists have produced of the patternings to be found at various levels of lin-
guistic analysis – phonological, morphological, syntactic, etc. – in the absence
of investigating what other neurological and/or psychological factors these
patternings might be layered upon or in. The modularity hypothesis, in short,
seems to be an interior projection of the perceived disciplinary need, traceable
to Saussure, to find a space (i.e., theoretical construct) that is the linguist’s
alone. Thus, the postulation of autonomous modules is a result of the premises
of twentieth-century disciplinary theory and practice, not a result of twentieth-
century psychological and neurological research.
The interdisciplinary problem with the Chomskyan model of language was
that it was intended for export – and then, indeed, often was exported. It was
Chomsky’s notion that the hypotheses and discoveries of what he called the
linguist-psychologist would set the stage for further inquiry into brain mecha-
nisms. He likened this notion to the ways “those who spoke of the valence of
oxygen or the benzene ring were speaking at some level of abstraction about
physical mechanisms, then unknown … [and thereby] set the stage for fur-
ther inquiry into underlying mechanisms [of chemistry]” (1988: 7). While
Chomsky’s notion might have sounded good, it received almost immediate dis-
missal from a scientist such as Francis Crick, who complained:
It comes as a surprise to neuroscientists to discover that many psychologists, linguists
in particular, have very little or no interest in the actual brain, or at least what goes on
inside it. The brain, they feel, is far too complicated to understand. Far better to produce
simple models which can do the job in an intelligible manner. That such models may
have little resemblance to the way the brain actually behaves is not seen as a serious
criticism. (1989: 131)
It is as if Chomsky handed neuroscientists a model of a closet with the clothes
hanging neatly on their hangers, but when neuroscientists look into the brain
they see a closet crammed with a wild tangle of wires. Now twenty-five years
later, distributed language phonologist Robert Port has observed: “For speech
scientists outside linguistics (cognitive psychology, engineering, speech sci-
ence, neuroscience, etc.), what is at stake in ‘what is the relevance of linguistic
theory and linguistic terminology for my research?’ The surprising answer …
turns out to be not nearly as much as we thought” (2010: 45). In short, the
twentieth-century refusal to engage with neurology in a meaningful way has
left the discipline with broken interdisciplinary bridges.
Modularity in biology
Before leaving the concept of modularity behind, we need to engage with the
concept in biology, where it is proving to be useful in explanations of develop-
ment and evolution. The purpose is to understand how developmental modules
differ from mental modules, so that we do not trip over our terminology in
moving from one area of science to another. What is counted as a module in
developmental and evolutionary studies can occur at multiple levels of bio-
logical organization from single molecules up to entire organisms, and they are
typically organized hierarchically, so that modules exist on a smaller physical
scale within individual, larger-scale modules. A module is considered to be a
module when the constituent components: (i) are physically proximate, such
as the cells within a single embryonic domain of the central nervous system, or
temporally correlated, such as the expression of the genes in the network that
operates during vertebrate muscle, eye and ear development; and (ii) contribute
to a common trait, e.g., the fully developed Drosophila wing, or to a common
process, e.g., the activation of the pathways of intercellular signaling proteins
in the morphogenesis of the Drosophila eye; and variations on the resulting
trait or process occur repeatedly within an organism and across the world’s life
forms, like the basic helix-loop-helix protein.
The second criterion seems to reflect what is counted as a connected trait,
described above, i.e., it is a trait that is the opposite of what a language mod-
ule is proposed to be. Furthermore, in finding what is common to these two
Modularity in biology 83
criteria, we might say that biologists are apt to recognize a module when a
few simple components can be seen to give rise to a range of life forms. This
makes a module what one might call a higher-level version of the fact that:
(i) the amino acid sequence of any protein is determined by the sequence of
bases in some region of a particular nucleic acid molecule; (ii) that twenty
different kinds of amino acid are commonly found in protein; and that (iii)
four main kinds of bases occur in nucleic acid. The description of the way in
which twenty or more things are determined by a sequence of four things of
a different type is commonly called the genetic code. In short, the recognition
of a module is another version of the observation that a relatively small set of
building blocks seems to underlie the enormous complexity of life.
We find ourselves back to Chapter 1 and the concept of code theorism,
which is based on the idea that language makes infinite use of finite means.
The American linguist Dwight Bolinger goes so far as to say: “The linguistic
code is like the genetic code – so much so that geneticists refer to ‘the syntax
of the DNA chain’” (1981 [1968): 5).2 Two points fall into place. First, we can
now see that neither Chomsky’s language faculty nor its subparts qualifies for
the status of a module in the biological sense, since mental modules are not
theorized to be connected traits. Mental modules are, moreover, exclusive to
humans, and the language faculty module is exclusive to only one aspect of
human cognition/behavior. By way of contrast, a module in developmental and
evolutionary studies refers to a region of strong interaction in an interaction
matrix that is required for the organism to develop, whereas a mental module in
linguistics refers to a feature of an adult brain whose properties or algorithms
are pre-given, so-called innately specified, i.e., are not developmental results.
The discussion of modularity helps us to see that descriptions of language/
ing and descriptions of life are not reducible one to the other. Another way of
saying this is: from the point of view of a DST linguistics, languaging is not
characterized by a scarcity of means. Languaging means are seen as vast, rich
and content-filled. The perception of the scarcity (or economy) of means – pre-
ferred sentence pattern forms, for instance – to produce an infinity of construc-
tions exists only in the domain of description of certain observers.
Second, while it is uncontroversial that humans the world over have created
large, flexible, and finely tuned orienting systems for coordinating behavior
whose parts can be used over and over and in various ways, the repetition of the
2
Typical of those who think of DNA as language are comments such as “The deciphering of the
DNA code has revealed our possession of a language much older than hieroglyphics, a language
as old as life itself, a language that is the most living language of all” (Beadle and Beadle 1966:
207) and “the ribosomes are nonspecific structures, which function as machines to translate the
nucleic language, carried by the messenger, into the peptidic language, with the aid of the trans-
fer RNAs” (Jacob 1966: 1472), which includes a later interpretation of a particular region in a
“nucleic text” as a “punctuation mark” (1475).
infinite-use-of-finite-means truism keeps alive the tradition of thinking about

language as a personal possession. The unspoken idea is this: just as my DNA
is mine personally, so must my language be. In addition, there is a third term
hovering around the equation, namely life, not only in the idea that DNA is the
most living language of all but also in the idea that DNA itself is the life giver.
In the collapsed DNA-life-language chain of associations, then, it is too easy
to slide into the perhaps only vaguely articulated idea that personal biology and
the ability to produce and understand an infinite array of utterances are what
produced, i.e., gave life to, language in the first place. Instead, in Linguistics
and Evolution, individual human biology is seen to have been dramatically
affected by the instantiation and development of languaging in our lineage,
and the ability to produce and understand an infinite array of utterances is not
in itself generative but rather epiphenomenal. In other words, the relationship
between life and language is opened up and a space is made between them. If
we see the current parallel of the infinite use of finite means is only the small
point of their intersection, then we can begin to explore both phenomena in the
fullness of their separate extensions.
We need a fresh perspective – literally – on our subject matter.
The observer
A problem has bedeviled the discipline of linguistics in the past several hun-
dred years, and it is the relationship between the observer, i.e., the linguist,
and the observer’s subject matter, i.e., language/ing. In Chapter 1, we saw that
the discipline of linguistics began with a methodology that was established
by nineteenth-century philologists who had nothing to do with the languages
they studied except study them. They were, in a sense, pure observers. We saw
how the philological spirit and practice, which served the purposes of histor-
ical reconstruction, was assimilated into the spirit and practice of synchronic
descriptions in twentieth-century structural linguistics. For the purposes of
writing a grammar, this is entirely appropriate, as has been said, and Leonard
Bloomfield, for one, was very aware that he was an observer. He and his fellow
structuralists understood that the relationship between the observer, i.e., the
linguist, interested in describing the structure of a particular language, and the
speaker, the one producing utterances, is one of precedence and of perspective:
precedence in that a linguist can only observe, record and/or describe after the
fact what a speaker produces; and perspective in that a linguist is not limited
to observing, recording and/or describing the productions of any one speaker
but can take a more global perspective. However, Bloomfield realized that this
supposedly global perspective would always be limited since any one observer
(or group of observers) could not necessarily have access to all utterances that
all speakers of a language have uttered.
The observer 85
In addition to effecting a dramatic turn to the inside, Chomsky collapsed

the relationship of the observer and the observed. He thereby solved the
Bloomfieldian problem of limitation by putting the conditions for generating
a complete grammar inside every human being and freed the generativists to
indulge their own linguistic intuitions about what utterances (sentences) could
and could not occur (were grammatically well-formed or not). The rules they
wrote made explicit the grammatical relations among the sentences. Bringing
the observer inside the speaker marked a logical end-point for philologism, for
now the observer’s descriptions are put inside the head of the languaging living
being and serve to account for the languaging living being’s behavior. When
the observer is both outside and inside the subject of study, we behold the phe-
nomenon of double vision.
The potential for double vision exists in a number of social and biological
sciences, for instance psychology and neurology. It is a problem that, when
unrecognized, causes a conflation of two domains that should, for scientific
purposes, be kept distinct: one is the domain proper to the observer, and the
other is the domain proper to the creature/behavior under observation. Double
vision has a strong presence in Western thought and has produced, among
other things, the pre-Darwinian psychology known as representationalism:
a view of the nervous system as an instrument whereby the organism gets
information from the environment out of which it builds a representation
of the world that it then uses to compute behavior adequate for its survival.
Representationalism is the psychology that floats code theorism, muddles the
study of meaning and makes formal syntax possible. In recent decades, its
popularity has waned in artificial intelligence/robotics circles, but it remains
the explanatory default setting for some (but certainly not all) cognitive scien-
tists and linguists, who find it difficult to speak of behavior in terms other than
information- and computer-processing metaphors.
A clear articulation of the problem of double vision begins with William
James and his chosen field of psychology. He notes, first, that the conflation
between the domain proper to the psychologist and the domain proper to the
psychological phenomenon under observation may occur when the observer
projects the knowledge he has as an observer into “the mental fact about which
he is making his report.” James red-flags this tendency:
“The Psychologists’s Fallacy.” The great snare of the psychologist is the confusion of
his own standpoint with that of the mental fact about which he is making his report …
The psychologist … stands outside of the mental state he speaks of. Both itself and its
object are objects for him. Now when it is a cognitive state (percept, thought, concept,
etc.), he ordinarily has no other way of naming it than as the thought, perfect, etc. of that
object. He himself, meanwhile, knowing the self-same object in his way, gets easily led
to suppose that the thought, which is of it, knows it in the same way in which he knows
it, although this is often very far from being the case. (1981 [1890]: 195)
When translating this into the field of linguistics, we might say, for instance,
that the linguist writing a treatise on syntax can all too easily conflate his
understanding of syntactic operations with the syntactic productions of any
given speaker who has provided the data for the linguist’s analysis. As James
notes, the two ways of knowing may be very far apart.
Second, the conflation can flow from the assumption that the “mental state
studied” is conscious of itself as the psychologist is conscious of it. James
cautions:
The mental state is aware of itself only from within; it grasps what we call its own con-
tent, and nothing more. The psychologist, on the contrary, is aware of it from without,
and knows its relationship with all sorts of other things. What the thought sees is only its
own object; what the psychologist sees is the rest of the world. We must be very careful
therefore, in discussing a state of mind from the psychologist’s point of view, to avoid
foisting into its own ken matters that are only there for ours. (1981 [1890]: 195–96).
Here we might say that a linguist has a view both inside and outside a language,
since it is an object that for a linguist can be seen in its relationship with other
sorts of things, e.g., the historical development of that language. Speakers, on
the other hand, only interact with their languages to get themselves through
their day, and they have their languages arranged in their heads only in the way
that they have them so arranged. Now, James exhorts the psychologist not to
foist onto the mental state matters that only exist for the psychologist, which
suggests that James wants the psychologist to add nothing, and that is a good
warning to keep in mind for linguists as well. However, I would also note here
that what speakers know about their languages surely also exceeds the lin-
guist’s ability to grasp descriptively.
Perhaps it is fitting that a philosopher/psychologist with a great interest in
the nervous system, such as James, would have called attention to the problem
of double vision so precisely and so long ago. Now these many decades later,
neurobiologists do not study the nervous system in terms of its taking “an out-
side in.” They are well accustomed to the idea that the operation of the nervous
system is expressed in its connectivity or structure of connections such that
an organism’s behavior arises from its nervous system’s internal relations of
activity. Certainly the nervous system is responsive to outside stimuli, but those
stimuli can only modulate the internal relations of activity. Reintroducing the
terminology of the neurobiologists Maturana and Varela, we now have the per-
spective and the vocabulary to state that while stimuli can trigger responses in
the organism, they cannot specify how the organism changes. Only the internal
organization of the organism specifies the changes, and these specifications
occur according to the organism’s own ontogenic drift, that is, the history of
interactions with its environment, and as a result of the phylogenic drift of the
lineage to which the organism belongs, which helps determine which triggers
The observer 87
the organism will be responsive to, that is, the evolutionary story that can be
told about an organism.
Let us say that someone observes the following situation: person A, seated
at a table, points to a salt shaker out of his reach and asks of person B, “Please
pass the salt.” B obliges by handing A the salt and says, perhaps, “Here it is.”
In describing this situation, it seems perfectly normal to say that B’s words and
actions were caused by A’s words and actions. However, when the observer
says that, s/he is operating wholly in his/her own domain of description. This
domain has its uses, for instance, in a court of law: Lawyer’s question: “Did
A say ‘Pass the salt’ to B, thereby causing B to pass A the salt?” Witness’s
response: “Yes, it looked to me as though A’s words caused B’s actions.”
However, such a description will have limited use in a neurobiological inves-
tigation of B’s nervous system. Strictly speaking, A’s words triggered events
in B’s nervous system. The so-called causes of B’s actions lie in the interior
organization of B’s nervous system, which has been shaped by a history of salt-
passing events (among others) and which belongs to the lineage of languaging
living beings.
A quite different account of what can be called the normal use of language
is offered by Chomsky. In Language and Problems of Knowledge, he invokes
Descartes and his followers who:
observed that the normal use of language is constantly innovative, unbounded, appar-
ently free from control by external stimuli or internal states, coherent and appropriate to
situations; it evokes thoughts in the listener that he or she might have expressed in simi-
lar ways in the same situations. Thus in normal speech one does not merely repeat what
one has heard but produces new linguistic forms – often new in one’s experience or even
in the history of the language – and there are no limits to such innovations. (1988: 5)
How this state of linguistic affairs that is “apparently free from control by exter-
nal stimuli or internal states” comes about we do not know, nor does Chomsky
attempt to explain it. We are told only that it comes about and that this state of
freedom is freer still. Chomsky continues:
Furthermore, such discourse is not a series of random utterances but fits the situation
that evokes it but does not cause it, a crucial if obscure difference. The normal use of
language is thus free and undetermined but yet appropriate to the situation, and it is
recognized as appropriate by other participants in the discourse situation who might
have reacted in similar ways and whose thoughts, evoked by this discourse, correspond
to those of the speaker. (1988: 5)
The possibility that the situation might evoke but not cause an utterance is an
obscure difference indeed, but, again, Chomsky does nothing to attempt to
dispel the obscurity. In the absence of explanation on his part, I will say that
the word evoke was chosen to suggest that an utterance can be appropriately
produced in a situation because the individual has within him/her the totality of
the environment whose contents he/she anticipates. In this explanatory frame-

work, A’s words do not cause B’s response, and B’s behavior is explained in
terms of the language faculty’s modules whose contents exceed the experi-
ences B could have in his/her lifetime and which are evoked at the appropriate
times. Chomsky’s turn to the interior was dramatic, indeed. For Maturana and
Varela and all other researchers pursuing a developmental understanding of
behavior, B’s behavior is explained precisely in terms of B’s experiences with
the world.
In contrast to Chomsky’s cause/evoke distinction, Maturana and Varela dis-
tinguish between the nervous-system-centric term specify and the observer-
oriented term trigger. They would say that A’s words trigger a response in B
whose neurological particulars are specified by the history of interaction B has
had with A and with similar such words. As usual, care must be taken to distin-
guish among epistemological frameworks in which similar terms or the same
terms are used in the elaboration of very different points of view. For example,
certain nativists also use the term trigger. They theorize the adult organism to
be a result of a kind of predetermined epigenesis and speak of the environment
as the trigger for an internally driven push-from-the-genes development of that
organism. The idea here is that these environmental triggers are like the drops
that water a plant. Surely the plant needs water, just as the developing organ-
ism needs environmental inputs, but the so-called real action is believed to
take place in the interior of the plant. When it comes to what is usually called
language acquisition, nativists believe that the infant comes into the world
equipped with all sorts of so-called pre-knowledge of the forms of language,
e.g., nouns, verbs, certain kinds of agreement, subordination patterns, and they
believe that the role of the particular language environment the infant is born
into is to simply trigger that knowledge, like drops of water falling on a plant.
No role is allowed for the progressive effects of earlier developmental events
and processes on later developmental events and processes. That is, and by
way of contrast to the approach outlined by Maturana and Varela (and then
also by Vygotsky), a nativist account does not entertain the possibility that the
changes specified in the organism’s interior as the result of some prior trigger
will affect how the organism’s interior specifies the effects of some later trig-
ger. In other words, for nativists, there is no possibility to see that the succes-
sion of triggers plays a role in structuring the organism’s interior. Furthermore,
nativists assume a fixed environment that preexists the organism that occupies
it. A rather static verticality is operating here: an organism appears on earth and
literally grows up in the midst of an already-given environment whose features
the organism anticipates.
We turn now to the framework in which the contrast specify/trigger has oper-
ational coherence and in which the term environment is seen to be in a dynamic
relationship with the organisms that occupy it.
Autopoiesis 89
Autopoiesis
In Autopoiesis and Cognition: The Realization of the Living (1980 [1972]),
Maturana and Varela help us to understand what James might have meant when
he said that “the mental state is aware of itself only from within.” To do this,
Maturana and Varela first make a conceptual shift away from understanding
living beings in terms of an enumeration of properties and toward an under-
standing of living beings in terms of circular organization and the operational
closure of the nervous system. They then give presence to this conceptual
shift in our languaging by offering the term autopoiesis to guide us in our
understanding.3
To begin with, Maturana and Varela were interested in coming to grips with
the notion of the living, and they noticed that traditional ideas about the notion
resorted to making lists of criteria such as chemical composition, the ability to
move, the ability to reproduce, responsiveness to stimuli and the like. Maturana
and Varela changed perspective and determined that merely asking the ques-
tion of how to recognize a living being indicates that the questioner has an idea,
even if implicitly, of its organization. To speak of living beings in the first place,
then, presupposes something in common among them; otherwise, they would
not be put in the same class designated with the name living. Maturana and
Varela proposed that the kind of organization that characterizes living beings is
a continually self-reproducing organization, an autopoietic organization. They
realized that they needed a term:
without a history, a word that could directly mean what takes place in the dynamics of
the autonomy proper to living systems. Curiously, but not surprisingly, the invention of
this word proved of great value. It simplified enormously the task of talking about the
organization of the living without falling into the always gaping trap of not saying any-
thing new because the language does not permit it. (1980 [1972]: xvii)
The concept of autopoiesis allows us to dispense both with the rationalist-
realist position that cognition consists in taking an outside world in and with
the solipsistic despair that we are only our interiors and that these interiors are
figments of our imaginations. The dynamics proper to the interior organization
of the living being are not in direct contact with an exterior, but these dynam-
ics have been significantly shaped by the living being’s history of interactions
with that exterior.
A key feature of an autopoietic unity is that it produces its own border. We
can say that a car has an internal organization and that we can replace parts
when needed to keep it running. We can say that human beings have an internal
organization as well, and we can and do replace parts in humans to keep them
3
Readers might find the autopoietic glossary at the end of this chapter helpful in explaining the
terms used in this section.
Dynamics Boundary
(metabolism) (membrane)
Figure 3.1 The dynamics of the boundary
The exterior arrow indicates the The interior arrow indicates circular
unity producing its own border organization with operational closure
of the nervous system
triggers → → unity’s effect on environment

→
environment
Figure 3.2 The dynamics of a living being with a nervous system
running. However, one of the differences between a car and a human is that the
various human parts, beginning with our cells, produce their own borders, in
addition to the most extensive border that we produce which is our skin.
Figure 3.1 here is meant to show that the membranous boundary is not a product
of cell metabolism in the way that fabric is a product of a fabric-making machine;
this membrane not only limits the extension of the transforming network that
produced its own components, but it participates in this network. Maturana and
Varela schematize an autopoietic unity in its environment as in Figure 3.2.
Maturana and Varela, like James, distinguish how an observer or a group of
observers interprets an autopoietic unity’s behavior from the internal workings
of the autopoietic unity itself. The observer has access not only to the nervous
system of the autopoietic unity but also to the structure of the unity’s environ-
ment. In addition, the observer has access to other observers’ descriptions. As
a result, it seems to the observer that the unity knows the outside world the way
the observer does. However, in the domain of the observed, the observed – that
is, the autopoietic unity – operates primarily in relationship to its interior. If
one imagines a car on the road as a body and the navigator at the wheel as the
nervous system, what the behaving body with a nervous system must do is to
maintain certain relationships between the dials and pedals as it tootles along.
That is, most of what the navigator has to deal with are the maneuvers that
maintain correlations internal to the vehicle/nervous system. The visual sys-
tem provides a case in point. The retina can only modulate – but not specify –
the state of the neurons in the lateral geniculate nucleus, which is the most
Autopoiesis 91
Figure 3.3 Structural coupling between living beings with nervous systems
prominent region of connections between the retina and the central nervous
system, because fully eighty percent of the internal interconnections – from
the hypothalamus, the superior colliculus, locus coeruleus, reticular nucleus
of the thalamus – come together at the same time (Maturana and Varela 1992
[1987]: 162). The overall job of the nervous system is to secure the synthesis
of behavior, not the representation of the world.
Autopoiesis can refer to: (i) the functioning of the nervous system (in
autopoietic unities which have a nervous system), which generates a behav-
ioral dynamic through generative relationships of internal neural activity in its
operational closure; and (ii) the condition of the living system which, at every
level, is organized to generate internal regularities. When interaction between
organisms acquires a recurrent nature, a reciprocal structural coupling results.
This gives rise to a new phenomenological domain, which may become par-
ticularly complex when there is a nervous system. The phenomena arising
from these structural couplings are of a third order and are usually investigated
under the heading “social behavior.” Here the environment of one living being
includes the existence of another living being (Figure 3.3).
Languaging as an orienting behavior arises when two or more organisms
who interact recurrently enough to generate a social coupling are able to gen-
erate a new domain that coordinates, let us say, actions of actions. Actions of
actions is not a phrase that is easily understood, and it is surely not the usual
way to describe how two or more languaging living beings organize themselves
with respect to each other’s words and deeds. The new domain could also be
described as the creation and elaboration of linked behavioral-cognitive loops.
What are we talking about here? Let us say that two or more living beings
are following one another down a path for some purpose – an observer might
identify that purpose as hunting, picking berries or looking for a good watering
hole to cool off. This action, again from an observer’s point of view, counts
as one domain of consensual coordination of actions, and this domain is very
common in the animal world. Now, these same organisms going down the path
might also coordinate their actions around another domain of actions – call
this new domain actions of actions – when they organize themselves around a
sound sequence such as Watch out for the low-hanging branch or even simply
Branch. The organisms that hear such a sound sequence and experience a new
orientation within their cognitive domains are called languaging living beings.
The cognitive orientations that take place with respect to the sound sequence
[waʧaʊtfɔrðələʊhæŋɪŋbrænʃ] and any accompanying gestures are of a dif-
ferent nature than the properties of the sound sequence itself and are a result of
the history of interactions that those living beings have with such a sequence. A
less complicated way of saying this is that all animals can organize themselves
with respect to a branch, while only languaging living beings can organize
themselves so thoroughly with respect to a sound sequence [brænʃ].
Humans are not the only creatures to have such social couplings, and any
creatures that we observe to have such behavior can be said to generate what
Maturana and Varela call a linguistic domain, because an observer is apt to
describe these recurrent interactions in terms of what the “meaning” of the
interaction is for each of the organisms rather than in terms of the ways they
operate to attain their respective poieses (= internal dynamics of the structural
coupling of the interacting organisms). Such behaviors constitute the basis for
language, but they are not yet what we observers call language. These kinds of
behaviors are to be distinguished from behaviors that arise in the development
of the organism independently of its particular ontogeny. The so-called lan-
guage of bees can be invoked here as an example of the latter kind, in that bee
dances are behavioral coordinations that result from bee phylogeny. However,
to the extent that bee dances exhibit group variations or dialects ontogenically
determined, we might consider them to be mixed ontogenic/phylogenic behav-
ioral coordinations.
The domain of language appears when the operations in a linguistic domain
result in coordinations of actions about actions pertaining to the linguistic
domain itself. We are in language, or, as we could now say, we language only
when, through a reflexive action, we make a linguistic distinction of a linguis-
tic distinction. A lengthy quote from Autopoiesis and Cognition helps here:
Linguistic behavior is orienting behavior; it orients the orientee within his cognitive
domain to interactions that are independent of the nature of the orienting interactions
themselves. To the extent that the part of its cognitive domain toward which the orientee
is thus oriented is not genetically determined and becomes specified through interac-
tions, one organism can in principle orient another to any part of its cognitive domain
by means of arbitrary modes of conduct also specified through interactions. However,
only if the domains of interactions of the two organisms are to some extent comparable,
are such consensual orienting interactions possible and are the two organisms able to
develop some conventional, but specific, system of communicative descriptions to ori-
ent each other to cooperative classes of interactions that are relevant for both. (1980
[1972]: 30)
The primary social results of such utterances are the cognitive turnings – we
may also call them tunings – that the organisms experience. Note that these
types of cognitive events count as types of social events.
Autopoiesis 93
The developmental story that can be told about the turning and tuning of
these organisms will be very different from the one that follows from a code-
theoretic account of our subject matter. To begin with, a developmental linguis-
tics foregrounds the primacy of action, real-world events, which means that it
foregrounds the peculiarities and particularities of bodies moving through time
and space. What we do relates to what we know, which, in turn, relates to what
we are, which feeds into what we do, and so forth. Maturana and Varela have
formulated this relationship as the “unbroken coincidence of our being, our
doing, and our knowing” (1992 [1987]: 25). Consequently, a developmental
systems linguistics recognizes that actions as well as actions of actions (= lan-
guaging) are never separated, such that the languaging sequences of a languag-
ing living being coming into languaging encounters are always embedded in
larger contextual motions and embodied in e-motioning. It follows that: (i) the
basic “unit” of investigation is not a sentence or a word or a phoneme but rather
an entire sequence of activity, for instance, a routine, such as a “passing the
salt” routine; (ii) languaging does not primarily serve a categorizing function
but rather an integrative function as it links cognitive activities across different
sequences of activity; this is also called socialization; and (iii) in its integrative
function, languaging has necessarily produced phylogenic effects in the neuro-
biology of humans in general and necessarily produces ontogenic effects in the
neurobiology of individuals.
In the epistemological framework in which languaging as an orienting behav-
ior has coherence, establishing knowledge becomes less about uncovering a
truth already there and more about bringing forth a world in our languaging
together. Fleck has called this a harmony of illusions, which can be described
as: a stable and effective congruence of ideas, observations and manipula-
tions; a consonance among beliefs, perceptions and action. Now, none of these
descriptions comes down to the matching of beliefs or statements to an inde-
pendent external reality, that is, to a classic commonsense idea of truth. But
“neither does any of them come down to a denial of external reality in the sense
of something other than our own statements, beliefs, or experiences with which
we interact” (Smith 2005: 51).
The concept of autopoiesis, furthermore, moves us definitively out of the
framework of code theorism and into a conceptual and discursive terrain
where languaging as an orienting behavior has operational presence and
coherence. What can be said about the distinction between code theorism and
languaging?
In code theorism, language – theorized as either a socially shared langue
or genetically shared competence – is presupposed to be the means by which
thoughts, propositions, ideas, feelings, intentions, any and all other states get
understood as they are transferred from one individual’s head to another’s. The
primary social purpose of both langue and competence is so-called knowledge
sharing. However, they differ in how that purpose is carried out cognitively.
Saussure’s langue, for instance, serves first the cognitive purpose of knowl-
edge creating, that is, creating ideas in the first place, because without langue,
“thought is a vague, uncharted nebula. There are no pre-existing ideas, and
nothing is distinct before the appearance of langue” (Saussure 1959 [1916]:
111). By way of contrast, the primary cognitive purpose of competence seems
to be first as a sorting mechanism, a way to categorize the already known out-
lines of the objects in the world, be they mental or physical, and second as a
generative mechanism, a way to assemble those categorized parts into thoughts,
propositions, ideas, etc. Competence is a vast, abstract mechanism for describ-
ing the ability an individual has to bring to articulation the thoughts, feelings,
propositions, etc. that are the exclusive property of that individual. Whether or
not or why the individual actually opens his or her mouth to articulate those
thoughts is another story.
An autopoietic framework makes no particular assumption about the func-
tion – that is, purpose – of languaging. The function of knowledge sharing is
too narrow and does not do justice to the range of functions that languaging
behaviors can be observed to serve. The phrase is also too hands-off, which
is to say that it is too friendly – which is, however, not to say that languaging
is somehow fundamentally unfriendly. Rather, what I am calling into ques-
tion, once again, is the assumption that conversation, as such, is or ever was
the prototypical languaging interaction or that languaging exchanges that can
be seen to be, in some sense, cooperative by the very fact of their occurrence
are necessarily mutually beneficial to all involved, as the phrase knowledge
sharing implies. This is also to note that languaging is necessarily invasive. It
is also nearly inescapable. Extreme cases of escape through deliberate depriv-
ation of a languaging living being from a languaging environment would be
the baby locked away from human contact, which is considered inhuman (the
sad case of Genie), and the convicted criminal in solitary confinement, which
is considered punishment. The cases of the hermit and the Trappist monk
belong to certain kinds of social and religious traditions, meaning that those
who remove themselves from other languaging living beings or take a vow of
silence have chosen a particular relationship to the languaging environment,
one that requires special effort to maintain. In sum, the most neutral term to
apply to languaging interactions is that they are human and have whatever
characteristics that may be understood by this designation.
One of the most startling aspects of autopoiesis, as Maturana and Varela
conceive of it, is the way it operates at the very core of the reproductive proc-
ess. According to them, everything that happens to make cellular fracture take
place in the reproductive plane is a result of autopoietic dynamics. No external
force or agent is needed. Even in the core of the reproductive process, then,
everything happens in the unity as part of the unity. There is no separation
Autopoiesis 95
between the reproducing system and the reproduced system. It is remarkable to

realize how the concept of autopoiesis – with its first-order, second-order and
third-order unities – allows us to move seamlessly from biology to sociology
without encountering disciplinary boundaries or without having to move from
one building on campus to another.
The concept of autopoiesis also opens up, once again, the relationship
between languaging and life which is conflated in the code-theoretic frame-
work. Instead of reducing the description of life to a reflection of a particular
conception of language (the syntax of DNA), autopoiesis – self-production – is
understood to be constitutive of the living. As long as a living being maintains
autopoiesis, the living being continues to live; autopoiesis ends only at the
moment of death. For languaging living beings, it is the case that a significant
aspect of the way they maintain their autopoiesis is through the activity of lan-
guaging. Languaging is a different activity than autopoiesis, but languaging is
indispensable to the realization of the autopoiesis of languaging living beings.
Languaging itself is not an autopoietic unity, by the way, although some theo-
rists have treated it as such. As Maturana and Varela have phrased it:
Language was never invented by anyone only to take in an outside world. Therefore, it
cannot be used as a tool to reveal that world. Rather, it is by languaging that the act of
knowing, in the behavioral coordination which is language, brings forth a world. We
work out our lives in mutual linguistic coupling, not because language permits us to
reveal ourselves but because we are constituted in language in a continuous becoming
that we bring forth with others. (1992 [1987]: 234–35)
So, in living our cognitive domain, we humans bring forth a world of continu-
ous becoming in our languaging together.
The purpose of running through the brain/body parts entertained by Saussure,
Bloomfield and Chomsky was to suggest, first, that the progress through the
twentieth century was not inevitably toward the inclusion of more brain/body
parts into linguistic descriptions and, second, that the number of brain/body
parts included was determined by the theoretical perspective taken by the lin-
guist. What is added by the inclusion of Maturana and Varela’s approach into
linguistics is a way to understand and speak about the interiors of various crea-
tures without reducing them all to instances of the same interior, one that takes
the (objective) outside world in and re-presents it as it is. Instead, by incorp-
orating autopoiesis into our linguistics, we are led to the idea that all creatures
live their cognitive domains, i.e., they bring forth a world through their interac-
tions with their environments, and that creatures that belong to the same lin-
eage will live their cognitive domain in similar (but not the exact same) ways.
It follows from this – here breaking down the term cognition – that all
(sighted) creatures live their visual domain: humans live their visual domain,
just as chimps and orangutans live their visual domains, which are different
but historically linked to ours. Having said this in just this way, it should come
as no surprise to discover that there are significant differences in the visual
cortices of various primates, as mentioned above. For instance, a layer in the
human visual cortex has a complex mesh-like pattern not found in other pri-
mates and is now hypothesized to explain humans’ superior ability to detect
objects against a background (Balter 2007: 1209).
The interiors of languaging living beings can now be seen as multidimen-
sional spaces – as opposed to two-dimensional maps – constantly being knit
by the languaging environments they find themselves in. That is to say that
dichotomized sets of oppositions – langue/parole, competence/performance –
not only do not help us in our investigation of the interior, they hinder our
investigation in that they propose an adult-oriented construct (langue, com-
petence) theorized without reference to brain structures or processing that is
nevertheless supposed to have explanatory validity in how living beings come
to be languaging living beings. Thus they severely reduce the interior spaces in
which the linguist can move comfortably, and they narrow the linguist’s job to
that of what might be called drawing the lines between the theorized features
of that particular construct and either the brain structures (rarely investigated)
or the brain mechanisms that supposedly process them. Just as these constructs
preempt a developmental perspective, so they also make the linguist’s job of
assimilating neuroscientific and psycholinguistic research more difficult, rather
than easier.
We now have an enlarged sense of the interior in which the full-bodied lan-
guaging living being engages with another’s or many others’ languaging, be
it written or spoken or gestured, at any given instant. Attention, recollection,
consciousness, memories shifting aimlessly below recollection, infrasounds
affecting our moods, feelings, perceptions, and judgments as lingering effects
of previous languaging experiences – all these interior states and dispositions
become relevant for understanding how any instance of languaging shapes and
is shaped by histories of interactions. In even the simplest of engagements,
histories may be called forth, such that, as political scientist William Connolly
puts it, “Y both listens to X and relives a sheet of his own past as he does. His
attention extends to past and future alike, not as the past was when experienced
but as it arrives now in the interstices of this conversation.” And, some lines
later, “for a memory to cross to the point at which it is called up as recollec-
tion, it must have already arrived near the tipping point. It was not plucked
immediately out of a huge storehouse covering every event in Y’s life. It was
crystallized from a wave of memory called into the vicinity by the conversa-
tion” (2005: 100). There is something in Connolly’s poetry that calls forth the
Jamesian vocabulary of psychic overtone, suffusion, fringe or even halo that he
applies to our words and verbal images that suggests how deeply rhizomed in
our synaptic tangles are the previous and ongoing effects of our languaging.
Autopoiesis 97
This is the poetry of a conception of our subject matter released back into
body and world, unfettered from the narrow theoretical jackets into which it
has been variously put. It links Proust to neuroscience, as suggested in the
title of a recent account of an fMRI study where subjects passively read both
odor-related words and neutral language items: “Reading Cinnamon Activates
Olfactory Brain Regions” (González et al. 2006).
Autopoietic glossary
Allopoietic unity a system, such as a car factory, which produces something
other than itself.
Autopoietic unity a system that produces itself; a network of processes of
production, transformation and destruction of components that produces
the components that: (i) through their interactions and transformations
regenerate and realize the network of processes (relations) that produced
them; and (ii) constitute it as a concrete unity in space in which they ex-
ist by specifying the topological domain of its realization as such a net-
work.
A first-order autopoietic unity is a cell.
A second-order autopoietic unity is any organism whose structure is char-
acterized by cell aggregates in close coupling, i.e., a metacellular. Many
animals, e.g., human beings, plants, fungi, etc. are metacellulars.
A third-order autopoietic unity is a unity made up of metacellulars, e.g., a
society. It has been extensively debated whether the definition can be
applied literally to social systems or only metaphorically (Mingers 2002,
Luhmann 1990). Maturana and Varela themselves remain noncommittal
whether they wish to use the term to describe social systems as “living.”
Behavior something ascribed to a system by an observer observing its inter-
actions with its environment. Behavior is not “in” the system.
Cognition the basic ability to respond to environmental events, aka perturba-
tions.
Congruence the result of interactions between a system that has been shaped
by its interactions with its environment over time and an environment re-
ciprocally shaped by its interactions with a system. This is different from
the concept of fitness, that is, when a system is said to be “fit” for its en-
vironment. Congruence arises from the changes that the system and the
environment prompt in each other.
Consensual coordination reciprocal structural coupling that is characteristic
of social life; any particular organism is a member of a social unity as
long as it forms part of that reciprocal structural coupling.
Drift we talk about genetic drift to refer to the genetic change that is pro-
duced from generation to generation; we talk about ontogenic drift to
refer to the history of structural changes of a system in its domain of exis-

tence that follows a course configured instant by instant following the path
in which, in its interactions, it conserves organization and adaptation.
Knowledge we admit knowledge whenever we observe an effective (or ade-
quate) behavior in a given context, i.e., in a realm or domain we define by
a question (implicit or explicit).
Language behavioral coordination through mutual and recursive structural
change; languaging is language as ongoing and situated activity.
Observer a system that through recursive interactions with its own linguistic
states may linguistically interact with its own states as if with representa-
tions of its interactions.
Operational closure a feature of metacellular systems whose identity is
specified by a network of dynamic processes whose effects do not leave
the network.
Organization the relations that must exist between the components of a sys-
tem in order for it to be in a certain class; the organization of an autopoi-
etic unity must remain consistent in order for it to continue to be that
unity.
Perturbation an outside event.
Phenomenological domain defined by the properties of the unity or unities
that constitute it, either singly or collectively through their transforma-
tions or interactions. Thus whenever a unity is defined or a class of unities
is established that can undergo transformations or interactions, a phenom-
enological domain is defined.
Phenomenology the domain of all the phenomena defined in the interactions
of a class of unities.
Reproduction when a unity undergoes a fracture that results in two unities of
the same class, that is, their defining relations are repeated in their respec-
tive extensions. The resulting unities are not identical with the original
one, nor are they identical with one another, but they belong to the same
class as the original; they have the same organization. If we take any cell
in its interphase and fracture it, we do not get two cells; we simply destroy
the original cell. Similarly, if we fracture a pen, a radio, a coin, a human
being, a declaration of human rights, we do not get two pens, two radios,
two coins, etc.; we get broken objects.
Specify a nervous-system-centric term that describes a sequence of events
within an autopoietic unity. For instance, when a living being’s hand
comes in contact with a hot stove, the living being’s nervous typically
specifies a series of modifications so that the proper muscular sequences
can be enacted in order for that living being to regain its preferred state
(‘preferred state’ in this case would mean that the hand is no longer in
contact with something hot).
Autopoiesis 99
Structure the actual components and their relations that make up a unity at
any given time; the structure of an autopoietic unity can change over time.
Systems are structure-determined, and a plastic structure is one that can
be affected by outside events, i.e., can be perturbed. By what and to what
extent a system can be perturbed is determined by its structure.
Structural coupling whenever there is a history of recurrent interactions
leading to the structural congruence between two (or more) systems. It
arises as a result of the plasticity of the structures of biological systems
and the plasticity of the environment. When two or more autopoietic sys-
tems perturb one another over time, behavioral coordinations tend to re-
sult. When two or more autopoietic systems of the same class perturb one
another over time, what is usually called “social behavior” arises.
Trigger an observer-oriented term to denote an aspect of the environment
the observer judges relevant to understanding a living being’s behavior.
An observer observing a living being quickly removing a hand from a
hot stove will identify the hot stove as the trigger for that living being’s
behavior. (It is the interior of that living being that specifies what the liv-
ing being does next.)
Unity that which is distinguishable from a background, the sole condition
necessary for existence in a given domain. The nature of a unity and the
domain in which the unity exists are specified by the process of its dis-
tinction and determination; this is so regardless of whether this process is
conceptual or physical.
Part II
A developmental systems linguistics

4 Evolutionary scenarios I: the standard story
and the self-reproductionist script
Linguistics and Evolution foregrounds the twin and intertwined temporal

dimensions of phylogeny and ontogeny in order to explain how a living being
becomes a languaging living being. Exploring the full extensions of this double
dimension of time requires: (i) an examination of a living being in its appropri-
ate setting, which for human beings ever was and always is the social, languag-
ing setting; (ii) an understanding of the mosaic-like development over time of
languaging abilities in both the species and the individual so that it has now
become a connected trait; and (iii) an appreciation of the psychological depth
and complexity of languaging behaviors and effects in the mature languager.
This is to say that, in a developmental systems linguistics, languaging is given
a kind of cognitive breathing room not found in the last fifty years of formalist
theorizing. A developmental systems linguistics is furthermore grounded in the
sense that aspiring languagers and mature languagers alike have two feet on the
ground (or are on all fours or in some conveyance) as they move around in their
world, acting on it and being affected – in their being, knowing and doing – by
the consequences of their actions.
We begin to explore the evolutionary dimension by moving through the
standard story for the evolution of language as it has played out in the past sev-
eral decades, and this for the perhaps very obvious reason that an exposition
of the more or less standard, more or less formalist story will bring into high
relief the features of the alternative evolutionary account promoted in the next
chapter. The standard story can be characterized by the following points that
are either assumed or made explicit:
(i) the purpose of communication is transfer of some entity (knowledge,
thoughts, feelings) from the mind of a speaker to the mind of a listener;
(ii) the mind is representational, and for some philologist-linguists the mind
may also be modular;
(iii) our knowledge of the world is accurate;
(iv) our descriptions of language reflect both the accuracy of our knowledge
and the means by which we are able to communicate;
(v) language is localized in the brain;
103
104 Standard story and the self-reproductionist script
(vi) the capacity for language is innate;

( vii) the capacity for language is a species-specific property; and
(viii) the evolutionary story to be told about how language came to be the way
current linguistic theory proposes it to be does not require the identifica-
tion of a non-linguistic capacity or behavior out of which the linguistic
domain emerged.
In the first three chapters, various aspects of a number of these points have
been examined. This chapter is engaged primarily with point viii), because it
is around this fundamental discontinuity that turns not only the whole of the
standard story in linguistics but also the self-reproductionist script of the (first)
neo-Darwinian synthesis.
Continuity versus discontinuity: a multilayered dichotomy

Studies on the question of the origin of language are often split into two groups:
continuity theories, which propose that the emergence of human language/ing
abilities can be told in terms of evolutionary processes that have operated on
the kinds of communicative display systems found in vertebrates in general
and in primates in particular, and discontinuity theories, which adhere to the
idea that so-called human language is different from other communicative dis-
play systems in the animal kingdom, i.e., that there is not just a quantitative
difference between, say, vervet monkey vocalizations and human language but
rather a qualitative and unbridgeable difference between them. The dichotomy
can be posed as a question: did human language develop out of animal commu-
nication systems in a continuous line of development or was there a break?
The practice of dividing studies in this way is sometimes said to begin with
Biological Foundations of Language (1967) by Eric Lenneberg, and that is
probably true for the second half of the twentieth century. However, the conti-
nuity/discontinuity dichotomy opened up much earlier, only a few scant years
after Darwin published On the Origin of Species in 1859, namely in 1861 when
the Oxford University philologist Friedrich Max Müller stated that “Language
is our Rubicon, and no brute will dare to cross it.” Since that pronouncement,
the term Rubicon has echoed in evolutionary discourse in an intraverbal chain
to the present.1
As Müller’s somewhat indignant declaration suggests, people in the Victorian
era readily understood the implications of Darwin’s evolutionary hypothesis,
and some immediately objected to the displacement of man’s privileged place
in the animal kingdom. New ideas, particularly ones that impact deeply held
1
A sampling: Critchley 1959: 289; Hill 1974: 137; Bickerton 1990: 104; Gibson 1994: 99; Deacon
1997: 23; the title of Ch. 6 in de Waal 2001; Bickerton 2009: 49.
Continuity versus discontinuity 105
convictions, are rarely adopted instantaneously. Uptake for them tends to

spread through a discipline and/or larger social group over a long period of
time. For instance, not all mid-sixteenth-century astronomers abandoned over-
night the Ptolemaic system, which posited that the cosmos revolved around
the earth the moment that Nicolaus Copernicus came forward to state the earth
revolved around the sun. The physicist Max Planck is known for this quote: “A
new scientific truth does not triumph by convincing its opponents and making
them see the light, but rather because its opponents eventually die, and a new
generation grows up that is familiar with it.” William James, with customary
intelligence and eloquence, stated the matter thusly:
Our minds thus grow in spots; and like grease-spots, the spots spread. But we let them
spread as little as possible: we keep unaltered as much of our old knowledge, as many
of our old prejudices and beliefs, as we can. We patch and tinker more than we renew.
The novelty soaks in; it stains the ancient mass; but it is also tinged by what absorbs it.
Our past apperceives and co-operates; and in the new equilibrium in which each step
forward in the process of learning terminates, it happens relatively seldom that the new
fact is added raw. More usually it is embedded cooked, as one might say, or stewed
down in the sauce of the old. (1981 [1906]: 78)
This quote does double duty as a description of the conservative nature of indi-
vidual cognitive ontogeny and as a description of the way evolution works –
and a Fleckian, non-Kuhnian understanding of the way science proceeds.
James understood that evolution does not produce novelties from scratch and
that even in the development of our mental processes – whether phylogenic or
ontogenic – the action of natural selection/experience works on what already
exists. From the tips of our toes to the tops of our heads, James might say, our
physical and mental lives are products not of engineering but of patching, of
tinkering. Summing it all up, we could say: “Fresh experience grafted onto old
knowledge makes new knowledge.” Or, reverse angle: “Old habits die hard.”
The point here is that, from the Greeks to the Victorians and still often today,
human languaging abilities were believed by many to be the shining monu-
ment to human rationality, the marker of the so-called cognitive divide between
humans and “the brutes,” in Müller’s phrase. Human languaging abilities are
often held up and out, even by those who subscribe to evolutionary theory, as
a kind of last bastion of human uniqueness in the animal kingdom. The phi-
losopher Alfred North Whitehead was, for instance, willing to accept behavior-
ism’s ability to account for human behavior provided one made an exception of
verbal behavior (Skinner 1957). And it was not by chance that in his review of
the behaviorist psychologist B. F. Skinner’s Verbal Behavior (1957), Chomsky
(1959) tossed in repeated references to rats pressing bars – namely thirteen
times in a thirty-one-page review – in order to discredit Skinner’s approach to
verbal behavior, although no rats appear in Skinner’s book. Of interest in the
present chapter is the fact that Skinner both opened and ended his account of
verbal behavior with recognition of the need to have an evolutionary frame-

work for it. For factoring in evolutionary considerations from the start Skinner
should get credit, and on this point he stands in stark contrast to the generativ-
ists, who have backed into the problem of evolution and are now hampered by
a theory of language that does not play easily on the phylogenic axis.
Indeed, Chomsky has long taken a discontinuist position with respect to
being able to account for his subject matter in an evolutionary line one could
trace back to simpler communicative systems. “It seems rather pointless,”
Chomsky wrote rather early on, “to speculate about the evolution of human
language from a simpler system” (1968: 62). Or, again: “[I]t is quite senseless
to raise the problem of explaining the evolution of human language from more
primitive systems of communication that appear at lower levels of intellec-
tual capacity” (1972: 67). The reluctance to accept a genealogical relationship
between “human language” and “primitive systems” led Chomsky to more
general skeptical pronouncements about the role of natural selection in shaping
the human mind: “In fact, the processes by which the human mind achieved its
present stage of complexity and its particular form of innate organization are
a total mystery … It is perfectly safe to attribute this development to ‘natural
selection,’ so long as we realize that there is no substance to this assertion, that
it amounts to nothing more than a belief that there is some naturalistic explan-
ation for these phenomena” (1972: 97). Chomsky clearly wants to preserve the
mystery of his chosen field of study, which he can do only if he evacuates the
concept of natural selection of all substance, dismissing it as “nothing more
than a belief that there is some naturalistic explanation for these phenomena.”
At this point it must be stated that no one thinks that natural selection is any-
thing other than a naturalistic explanation. That is precisely its purpose: to
offer naturalistic explanations for how species have come to be the way they
are without recourse to non-naturalistic explanations (e.g., the hand of God).
Now, despite Chomsky’s pronouncements in the 1960s and 1970s, linguists
and interested others continued to desire an explanation of the relationship
between (some version of) human language and evolution. In Language and
Problems of Knowledge, Chomsky chose to respond to his detractors thusly:
It is sometimes argued that even if we succeed in explaining properties of human lan-
guage and other human capacities in terms of an innate biological endowment, nothing
has really been achieved because it remains to explain how the biological endowment
developed … This is a curious argument. By the same logic we might argue that noth-
ing is explained if we demonstrate that a bird does not learn to have wings but rather
develops them because it is so constructed by virtue of its genetic endowment … True,
it remains to account for the evolution of language, wings, etc. The problem is a serious
one, but it belongs to a different domain of inquiry.
Can the problem be addressed today? In fact, little is known about these matters.
Evolutionary theory is informative about many things, but it has little to say, as of now,
about questions of this nature. The answers may well lie not so much in the theory of
Continuity versus discontinuity 107
natural selection as in molecular biology, in the study of what kinds of physical sys-
tems can develop under the conditions of life on earth and why, ultimately because of
physical principles. It surely cannot be assumed that every trait is specifically selected.
(1988: 166–7)
On one point one must agree with Chomsky: his understanding of human
language cannot be accounted for by the usual continuist evolutionary frame-
work which features a gradualist, descent-with-modification story about the
phenomenon in question. His position on that score is completely consistent
with the rest of his theory in that his view of language phylogeny is intimately
bound up with his account of what it means to know a language and his account
of how humans learn language, that is, that they do not learn it at all. This view
is part and parcel – both cause and consequence – of his broader epistemologi-
cal framework, a framework that comprises a network of mutually supporting
assumptions: that is, if one views language as a discrete species-specific capac-
ity, it makes sense that it is physically realized in the mind as a discrete mental
organ. In turn, if one accepts that language is realized as a discrete mental
organ unique to the human species, it makes sense to doubt that it has any
homologues or analogues in the natural world.
As a terminological aside, a homologue is historically defined. A homo-
logue is not defined by its intrinsic properties but in terms of its evolutionary
origins, while an analogue is functionally defined. Analogous characteristics
resemble one another but are not descended from a common ancestral form;
rather, they are seen to serve the same evolutionary purpose. Note that biologi-
cal categories are historically defined but that any particular feature or trait can
be classified either by evolutionary homology or by analogy – the latter now
often called homoplasy. An arm is a vertebrate limb, i.e., homologous, because
it is united in a pattern of ancestry and descent with all the other vertebrate
limbs. The flight feathers of birds and the wing membranes of bats are analo-
gous. Classifications by homology and classifications by analogy are typically
viewed as complementary to one another.
Chomsky has long asserted that people’s use of language (as he defines it)
only incidentally serves the utilitarian goals of communication and is prima-
rily and presumably originally an autonomous competence used gratuitously
to express propositions about the world. If that is the case, then it does, indeed,
seem rather pointless to speculate about its development from simpler systems.
Human grammar, such as Chomsky sees it, has a unique and universally idi-
osyncratic logic all its own that is not explicable as an adaptation, since there
cannot by definition be any preexisting trait or behavior out of which grammar
could have developed. The naturalistic explanation that seems to suggest itself
here, given Chomsky’s appeal to molecular biology in the above quote, is the
one he offers in perhaps his most oft-quoted position on the subject which
appears in Pinker’s The Language Instinct (1994: 362–63), as well as in at
least one popular introductory linguistics textbook (Fromkin et al. 2003: 60).
It furthermore exists in an intraverbal chain that extends through Jackendoff by
way of Newmeyer (1998), who cites Chomsky (1975) to the effect that:
We know very little about what happens when 1010 neurons are crammed into some-
thing the size of a basketball, with further conditions imposed by the specific man-
ner in which this system developed over time. It would be a serious error to suppose
that all properties, or the interesting properties of the structures that evolved, can be
“explained” in terms of natural selection. (Jackendoff 2002: 234)
Moving forward thirty years, Chomsky’s most recent statement underscores the
mystery in the unsolved problems that he notes have been raised for hundreds
of years: “Among these are the question how properties ‘termed mental’ relate
to ‘the organical structure of the brain,’ problems far from resolution even for
insects, and with unique and deeply mysterious aspects when we consider the
human capacity and its evolutionary origins” (2006: 185). This is his conclud-
ing comment to the chapter entitled “Biolinguistics.” The term mystery seems
entirely appropriate to describe the Chomskyan evolutionary state of affairs.
Continuing continuity versus discontinuity

Steven Pinker’s work is pertinent in this chapter because he began with a com-
mitment to arguing against Chomsky’s evolutionary views while upholding
Chomsky’s linguistic views. Pinker originally undertook the task of (attempt-
ing to) account for the Chomskyan view of language within the framework of
contemporary evolutionary theory. The Pinker/Chomsky divide gives us the
first fold in the multilayered continuity/discontinuity dichotomy, one that might
have been easily overlooked in the opening description of continuity theories.
To repeat: continuity theories propose that the emergence of human language
can be told in terms of evolutionary processes that have operated on the kinds
of communicative display systems found elsewhere in the animal kingdom. The
italics bring the fold out of the shadows. In other words, the opening ques-
tion – did human language develop out of animal communication systems in a
continuous line of development or was there a break? – joins two issues. The
one pertains to whether or not language has evolutionary antecedents in the
animal kingdom, specifically in the primate lineage. (Chomsky and Pinker both
say No.) The other pertains to whether or not language can be accounted for in
the usual terms of evolution. (Chomsky says No; Pinker says Yes.) The bottom
line here is that Pinker wishes to demonstrate that Chomskyan linguistics is not
a case of special creation theory. Because, in Pinker’s view, language has con-
ferred great survival value on the species, it can be accounted for by the usual
means by which any trait that is deemed to have survival value can be accounted
for, namely by natural selection. And because human language shows evidence
of design, this evidence in and of itself means that language must have been
Continuing continuity versus discontinuity 109
produced by Darwin’s theory of natural selection, since natural selection is the

principal explanation for “any other complex instinct or organ” (1994: 333).
It is the case that even Skinner would agree with Pinker that (Skinner’s ver-
sion of) our subject matter, namely verbal behavior: (i) has an evolutionary
explanation – Skinner viewed operant conditioning as a version of natural selec-
tion on a foreshortened time scale; and (ii) does not derive from vocalizations –
cries of alarm, mating calls, aggressive threats, etc. – that are widely found in
the animal kingdom, since Skinner observes that such behaviors are difficult,
if not impossible, to modify by operant conditioning (1957: 463). When they
are modifiable, they are so “apparently only with respect to the occasions upon
which they occur or their rate of occurrence” (1981: 502). Commentary on this
particular issue got a charge from the widely cited Hauser, Chomsky and Fitch
(2002) article, along with the two-pointed responses by Pinker and Jackendoff
(2005) and Jackendoff and Pinker (2005). It is the case that, on the one hand,
Hauser, Chomsky and Fitch, and on the other Pinker and Jackendoff, all agree
that primate vocalizations are not a homologue of (the researchers’ respective
versions of) human language. By the way, even Max Müller was against the idea
that natural selection could have converted the cries of animals into meaningful
speech. Good. It seems that everyone agrees that the evolutionary story to be
told about human language/ing is distinct from the one to be told about primate
vocalizations – although they agree on that point for different reasons.
The dissociation between human language and primate vocalization seems to
be well justified on the cognitive-processing side of things. Although a monkey
homologue to Broca’s area has been identified, namely a place where audio-
visual mirror neurons respond to actions independently of whether the monkey
sees, hears or performs those actions (Kohler et al. 2002), the neural basis of
vocalization is distinct. In non-human primates, the cortical area that medi-
ates vocalization is close to the midline in the cingulate region in the limbic
association cortex, while in humans, it has moved out to include the premotor
cortex on the left side (Crow 2005: 145). Certainly, there are distinctions to be
made between non-human primate and human vocalizations: non-human pri-
mates do not use communicative signals the way humans do, namely to orient
the orientee within his/her cognitive domain. However, non-human primates
do use their vocalizations to affect the behavior or motivational states of their
conspecifics. This point leads Tomasello to note that “the deep evolutionary
roots of human language lie in the attempts of primate individuals to influence
the behavior, not the mental states, of conspecifics” (2003: 11). In other words,
the lack of a current homology between human languaging and non-human
primate vocalization does not mean that human languaging abilities have no
phylogenic relationship to other primate activities and behaviors.
Bringing these points together, MacNeilage agrees with all of the above and
shifts the discussion by noting that looking for precursors to speech in primate
vocal communication is looking in the wrong place. Rather, he locates the

prelinguistic source of human vocalizations in the extremely common primate
visuofacial communicative gestures, such as lipsmacks, tongue-smacks and
teeth chatters. The beauty of this account lies in their similarities to speech in
that: (i) the “up-and-down movements of the mandible are typically repeated
in a rhythmic fashion in the lipsmack, just as in syllables”; (ii) they typically
occur in social situations, particularly positive ones; (iii) they involve eye con-
tact and sometimes appear to occur in what humans would call turn-taking; and
(iv) they are sometimes (but not always) accompanied by vocalization (2008:
93–94). As such, these visuofacial communicative gestures could have been
exapted for vocal-auditory communicative purposes, which would account not
only for how speech got started, but also why human speech and primate vocali-
zations are now processed differently. What is most intriguing in MacNeilage’s
account is the discontinuity he proposes between the actually very restricted
set of sounds that are favored in babbling (which is opposed to many linguists’
belief that in babbling all the sounds of the world’s languages can be found)
and the syllable preferences found in the world’s languages. The point here
is that, despite the proposed discontinuity in syllable patterns between bab-
bling and speech, MacNeilage takes the continuist, descent-with-modification
approach to explaining how speech got started. His work will return in the next
chapter. For now, we still have the standard evolutionary story to tell.
Returning to Skinner, I note that he proposes verbal behavior arose with the
extension of operant control to the vocal musculature and a supply of vocal
behavior not strongly under the control of stimuli or releasers – a position
which is not inconsistent with MacNeilage’s. The emergence and then develop-
ment of this behavior extended the ways humans could influence the behavior
of one another. For instance, it greatly extended the help one person receives
from others, and from this small step the whole of human sociocultural history
can be said to blossom:
By behaving verbally people cooperate more successfully in common ventures. By tak-
ing advice, heeding warnings, following instructions, and observing rules, they profit
from what others have already learned. Ethical practices are strengthened by codify-
ing them in laws, and special techniques of ethical and intellectual self-management
are devised and taught. Self-knowledge or awareness emerges when one person asks
another such a question as “What are you going to do?” or “Why did you do that?” The
invention of the alphabet spread these advantages over great distances and periods of
time. They have long been said to give the human species its unique position, although
it is possible that what is unique is simply the extension of operant control to the vocal
musculature. (Skinner 1981: 502)
So, while human languaging abilities and non-human primate vocalizations

have diverged significantly enough to be now appropriately classed as sepa-
rate phenomena, it is nevertheless the case that it is not difficult to understand
human languaging to have arisen from general ways that primates affect the
behavior of their conspecifics. On this score, the difference between Skinner
and Tomasello seems to be that Tomasello believes that human languaging
interactions evolved from “a procedure to getting something done” to “an invi-
tation to share attention” (2003: 33), whereas Skinner does not seem to envi-
sion an instance of verbal behavior in a (possibly disinterested?) invitation- or
(equal?) sharing-mode. For Skinner, the evolution has been toward more suc-
cessful cooperation, as he describes it in the above passage, which suggests a
change in quantity (to wider and more refined procedures) rather than a change
in quality (from procedural to invitational).
Although nowhere in Pinker’s and Jackendoff’s writings do they support a
behavior(ist)-inflected account of languaging abilities, their joint critiques of
Hauser, Chomsky and Fitch (2002) marshals a wealth of information about just
how much the perceptual and production machinery of human speech differs
from that of our nearest relatives: perception of consonants by formant transi-
tions and of vowels by frontness and backness, phoneme discrimination, breath
control, articulatory control, etc. (Pinker and Jackendoff 2005: 206–12). They
argue convincingly that these differences must be due to the long, behavioral-
cognitive interlacing of languaging in the species (my terms, not theirs). They
do not go so far as to say that humans “live their auditory domains, just as other
species live theirs,” but that is my uptake, which is further bolstered by recent
work on the human supralaryngeal vocal tract. Lieberman, for one, argues that
humans do not perceive linguistic distinctions directly from the spectrum of the
acoustic signal but are rather able to extract formant frequencies that are con-
stantly subject to distortion from environmental sources. Although Lieberman
shows what human speech perception and production have in common with
other mammalians’ abilities, he points to vocal tract normalization as a particu-
lar attribute of humans (2006: 121).
Pinker and Jackendoff further distance themselves from the Hauser, Chomsky
and Fitch account when they invoke evidence of the lack of recursion that has
been reported in the Amazonian language Pirahã (Pinker and Jackendoff 2005:
216). Indeed, recursion is the last stand of a generativist-inspired linguistics,
and Hauser, Chomsky and Fitch hold on to it until the end, suggesting that
recursion may have evolved “to solve other computational problems such as
navigation, number quantification, or social relationships” (2002: 1578).2 This
suggestion is made to save the long-held Chomskyan position that the language
2
The topic of recursion is rather fraught at the moment, with respect both to the somewhat slippery
definition of it and to its presence or absence in Pirahã (see Nevins et al. 2009 and Everett 2009).
The position taken in Linguistics and Evolution is similar to the one proposed by Marianne
Mithun that “recursive structures are in a sense epiphenomenal, the products of a host of cog-
nitive abilities” (quoted in Everett 2009: 438). A special, forthcoming issue of The Linguistic
Review is devoted to it, and so interested readers may turn to that issue for more information.
faculty is not a direct product of evolution but rather a by-product of other abil-
ities such as navigation, which they further speculate is a modular system, one
that in other animals is impenetrable with respect to other systems. However,
in humans, “during evolution, the modular and highly domain-specific system
of recursion may have become penetrable and domain-general. This opened
the way for humans, perhaps uniquely, to apply the power of recursion to other
problems” (2002: 1578). How the domain-specific system of recursion may
have become penetrable and domain-general is not addressed.
Nowhere in Hauser, Chomsky and Fitch (2002) is there a hint of one human
interacting with or affecting another – bodily, vocally or otherwise. Yet another
logical end-point has been reached here. Call it Extreme Minimalism; and
Pinker and Jackendoff do take the gloves off when they take issue with the
Minimalist Program in the last part of their response to Hauser, Chomsky and
Fitch (2002). In their return response to Pinker and Jackendoff (2005) enti-
tled “Appendix: The Minimalist Program,” Chomsky, Hauser and Fitch (2005)
explain that the Minimalist Program has succeeded in reducing the complexity
and variety of the earlier phrase structure and transformation rules to more
elementary and far-reaching principles. They then comment: “These moves
have also had the positive effect of excluding impossible structures and enhan-
cing explanatory force – and, accordingly, improving the prospects for study of
the evolution of language, always a background consideration from the earliest
days” (2005: 6). The gesture toward the “study of the evolution of language”
is not only empty given the historical record of Chomsky’s writings, it also
marks yet another deferral – not unlike his continuous deferrals of psychology
and neurology – since it is only the prospects for the study of the evolution of
language that are improved. The Chomskyan program is no farther along the
evolutionary road than it was when it started out fifty years ago.
Recognizing dead-ends is always refreshing, and in that spirit, scholars
interested in evolutionary questions have urged that we all move on from the
continuity/discontinuity debate. Aitchison (1998), in particular, has reason-
ably argued that complex research topics often begin by exploring dichotomies
which then naturally fade as the complexities become unraveled. She points
out persuasively that different degrees of continuity exist between the three
main constituents in the physical instantiation of our subject matter: auditory
mechanisms, articulatory mechanisms and a brain that coordinates them. Here
she notes that, in the primate lineage, there is great auditory continuity between
humans and other primates, moderate brain continuity, and great articulatory
discontinuity. (The auditory continuity is only relative to the great articulatory
discontinuity, since there is subtle, but marked auditory discontinuity, as stated
above.) In addition, she cites cases of ambiguous continuity, where the first
step is a necessary condition for a new development but not a sufficient con-
dition: it enables a leap to be made, but does not inevitably cause it. This, as
she notes, makes the dichotomy over simple. Finally, in recent decades, newer
issues have come to the fore: was language as we know it today the result of a
so-called slow haul or could it have popped, as some say, into existence rela-
tively quickly? Was there a single emergence in the species or multiple? In
sum, for Aitchison, the old question about continuity versus discontinuity has
now outlived its usefulness, and it is time to put it aside.
The spirit and details of Aitchison’s critique are apt, and it is useful to
problematize the very question of how we are to think about continuity and
discontinuity. However, Linguistics and Evolution presses the continuity/dis-
continuity dichotomy into service: (i) because theorists involved in it have not
yet put at issue the very phenomenon that we are trying to account for evolu-
tionarily; and (ii) because it is implicated in an even bigger ball of wax which
can be summarized by the following five questions:
(i) Is human language/ing continuous with or discontinuous from the com-
municative interactions of other animals, e.g., non-human primates?
(ii) Is human language/ing continuous with or discontinuous from other
human social practices, e.g., communicative bodily behaviors such as a
social smile, hand gestures, bowing, other communicative oral emissions
such as deprecatory coughs?
(iii) Are the cognitive/neurological phenomena associated with human lan-
guage/ing continuous with or discontinuous from other cognitive/neuro-
logical phenomena?
(iv) Is the history of the human species continuous or discontinuous? That
is, was there “a time when” the species was in a state of so-called nature
and subsequently crossed some sort of threshold into so-called culture,
thereby marking a before and after period?
(v) Is the history of organisms continuous or discontinuous? That is, is evolu-
tion continuous, with minute changes accumulating in each generation, or
is evolution discontinuous, involving production of totally new forms?
These five questions are not necessarily asked explicitly. Rather implicit
answers to them inform many an evolutionary approach. Not surprisingly, the
differing assumptions necessarily produce differing evolutionary frameworks.
Thus, in order to tease out the differences among these frameworks, keeping
open discussion of the continuity/discontinuity divide is warranted.
A reading of Pinker and Bloom (1990) gives a sense of how the continuity/
discontinuity divide traces the broad outlines of an entire evolutionary story.
For Pinker and Bloom, language is a capacity unique to humans that differs
from all other forms of communicative interactions in the animal world. Its
differences lie in its complex design features, e.g., those of its structural levels
(syntax, morphology, phonology), which reflect the purpose for which it was
designed, namely “the transmission of propositional structures through a serial
interface” (1990: 707). Language is, thus, appropriately described in terms

exclusive to itself, e.g., noun and verb and dependent clause. The capacity for
human language is, furthermore, linked; (i) to localized centers in the brain
that operate modularly according to innate, richly structured, content-specific
mechanisms; and (ii) to the genome. This capacity is, in a sense, doubly disem-
bodied in that it is pushed both neck up and into the genes. Since the capacity
for language has no non-linguistic neurological reflexes in the human brain or
behavioral precursors in the primate lineage, it follows that the history of the
linguistic lives of our human ancestors is marked by discontinuity.
The necessity of this historical linguistic discontinuity is most readily
apparent when Pinker and Bloom try, among other things, to explain how it
is that speakers converge on what they call “form–meaning pairings.” Pinker
and Bloom are aware that there are various possible rationales for any form–
meaning pairing. “[A]nd that,” they write, “is precisely the problem – differ-
ent rationales can impress different speakers, or the same speaker on different
occasions, to different degrees.” Having made that observation, what they do
in the very next sentence is to efface those interesting differences: “But such
differences in cognitive style, personal history, or momentary interests must
be set aside if people are to communicate” (1990: 718), i.e., if people are to
“transmit propositional structures through a serial interface.” For Pinker and
Bloom, the setting aside of those differences was precisely the work of nat-
ural selection, the encoding of the language code on the genes, so to speak.
The genetic “encoding” of the “propositional structures” of language ensures
that all members of a community of speakers will converge on the same form–
meaning pairings.
This line of argument necessarily entails a time when those differences had
not yet been put aside, when humans were learning language non-or-quasi-
genetically and more or less communicating, but not “really” communicating.
During that time, Pinker and Bloom concede that the communicative functions
of grammar must have played “an important role,” but now these functions
may have relatively little role, that is, they “may play no role in acquisition”
(1990: 720). It has been the work of natural selection, then, to gradually deter-
mine (“hard-wire”) more and more grammatical knowledge (“correct connec-
tions”) innately, because “with more correct connections to begin with, it takes
less time to learn the rest, and the chances of going through life without having
learned them get smaller” (1990: 723). So, there was a time when commu-
nicative utterances were more open to interpretation than they are today. For
Pinker and Bloom, the closing of the interpretive gap – the setting aside of the
differences – has been the work of natural selection.
If the above outline is correct, a discontinuity within the history of the species
necessarily exists. On the one hand, we have evolutionary time, i.e., the time of
the continuous-genetic-setting-aside-of-the-differences. On the other hand, we
have cultural history, i.e., the time marked by the changes that resulted from
the ability to transmit propositional structures, aka progress. This discontinu-
ity captures the striking differences in human habitat and lifestyle that existed,
say, between the Pleistocene, when human evolution was presumably at work,
and those developments in human culture and civilization that have occurred
since that time. In other words, although Pinker and Bloom are committed to
the traditional adaptationist idea that human language was “selected for” in
the usual, gradualist (i.e., continuist), descent-with-modification fashion, they
would nevertheless answer “discontinuous” to questions (i)–(iv) above. Given
this, it is not surprising that Pinker and Bloom do not identify some behavior,
e.g., a precursor to a “setting-aside-of-the-differences” behavior, out of which
the communicative exchange of propositional structure arose.
As for question (v), Pinker answers it indirectly in The Language Instinct
when he identifies his work with “the mainstream in modern evolutionary biol-
ogy,” namely “George Williams, John Maynard Smith, and Ernst Mayr” (1994:
359). Mayr makes the intersection between discontinuity and evolution expli-
cit: “The origin of new species, signifying the origin of essentially irreversible
discontinuities with entirely new potentialities, is the most important single
event in evolution” (1963: 11). Tree structures are particularly good at repre-
senting the splitting process, where each branching point is meant to represent
one or more modifications that is/are deemed significant enough to warrant the
drawing of a new line. They thus illustrate lineages defined by descent with
modification. It is worth noting that the only figure to appear in On the Origin
of Species is Darwin’s tree diagram. As such, it ranks as biology’s premier
ideogram, and the absence of an alternative model warrants the assumption
that vertical transmission has been the dominant process over all of evolution.
By Darwin’s time in the mid nineteenth century, the tree model was already
in use by philologists to represent the historical relationships among daughter
languages in a language family. It so happens that the late eighteenth-century
philologists borrowed the tree model from early eighteenth-century textual
critics who were concerned to determine the lineages of biblical and classical
manuscripts. Textual critics knew that a shared modification in a set of manu-
scripts – i.e., a mistake or an innovation introduced by a scribe in copying a
text that was then reproduced in all subsequent manuscripts of that text – was
a mark of common ancestry. In other words, Darwin’s concept of evolution as
descent with modification, where shared modification (mutation, innovation,
mistake) is the mark of common ancestry, was an extension to the new disci-
pline of evolutionary biology of a principle already established in philology
that had earlier borrowed its methodology from textual criticism. Here, then, is
another example of the tight interweaving of biological, linguistic and textual
notions. The tree structure is a useful enough model for representing biologi-
cal genealogies, as long as we realize that it: (i) is insufficient for capturing
significant continuities that can be observed when one undertakes a more com-
plete understanding of linguistic facts and phenomena as well as of biological
processes; and (ii) obscures the recognition of patterns of resemblance that
occur for reasons other than “vertical” descent with modification, such as envi-
ronmental constraint and convergence, parallelism or reticulation.
The figure of the tree has come under recent critique in the field of molecular
phylogenics, which asserts that phylogenic trees should, in principle, be defin-
able in terms of molecular information alone. Now, if it is the case that the trees
drawn by molecular phylogenicists match those proposed by organismal biol-
ogy, then this agreement would provide independent confirmation that their
trees corresponded to something of the reality of macroevolution. However,
the process of lateral gene transfer – i.e., non-vertical descent, which is both
more frequent and much more extensive among the prokaryotes (organisms
without cell nuclei) than the eukaryotes (organisms with cell nuclei) – signifi-
cantly scrambles the model of the tree, with the trunk and the branches sprout-
ing twigs. Because two-thirds of the history of life, and thus of the tree of life,
belongs to the prokaryotes, the complete tree can be said to look less like the
poplars of Lombardy and more like the live oaks of Louisiana (Dootlittle and
Bapteste 2007). I will return to this point about pattern pluralism, that is, the
recognition that different evolutionary models and representations of relation-
ships will be appropriate for different taxa or at different scales for different
purposes, in Chapters 5 and Chapter 9.
Continuous or discontinuous with respect to what?

In unpacking what is dubbed here as the continuity/discontinuity conceptual
complex in evolutionary explanations, the obvious question must be asked.
When speaking of continuity and discontinuity, what is our dimension of
assessment, as J. L. Austin might say?
First, as always, we want to remember that this complex is a product of
observers’ – and groups of observers’ – observations. The observers tend to
be evolutionary biologists, evolutionary anthropologists and interested oth-
ers, myself included. Having said that, then, we can state that a continuity is
a perception of sameness or similarity between or among organisms and/or
behaviors, and a discontinuity is a perception of difference between or among
organisms and/or behaviors. So, the first dimension of assessment in evolution-
ary explanations is that of dimension, namely size of perceived change from one
generation to the next. If the perception of change is great, evolutionary profes-
sionals tend to speak of a discontinuity. If the perception of change is small,
evolutionary professionals tend to speak of a continuity. Perceived changes
may be accompanied by the discovery of genetic changes, aka mutations, and
presumably there are such things as large mutations and small mutations. Now,
Continuous or discontinuous with respect to what? 117
the relationship between a large or small perceived (phenotypic) change and a

large or small (genetic) mutation is not at all straightforward.
A case in point is the oft-cited statistic that humans and our nearest ancestors,
chimpanzees, have (what most people would say is) a very small difference in
DNA (about 1 percent) and yet can be distinguished by (what most people
would say is) quite a bit of difference in how the two species look and behave.
Recently, this 1 percent difference, first proposed in 1975 when the similari-
ties between humans and chimps were underappreciated, has been reassessed.
Geneticists have been reviewing not only the base substitutions, which were
used to calculate the original percentage difference, but also the many stretches
of DNA that have been inserted or deleted in the genomes as well as ones that
might be randomly duplicated or lost. One team focused on gene copy numbers
and found a 6.4 percent difference between humans and chimps. Now, whether
or not 6.4 percent is considered large or small is a product of the observer’s
observations. As the geneticist Svante Pääbo said, “In the end, it’s a political
and social and cultural thing about how we see our differences.” He concluded,
“I don’t think there’s any way to calculate a number” (Cohen 2007). Indeed,
the determination of similarity or difference is made by the observer.
The broad large/small // different/same dimension intersects with another
broad dimension, that concerning whether whatever subject we are addressing
is one that pertains to all creatures great and small or whether it is one where
a distinction between the human realm and the non-human realm is pertin-
ent. Here perceived qualitative differences between the human and non-human
realms tend to be cast in terms of discontinuities, while perceived quantita-
tive differences tend to be cast in terms of continuities. Hauser, Chomsky and
Fitch mention a number of traits and abilities that humans may, or may not,
share with other animals, including chimpanzees, in terms of continuities and
discontinuities. Their discussion of the representation of number in humans
versus other animals, in particular chimpanzees, boils down to determining
whether the human ability to perceive quantities in terms of number is a quanti-
tative elaboration of an ability found in other animals or whether – and how – it
is, somehow, qualitatively different (2002: 1577).
The continuity/discontinuity conceptual complex is crosscut by two more
dimensions, that of civilized (language is not accountable by natural selection) /
naturalized (language is accountable by natural selection) and that of rationalist
(the emergence of language in the species is tied to cognitive advance) / ecological
(the emergence of language in the species is not tied to cognitive advance). One of
the most interesting theorists is Pinker, who opts for a naturalized account of our
subject matter in terms of natural selection but insists on a rationalist explanation
when he posits a “grammar mutant” (1994: 365), “jumping genes” and “trans-
posons” to account for how language got to be the way it is. The naturalized and
ecologically oriented story will be told in Chapter 5.
The self-reproductionist script

Modern evolutionary theory is no more a monolithic whole than is the discip-
line of linguistics. However, just as there once was a mainstream center for the
discipline of linguistics (Chomskyan linguistics, widely conceived), so there
has been a mainstream center of modern evolutionary thought whose begin-
nings can be identified as being consolidated by the British evolutionary biolo-
gist Sir Julian Huxley and his Evolution: The Modern Synthesis (1942 [1963]).
The (first) neo-Darwinian synthesis has been, since its creation, very gene-cen-
tric, appropriately so, perhaps, given the complexities of the genetic structure
of any given population whose task it is for researchers to understand. Early on
in the synthesis, population geneticists had thought that selection worked on
individual genes, each of which could be assigned a fixed adaptive value, such
that evolution would be no more than the addition and subtraction of genes
from the so-called gene pool. Henceforth, the (first) neo-Darwinian synthesis
will be labeled the self-reproductionist script, in reference to what is axiomatic
in this approach’s understanding of an organism.
A summary of the network of ideas that forms the self-reproductionist script
goes like this: in this framework, the understanding of the organism is organ-
ized around the concepts of competition and self-reproduction. That is, an
organism is understood to be an organism (or is understood to be an organism
worthy of consideration) to the extent that it is able to compete and that it sur-
vives to reproduce another organism of its kind. The reproductive fitness of the
organism is determined when the organism’s offspring have offspring.
This competition-and-reproduction framework has had the longest theoreti-
cal history, the widest disciplinary influence and, therefore, the most presence
in the contemporary evolutionary biological literature. Its theoretical basis
can be traced most directly to the late eighteenth century and the thought of
Thomas Malthus in his Essay on Population written in 1798. It is Malthus
whom Darwin quotes in his Origin, specifically in his third chapter entitled
“Struggle for Existence” where Darwin correlates the geometric increase of
organic beings with the struggle for existence, “either one individual with
another of the same species, or with the physical conditions of life.” If no such
competitive struggle existed, the number of organisms would “quickly become
so inordinately great that no country could support the product” (1968 [1859]:
117). In this framework, reproduction is: (i) measured as over-reproduction
with respect to (ii) some preexisting and fixed conditions of the “country” or
environment, e.g., a limited food supply.
If the organism’s capacity to reproduce and then, in fact, to over-reproduce
creates the competitive struggle for existence in the context of preexisting and
fixed environmental conditions, it makes sense to speak of those organisms
that do reproduce as those that are the most able or fit to reproduce another
The self-reproductionist script 119
organism of their own kind. That is, it makes sense to speak of them as the
best self-reproducers. Here the term best self-reproducers need not be under-
stood in any absolute sense but rather in the relative sense of better or more fit
self-reproducers than their competition, given the environmental conditions in
which they are competing. Relative fitness, then, can be appropriately spoken
of in terms of advantage along two dimensions: (i) the advantage one organism
has with respect to either another of its kind or another of a different kind (e.g.,
relative strength, speed, visual acuity); and (ii) the advantage one organism has
with respect to environmental conditions (heat, cold, food supply, etc.).
Classic examples of the first type of advantage might be: (i) the case of
sexual competition where an organism successfully competes with another
organism of its kind to mate with the opposite-sex organism of choice and is
thereby able to reproduce; or (ii) the case of an organism who outruns/outfights
a predator (presumably, but not necessarily, an organism of another kind) and
survives long enough to engage in case (i) competition. Classic examples of
the second type of advantage might involve an organism either which is able to
survive the particular environmental rigors of infancy or which resists and/or
recovers from disease in childhood or early adulthood (in any case, who lives
long enough to reproduce). In practice, of course, the two types of advantage
cannot be easily distinguished, and it makes sense to speak in general of adap-
tation, to say that the organism who survives and reproduces is adapted to the
conditions that allow just that survival and reproduction.
It also makes sense to speak of those conditions that allow just that survival
and reproduction as selecting the organisms that are best adapted to them. This
selecting process, called natural selection, is a causal force or pressure which,
under the form of competition, leads continuously to the survival of the (rela-
tively) best adapted. This selecting process defines the bare bones of evolution-
ary theory. Of course, there can be random mortality without the intervention
of selection, e.g., the case of a meteorite falling on the head of an otherwise
healthy and sexually viable organism. In this case, then, non-selective elimina-
tion is called drift. This means that branching species divergence is not inevi-
tably connected to natural selection, and, indeed, modern evolutionists accept
this uncoupling of selection from divergence.
Since the incorporation of Mendelism, self-reproductionist evolutionary
biologists have identified the physical basis underlying the evolutionary mech-
anism that was unavailable to Darwin and can express adaptive advantage in
terms of genes, i.e., as being due to one mutation over another. Mutation is an
intrinsic capacity of genes that provides the variability from which are selected
the so-called best variants of the trait or traits which allow and are definitive
of the organism’s adaptation. These genetic traits are, furthermore, heritable.
Evolution can now be seen as a two-stage process: (i) the production of genetic
variation; and (ii) the sorting of the variants by natural selection.
The development of twentieth-century genetics within the self-reproduc-

tionist framework has led to the identification of a species in terms of its gene
pool and the presence or absence of gene flow, i.e., reproductive isolation. The
proper evolutionary unit is now the population (the closed net of possible genetic
flux), and the evolutionary process is ( = ) the process of genetic change. All
these terms come together, with fitness more or less strongly defined in terms
of differential reproductive advantage. It is important to note that the self-re-
productionist framework described thus far is wholly continuist, and that this
continuity is ensured, in fact, by the genes. Although mutations are understood
as discrete quanta of change, the majority of them appear to be of small extent
such that the variability of a species at any given moment will be continuous
and evolutionary change will be gradual. Whatever else can or cannot be said
of it, the self-reproductionist framework is thoroughly gene-centric.
Despite the continuity supposedly ensured by a gene-centric view of evolu-
tion, evolutionary biologists in the self-reproductionist framework have trad-
itionally posited an abrupt discontinuity when they have turned to describing
the human reproductive condition. First, they consider humans unique among
organisms in showing an enormous range of individual variation in fertility,
instead of a single optimum value with low variance; and second, they con-
sider humans unique in having reduced the impact of natural selection on the
survival of individuals by artificial means, such as medical care and sanitation.
Huxley goes so far as to say that: “the human situation is so different from the
biological that it may prove best to abandon the attempt to apply concepts like
natural selection to modern human affairs” (Huxley 1942 [1963]: 261). Here
we see clearly, in the earliest formulation of the self-reproductionist frame-
work, the non-trivial effects of traditional assumptions that oppose the natural/
biological/animal world to the artificial/social/human world.
Over the past half-century, biologists have come to perceive the problems in
the nature/nurture opposition and have worked to overcome them. The results
of their efforts can be seen in such a standard college textbook as Douglas
Futuyma’s Evolutionary Biology (1998). Futuyma certainly gives a balanced
overview of the various nature/nurture positions that have been advanced in
recent years to account for human behavior and he takes care to negotiate a
path between the extremes (1998: 741–44). He notes that many conditions,
such as nearsightedness, which would have increased the risk of death for our
hunter-gatherer ancestors, can now be remedied by medical or technologi-
cal means, suggesting the human triumph of nurture over nature. However,
he pulls back from Huxley’s position that “it may prove best to abandon the
attempt to apply concepts like natural selection to modern human affairs” by
noting new pathogens that have emerged, e.g., HIV, that offer “considerable
scope for natural selection to act on genetic variation for resistance” (1998:
740). Interestingly enough, however, when it comes to human reproduction,
The self-reproductionist script 121
his position in an earlier version of the textbook was consonant with Huxley’s:
“Modern technology, education, medicine, and especially improved sanita-
tion have greatly reduced prereproductive mortality in industrialized nations,
reducing the opportunity for natural selection” (1986: 516). In the new edition,
he remains silent on the issue, which suggests that some rethinking of the dis-
continuity at the center of the self-reproductive framework concerning human
reproduction is occurring.
Evolutionary biologists and anthropologists are coming to realize that
humans have not ceased to evolve. For instance, evidence that skin color in
Europeans paled only recently alters the more standard hypothesis that modern
humans achieved their current state about 45,000 years ago and that they have
not changed much since. In addition, skin color in populations is not stable
and can go from dark to light to dark again, depending on the circumstances.
In evolutionary terms, what seems to be at issue is the current speed of human
evolution. One of the founders of the discipline of population genetics, R. A.
Fisher in Britain, argued that large populations with random mating – just what
globalization and air travel are helping to bring about – are the best fodder
for rapid evolution. Another founder of population genetics, Sewall Wright
in the United States, held the opposite view: that small populations with little
gene flow between them, such as was the case until the agricultural revolu-
tion 10,000 years ago, was the best situation for rapid evolution. To determine
whether human evolution is now – or ever was – rapid or slow, we would have
to know: rapid or slow with respect to what? There does not seem to be an
answer to this question.
In any case, the supposed human reproductive discontinuity opens the door
to – or, rather, is already a part of – the age-old assumption in Western thought
that humans are exceptional in the animal kingdom. From the very beginning
of the elaborations of evolutionary theory in Darwin’s Origin, theorists per-
ceived the difficulties of explaining the so-called human condition. Darwin
himself left the discussion of it to a later book. Furthermore, since disciplines
devoted to explaining various aspects of the human condition – e.g., history,
literary studies, anthropology, linguistics – were already in place or in the pro-
cess of establishing themselves in the nineteenth century, there was no great
institutional pressure to offer evolutionary accounts across all disciplines. In
fact, there was pressure to exclude evolutionary explanations, as was the case
in late nineteenth-century linguistic theory and practice. Another way of say-
ing this is that big ideas take time to be assimilated, to work themselves out in
specific disciplinary contexts and to find uptake across disciplinary boundary
lines. So it has been with the theory of evolution, which is still working itself
out and which is now being assimilated to a greater or lesser degree by (some)
linguists. In the early twenty-first century, the evolution of language/ing is def-
initely a topic whose time has come. However, and to repeat, the traditional
view of language that linguists have inherited was shaped in the deliberate
absence of evolutionary explanations, and the inspiration for the Chomskyan
model is overtly Cartesian rather than Darwinian. Thus, we cannot be surprised
that when it comes to offering evolutionary explanations of our subject matter,
linguists are faced with a multidimensional gap to fill.
For all the lack of discursive interaction until recently between evolutionary
theory and linguistic theory, these two fields have been running on parallel
lines for a very long time. Briefly, evolutionary biology and linguistics have
both been: (i) inspired in the twentieth century by a strategy of decontextual-
ization, where abstract structures upon which sentences in isolation can be
pitched – and/or the information flowing from genes is imagined to be a fixed
point around which whatever needs to be explained can be explained; and (ii)
informed by explanatory frameworks that have not included a developmental
perspective. Competence and the genotype thus line up to provide neat, linear,
feed-forward explanations of an end product. They provide the plan, the recipe,
the blueprint, the program – or, in the case of Saussurean langue, the musical
score – which, once deployed, simply (although nothing is simple) happens.
Whatever occurs during the deployment or after full deployment is not relevant
to and does not affect the plan, recipe, blueprint, program, musical score, etc.
Given this, Pinker and Bloom (1990), Pinker (1994, 2003) and Pinker and
Jackendoff (2005) could hardly do anything other than to tell their story at
the intersection of the mutually supporting frameworks for self-reproductionist
evolutionary biology and Cartesian/Chomskyan grammar, an intersection that
is a completely blank space. In sum, they: (i) suppose that the genotype deter-
mines the phenotype and that advantage is conferred due to one mutation over
another; (ii) recognize that evolution is not purposive, not teleologic, but never-
theless argue teleonomically; that is, they argue that the so-called language
instinct or faculty of language has at least the appearance of design that is the
hallmark of adaptive complexity, and although they deny the existence of a
designer (e.g., God), they believe that it was designed for some purpose; (iii)
determine that what language was designed for was the trading of hard-won
knowledge with kin and friends whose payoff was a greater enhancement of
fitness; (iv) speak of language and the brain on the analogy of the computer,
which was designed for information processing; (v) assume, on the analogy
of the computer, that language is a code encoded on the genetic code whose
phenotypic appearance, as stated in (i), is determined by the genotype, which
will include genes for language, with the more genes that are necessary for
language, the lower the probability that they accumulated by chance and the
higher the probability that they are the results of natural selection (see also
Pinker 2003: 33–37).
Pinker’s most recent article on the subject “The Cognitive Niche:
Coevolution of Intelligence, Sociality, and Language” is an admirable update
Done with Descartes 123
on his earlier work. He notes, for instance, that “many biologists argue that
a niche is something that is constructed, rather than simply entered, by an
organism” (2010: 3). The terminology suggests that he is now open to some
elements of constructivism. Yet he remains committed to the rationalist idea
that language is a “means of transmitting information to another brain” (2010:
3) whose power is “to convey news” (2010: 2) but says nothing of the bod-
ies that may be involved in these transmissions or conveyances, for all his
emphasis on cooperation. Pinker acknowledges that “language may foster
cooperation, but it also depends on it” (2010: 4), but seems to assume both,
along with human reasoning abilities, for he leaves absent the identification of
any living beings encountering any other living beings of a similar kind and
engaging in any activities that would count as prelinguistic.
This absence is a blank space, and it looks something like this: here we are
now in language. Since we are here now, we must have got here somehow.
The acceptable scientific way of explaining the existence of a physical struc-
ture, trait, character or behavior in an organism is to give (some version) of
a Darwinian account of it. It is the case that natural selection works on what
is already there and that adaptations begin as by-products or modifications
of structures, traits, characters or behaviors. What natural selection acted on
(besides the genes) and modified in the case of human language is unknown,
unknowable and possibly even nothing, if we accept that language is a wholly
new cognitive and/or behavioral domain. Thus, we cannot tell how we got
here, but the fact remains that we did get here, because here we are now in
language.
A third historiographic review: done with Descartes

Attempts to account evolutionarily for any Chomsky-oriented approach to our
subject matter always founder on the fact that they are inspired by Descartes
and not Darwin. That is, they operate in a rationalist-realist epistemological
framework in which the speaker is conceived: (i) to be motivated to speak
primarily in order either to transfer the picture of the world, or knowledge
of it, from his/her head to the head of his/her listener or to entertain interior
monologues, and not to affect the environment in some way, e.g., to affect
the behavior of his/her listener; and (ii) to hold in his/her head quasi-pictures
(representations) of the world rather than adaptations to the environment pro-
duced by interactions with it. What is of interest to us now, as we delve fur-
ther into the linguistics inspired by Descartes rather than informed by Darwin,
is the significant epistemological linkage between vision and (the Cartesian/
Chomskyan version of) language. Specifically, the conception of the represen-
tational function of language is supported by the conception of the supposedly
representational function of vision. The vision–language linkage takes us back
to the Enlightenment where we trace today’s sources of evolutionary scenarios

for language, whose initial stages are imagined, by contemporary theorists, to
play out not in noisy interaction but in watchful silence on the part of the earli-
est members of our species.
The master metaphor of the eighteenth century was of the world as a grand
timepiece, or clock, and that everything in the universe, including human per-
ception, could be explained in terms of a machine that could be depended
upon to work regularly. It followed that if nature was made up of separate,
given parts, man’s perception of it must also be composed of separate and
given parts. Even a particular sensory perception was deemed to be a com-
posite of elements that the brain of man-the-machine mechanically registered.
Now, these individual sensory elements were theorized not to arrive at the
brain as a jumbled mass but rather to be received one by one and in an ordered
sequence dictated by nature to form a mental image or picture of a particular
perception.
In the eighteenth century the term tableau (“picture”) was used to refer both
to nature as it exists and to the mental image of it. The tableau thus came to
serve as the metaphor to unite the mechanistic belief in an organized and uni-
fied reality with the mechanistic theory of perception, in that the tableau rep-
resented both the reality “out there” as well as the mental tabula rasa (“blank
slate”) receiving the data about that reality. This use of the term tableau per-
vades all work on language in eighteenth-century France. The metaphor of the
tableau indeed captures the essence of mechanistic explanations. It implies
first that understanding is primarily a visual faculty, and second that the spa-
tial, geometrical way of thinking prevails over the perception of time, for with
the tableau one is able to convey that everything in nature has been given and
unfolds completely within its space. In a mechanistic view of the world, time
may create the appearance of change, but it has no power to affect the underly-
ing, eternal organization of the universe.
The universe so organized is therefore spatially representable in terms of
an abstract and continuous geometric design that knows no rupture from the
lowest matter to the most perfect beings. Since the time of the Neoplatonist
Plotinus, this design was known as the Great Chain of Being. Theorists in the
eighteenth century, operating in a spatial domain, imagined that all evolution-
ary possibilities could exist on the earth at the same time. Thus, an orangutan
(Malay orang “man” + hutan “forest”) could be imagined to be on a rung of
the ladder just below, say, the American Indians. (Europeans were generally
believed to perch on the top rung.) By the end of the nineteenth century, the
totality of species, including so-called missing links in the primate lineage,
could only be conceived as having existed in historical time. The image of
the ladder disappeared in scientific discourse, and relationships among species
transformed into now-familiar trees.
With the spatial and geometric view of the organization of nature in play
in the eighteenth century, intelligence was seen to be as vast as reality, for it
was (and perhaps still is) incontestable that what is geometrically organized is
entirely accessible to human intelligence. At the bottom of these speculations
lay the conviction that the function of intelligence was to embrace nature in its
entirety, thereby identifying the knowing with the totality of experience. (We
are speaking about an age that referred to itself as the Enlightenment, after all.)
The principle of mechanical regularity, seconded by those of geometry, assured
the grammairiens philosophes that the past and future were calculable in func-
tion of the present and visible all at once to the (enlightened) individual who
could apply the calculus. The enlightened mind could then grasp and under-
stand the totality of reality just as sight embraces the horizon.
The grammairiens philosophes had confidence in the mechanistic operation
of the world, confidence in the mind’s ability to seize that operation, and above
all, confidence that a langue bien faite (“well-made language”) would sup-
ply the calculus, i.e., be the tool that would implement man’s enlightenment.
For these reasons, they placed the study of language at the epistemological
crossroads of their thought. By mediating the mental tableau and the tableau
of nature, everyday language was believed to take a first step toward grasping
the operation of the world. This mediation allowed man a certain measure of
control over the succession of tableaux, so that instead of being swept along
by passing desires, feelings and sensations, man could dominate them, halt
the interminable flow of experiences and examine them. But the philosophes
wanted more than mere control; they were striving for enlightenment, and that
required an absolute mastery over reality that, in their opinion, would come
only with a well-made language.
The study of language, thus, became a quest for the ideal language, the one
that would give man the most perfect understanding of reality, the one that
would render most accurately the original tableau. Since the mental tableau
must necessarily be composed of individual parts which (supposedly) corre-
sponded to those in nature, the philosophes’ task was to decompose the men-
tal tableau into its constituent elements, identify each with a name and then
reassemble the elements to reproduce, as faithfully as possible, the original
tableau. The emphasis rested on the naming process, for the more precisely the
word corresponded to the thing – be it an object, idea or sensation – and the
more a word’s name made transparent the composition of the thing, the better
the reproduction of the original picture. By the turn of the nineteenth century,
students of language had come to realize that language was not transparent and
that if we do indeed see through language into either nature or the mind, we
“see through a glass darkly.” However, this eighteenth-century emphasis on
the naming process, whereby names can transparently reflect the composition
of their referents, helped to bring forth the modern discipline of chemistry.
Antoine-Laurent Lavoisier, for instance, was a great admirer of Condillac’s

writings on language.3
The rise of the interest in philology at the end of the eighteenth century with
its goal of reconstructing protolanguages required serious linguists to abandon
the attempt to construct the perfect language that might exist in the future. If
anything, there was a belief held by certain philologists, e.g., Grimm, that the
Germanic protolanguage, die Ursprache, was somehow better and that con-
temporary German, English, etc. were decayed remnants of a formerly glorious
linguistic past. While the project of constructing an ideal language gave way to
the adventure of reconstructing a protolanguage, what was not reexamined in
the main line of linguistic thinking from the eighteenth century onward was the
connection between some construct language and some construct thought to
which cling significant chunks of Enlightenment epistemology. These chunks
have recently become most visible in the litmus of evolutionary scripts that do
not question that the evolutionary account of language/ing will have to have a
connection to the evolution of human intelligence. What is called here ration-
alized evolutionary accounts of language thus assume that the story is about
the evolution of intellectual capacities and mental organization – i.e., cognitive
specialization – one way or another.
The most striking conceptual carryover from Enlightenment epistemology
to contemporary linguistic theory, and then to evolutionary explanations, is its
spatial aesthetic. Nineteenth-century philological method, working with the
formal residues of languaging living beings long dead, did nothing to efface
that aesthetic. Chomskyan linguistics is fundamentally geometric, which
means that it is first and foremost even obviously and therefore unremark-
ably visual, which also means that it is silent. In Chomskyan linguistics, the
finished, monologic utterance, divorced from its verbal and actual context and
standing open not to any possible sort of active response but to passive under-
standing on the part of a philologist, is intended only for visual contemplation
in its representation by a syntactic tree. Since it is intended neither as any kind
of stimulus nor as any kind of response, it has no auditory and therefore no
temporal life. Current linguists working in this framework do not imagine or
present a social framework in which the faculty of language, broad or narrow,
has evolved or currently operates, since the center court of explanation is the
mind/brain and the central question of how it does and does not differ from
other species’ minds/brains, notably those of non-human primates.
3
There is a far more textured story to be told about the emergence of modern chemistry than
the one just told in terms of hands-off theorizing about the naming process. The more textured
story involves acknowledgement of a great deal of hands-on activity in the form of exuber-
ant experimentation pursued by apothecaries, metallurgical officials, consultants, entrepreneurs,
learned practitioners, teachers and professors, as well as the social dimensions of scientific pur-
suit through academies and scholarly societies.
The point here is that evolutionary stories told in terms of either the con-
vergence or divergence of cognitive capacities and language have tended to
carry forward the baggage of a representational psychology which emphasizes
vision and space over hearing and temporality, which, in turn, tends to generate
evolutionary scripts for language whose first stages either are ignored or are
imagined to unfold in watchful silence, i.e., those learning language learn it by
watching, not by speaking. The most strikingly quiet script produced within a
generalized Cartesian/Chomskyan – in any case, rationalist and comparative-
intelligence-based – model of language and culture is that of the monograph-
length article by the sociobiologists/evolutionary psychologists John Tooby
and Leda Cosmides entitled “The Psychological Foundations of Culture” pub-
lished in The Adapted Mind: Evolutionary Psychology and the Generation of
Culture (Barkow et al. 1992). Pinker and Bloom (1990), reprinted in the same
volume, is in synch with Tooby and Cosmides’ framework.
Tooby and Cosmides propose a model to account for humans, their minds
and their collective interactions that recognizes, among other things, that “the
human mind consists of a set of evolved information-processing mechanisms
instantiated in the human nervous system” (1992: 24). Part of their task,
then, is to account for how such a set of information-processing mechanisms
evolved. They focus in on a kind of culture characterized by representations or
regulatory elements that reappear in chains from individual to individual, i.e.,
culture in the classic sense. Instead of calling it “transmitted culture,” they pre-
fer to call it reconstructed culture, adopted culture or epidemiological culture
because:
The use of the word “transmission” implies that the primary causal process is located in
the individuals from whom the representations are derived. In contrast, an evolutionary
psychological perspective emphasizes the primacy of the psychological mechanisms in
the learner that, given observations of the social world, inferentially reconstruct some of
the representations existing in the minds of the observed. Other people are usually just
going about their business as they are observed, and are not necessarily intentionally
“transmitting” anything. (1992: 118)
So, if no one is intentionally transmitting anything, how do those representa-
tions get from one head to another? The answer is that “an observer (who …
we will call the ‘child’) witnesses some finite sample of behavior by others
(e.g., public representations, such as utterances or other communicative acts;
people going about their affairs, people responding to the child’s behavior).”
From these observations the child reconstructs the set of representations or
regulatory elements that s/he needs, making the “learning of culture” equiva-
lent to “deducing the hidden representations and regulatory elements embed-
ded in others’ minds that are responsible for generating their behavior” (1992:
118, emphasis mine). Now, it is clear that a child can be said to witness peo-
ple going about their affairs, but how does a child witness an utterance or a
response to his/her behavior? In the latter two cases, the child is surely not a
silent, watchful witness but an active, perhaps even noisy, participant, whose
behavior shapes and is shaped by whatever activity, including verbal activity,
is at hand. By taking the child out of the interactive loop, Tooby and Cosmides
are not going to be able to explain how the representations in the heads of the
people whose behavior the child is witnessing got there in the first place when
they were children, and Tooby and Cosmides do not undertake such an expla-
nation. They propose an internal mechanism that is all feed-forward and with
no feedback.
The situation is not dissimilar to Pinker and Jackendoff’s uncritical accept-
ance of Saussurean code theorism when they cite studies that have shown that
children, upon hearing words used by a speaker, “can conclude that other speak-
ers in the community, and the child himself or herself, may use the word with
the same meaning and expect to be understood … This is one of the assump-
tions that allows babies to use words upon exposure to them, as opposed to
having to have their vocal output shaped or reinforced by parental feedback”
(2005: 214). First a quibble. Are children, namely young children and particu-
larly babies, necessarily able to conclude anything about other speakers’ abili-
ties to understand them or anyone else? With experience and exposure, more
mature languagers do often come to that conclusion, because their behaviors
are so well coordinated with other languagers’ behaviors around those words.
And with even more experience and exposure, some languagers come to fur-
ther conclusions, namely that other languagers often do not use the same words
with the same meanings. However, more to the point in this passage, is the
continuing denial of shaping and reinforcing by parental feedback. One can
only wonder why. Is there truly too much feed-forward theoretical machinery
in mainstream linguistics to preserve at this point? When it comes down to the
bottom line, it is the feedback loop, in evolutionary explanations sympathetic
to the dynamics of languaging as an orienting behavior, that will explain how
there comes to be a lineage of living beings born into this world who are capa-
ble of becoming languaging living beings.
The feed-forward conceptual impasse is repeated in Newmeyer’s revised ver-
sion of the presidential address he gave at the annual meeting of the Linguistic
Society of America in 2003. He notes that when linguists started turning to
the questions of the origin of language and evolution in the early 1990s, “a
near consensus began to develop on a central issue. In a nutshell, the idea
was that the roots of grammar lay in hominid conceptual representations and
that the shaping of grammar for communicative purposes was a later develop-
ment” (2003: 698). As is customary for a theorist committed to a version of the
Chomskyan paradigm broadly defined, Newmeyer contends, in effect, that the
roots of grammar take shape not in communication (verbal interactions) but in
prehuman conceptual structures. Of course, there are linguists these days who
argue the opposite – namely, that language evolved primarily as a vehicle of

communication – and they charge the claim that language evolved primarily
as an instrument of thought with being intuitive and a priori. To these critics
Newmeyer says: “It is not clear to me why the claim that language emerged
from thought is any more ‘intuitive and a priori’ than the claim that it emerged
from communication” (2003: 699, n.13). Let us make it clear: in order for lan-
guage (in Newmeyer’s sense) to have developed from conceptual representa-
tions (in Newmeyer’s sense), prehumans had to operate essentially in silence,
and they could not have been affecting one another bodily or otherwise. Since
Newmeyer’s stated goal in his article is to argue in favor of the classical
Saussurean position with respect to the relationship between knowledge of lan-
guage and use of language, it is not unfair to think that Newmeyer subscribes
to some version of the code-theoretic premise that the purpose of language is
the transfer of conceptual representations from one head to another. So, now,
where did hominid conceptual representations come from? Newmeyer does
not speculate on that, but the available explanation is that the development
of these conceptual representations depended primarily upon a visual strategy
(e.g., witnessing?) such as that spun out by Tooby and Cosmides.
We are at the rock-bottom of realist representationalist rationalist explana-
tions: that of vision. We cannot proceed without mentioning the work of the
influential British psychologist David Marr, whose Vision: A Computational
Investigation into the Human Representation and Processing of Visual
Information (1982) is a landmark study that some say has given rise to the
currently vibrant field of computer vision. Here we must take a quick pass
through the field of cognitive science that began in the 1950s, during which
time: (i) the human mind was assumed to be understood best by comparison
with a computer; (ii) the mind was proposed to operate with mental represen-
tations; and (iii) mental representations were posited to be results of specific
(algorithmic) kinds of processing. For Marr, vision was an information-
processing task, and his three-tiered description of this processing quickly
became a model for how all cognition supposedly operated, namely in terms of
a computational level, an algorithmic level and an implementation level. Marr
was working some twenty years into the so-called cognitive revolution, and
he had come to see that progress on the understanding of vision had stalled.
The problem, as he saw it, was “that people were only describing vision and
its neurophysiological and anatomical underpinnings, but not explaining how
visual information, visual physiology, and perception are related” (Purves and
Lotto 2003: 4). He proposed to bring it all together by offering his model
that information available in the retinal image was subjected to algorithmic
processing at three levels.
The crucial point to make here is that Marr, among many other vision research-
ers, assumed that the purpose of vision was to offer an accurate reconstruction
in perception of the physical world based on the information that generates

the retinal image. Such a view certainly makes sense; in fact, it makes com-
monsense. What other purpose could vision serve than to see the way the world
“really is”? Vision researchers are not the only ones to hold such a common-
sensical position. In a purportedly evolution-inspired work, the anthropologist
Donald Symons refers to our “astonishingly accurate” perceptual mechanisms
and the “accurate representations” of the world that they provide in an article
entitled “If We’re All Darwinians, What’s the Fuss About?” (1987: 131 – not
incidentally, Symons works congenially with Tooby and Cosmides). It is hard,
at first, to take issue with the accuracy of vision: when my eyes are open, I get
around remarkably well. I really do see what I see. “Seeing is believing” is a
cliché in the way that “hearing is believing” is not – although the negative “I
can’t believe my ears” does exist as a response to something that one has, actu-
ally, heard correctly. But just as the commonsensical nature of code theorism
obscures a way of approaching our subject matter and prevents a developmen-
tal linguistics, so the commonsensical nature of visual perception obscures an
account of human visual perception inspired by biology rather than by com-
puter science. The computer science model, in short, derives from Descartes
rather than Darwin.
Marr’s descriptions of visual processing, with their mathematical rigor and
formal systematization, have much in common, in style and assumptions,
with Chomskyan linguistics. They also act as a kind theoretical backstop to
the evolutionary scenarios that support the development of cognition as the
precursor to “language.” The (Chomskyan-style) language-processing/pro-
duction machinery was proposed to exist in order to process/produce accurate
re-representations (in language) of the accurate cognitive representations (in
our minds) that are either being entertained in the free expression of thought
(whether or not they get articulated) or are being transferred from one head to
another. However, if our vision is not a result of rule-based schemes that oper-
ate on information “out there” in the world, and if our minds are therefore not
operating with “accurate representations” of that world out there, then the bot-
tom falls out of the scenario where cognitive development precedes languaging
interactions. This is so because if cognitive development proceeded by means
other than as an increasingly good (better and better) conceptual/perceptual
adjustment to reality, then the rationale for a certain conception of language
(the increasingly good transfer of representations from one head to another)
is lost. This does not mean that Marr’s three-tiered account of processing is
wrong. It only means that we are not licensed to feel with respect to our percep-
tions that they are passively taking in an outside, independent world – what in
rationalist-realist terms is typically called reality.
Now, why not go along with the computer science/cognitive science approach
to the mind/brain and understand cognition, and in particular vision, in terms
of computers? The neuroscientists Purves and Lotto cast the question thusly:
what is so wrong with:
the seemingly sensible idea that the purpose of vision is to perceive the world as it is
and that this obviously beneficial goal is achieved by neuronal hardware that detects the
elemental features of the retinal image and, from these, reconstructs a representation of
the external world according to a set of more or less logical rules instantiated in visual
processing circuitry? (2003: 5)
The answer is that the significance for subsequent action of any retinal stimu-
lus is unknowable directly. In other words, the significance of a stimulus for an
organism has nothing to do with the properties of the stimulus itself. That is,
a stimulus can trigger something in an organism, but it is only the organism’s
internal organization, which is partially determined by its prior interactions
with that stimulus, that specifies what that stimulus “means” to the organ-
ism. What stimuli are relevant, i.e., can trigger something in an organism, are
determined: (i) by the lineage to which this organism belongs, namely by the
ways the organism’s ancestors’ interiors have been shaped by prior interactions
with those stimuli; and (ii) by the organism’s own history of interactions with
those stimuli. It may be difficult to accept, Purves and Lotto acknowledge, “the
admittedly strange idea … that what we see is a probabilistic manifestation
of the past rather than a logical analysis of the present and that the discrepan-
cies between reality and perception (visual illusions) are the signature of this
process” (2003: 11). Nevertheless, neurobiological evidence indicates that the
visual system is not organized to generate veridical (= truth-giving, real or
actual) representations of the physical world “but rather is a statistical reflec-
tion of visual history” (2003: 227). What we see is as much a constructed result
of experience – which implies the circular causality by which bodily actions
under brain control modify the brain through sensation, perception and learn-
ing – as every other aspect of our being, our knowing and our doing. When we
open our eyes, we do not “bring an outside world in.” Rather, we bring forth
our visual domain through our visual experiences as living beings belonging to
a certain lineage, leading certain lives.
And the lives that humans lead are generally very noisy – lipsmacks, tongue-
smacks, teeth chattering (maybe not) and grunts included. So let’s acknowl-
edge it.
5 Evolutionary scenarios II: the emerging
story and the self-realizationist script
The story to be told in this chapter is emerging in two senses: it is gaining

acceptance as an explanatory framework, and it suggests that the human lan-
guaging network is a phenomenon that emerged over several million years
to become recognizably modern in the last several hundred thousand years.
Although this seems like a fairly straightforward statement to make, nothing
about this topic is straightforward. The editors of Language Evolution, Morten
Christiansen and Simon Kirby, are well aware of that fact when they ask the
deliberately provocative question: “Language Evolution – the hardest question
in science?” Of course, they realize that it is unanswerable, since every discip-
line has its difficult questions, and they concede that those studying conscious-
ness have already “laid claim to the phrase ‘the hard problem.’” Nevertheless,
they consider the problem of language evolution to be a unique scientific chal-
lenge, given that language, being difficult enough to define to begin with, is
also “by far the most complex behaviour we know of ” (2003: 15). As such, it
requires examination from an extremely wide variety of disciplines.
Certainly this is true. However, the difficulty of the problem does not derive
solely from the complexity of the behavior or the variety of disciplines involved
in tackling it – though these aspects of the undertaking surely do pose distinct
challenges. Rather, the difficulty of the problem lies in the fact that the ques-
tion of the evolution of language – or, rather, the question of the instantiation,
development and maintenance of the human languaging network – has some
of the most historically embedded conceptual baggage to unpack before one
can move forward.
In the previous chapter, arguments surrounding the relationship between
the development of certain cognitive abilities and a certain conception of lan-
guage, framed in terms of various dimensions of the continuity/discontinuity
dichotomy, were discussed. What is characteristic of any version of the stand-
ard story is that it is built up and out from an inherent discontinuity in its expo-
sition. That is, the fulcrum upon which the whole story is balanced is either
an exception to or an exemption from an evolutionary process, a suspension
of the usual dynamics of natural selection, a special pleading for the unique-
ness of some facet of human existence (e.g., language) and/or some facet of
132
Emerging story and the self-realizationist script 133
language (e.g., recursion, phrase structure rules with X-bar syntax, long-range
dependency relations) or even finally (or, rather, to begin with) an inexplicable
mystery. This is the civilized script, because it requires that, somewhere along
the way, there not only is but also must be a difference in kind between (some
of) the ways humans go about living their lives and (some of) the ways non-
human organisms live theirs. In the standard story, a lot of effort is devoted to
filling or leaping across that gap, but the realization does not seem to come
that it is the very conception of language and its linkage to some prior but
yet-to-be-determined cognitive specialization – aka thought – that caused the
gap in the first place. Although no one telling an evolutionary story these days
invokes reason as the unbridgeable dividing line between humans and brutes,
the power of recursion in the Hauser, Chomsky and Fitch story serves a func-
tion similar to reason in its account of the human/non-human divide, i.e., that
which we cannot attribute to a non-human organism without losing that which
we consider to be uniquely human.
In Linguistics and Evolution, by way of contrast, when the evolutionary
script for language/ing is: (i) unlinked from thought (however we want to
understand that term), from code theorism, and from silence; and (ii) restored
to a setting in which brains are embodied and bodies are interacting from the
beginning, an understanding arises that tends, first, to emphasize sameness and
similarity between even complex human behaviors and those of non-human
organisms and then to produce a continuist plotline. What now in retrospect
reads like a preemptive refutation of Chomsky-inspired evolutionary consider-
ations of the 1990s is Jane Hill’s “Possible Continuity Theories of Language”
in which she observes that “people create much of their environment, using
language as a symbolic tool for the creation, and then proceed to live in it. For
other animals a similar point could be made. What an animal’s world consists
of, what is relevant for its survival, is in large part determined by the properties
of the animal itself” (1974: 135). Hill, among others, is pursuing descriptions
in what can be called a naturalized dimension of assessment in that the world
humans live in is not different in kind from the worlds that any other animals
live in, since they are all particular to themselves. Hill calls her approach eco-
logical. This ecological or naturalized approach requires a different languaging
about our subject matter. Where, for instance, Hauser, Chomsky and Fitch con-
tend that if a human and a non-human primate share a cognitive ability, then
such an ability cannot, by definition, have been developed for language, an
ecologically oriented description requires more nuance. That is, if a human and
a non-human primate share a cognitive ability, then we can state that those abil-
ities emerged from a common origin. However, the human ability for numbers,
for example, may well have been developed through language/ing.
An ecologically oriented description does not require a break of any kind
or a line to be drawn between or among abilities, but rather sees threads that
134 Emerging story and the self-realizationist script
may be looped one way through one group of organisms, i.e., a species, and a
different way through another. This means that culture is as natural an activ-
ity as any other for human beings. In many cultures, for instance, there is an
open-ended counting routine (“1, 2, 3, 4 …”) that children typically learn first
and then later individuate, and this is a routine that chimpanzees and some
human cultures, e.g., the Amazonian Pirahã, lack. However, for a develop-
mental linguistics, the presence of an open-ended counting routine is not “an
interesting quirk of the child’s learning environment” as Hauser, Chomsky
and Fitch phrase it (2002: 1577); it is rather constitutive of it, in that it scaf-
folds the child’s coming into a particular and important aspect of languaging
and culture. And nothing in the understanding of languaging as an orienting
behavior depends upon positing any particular prior cognitive development in
the human species. Rather, the social and cognitive dimensions of languaging
behaviors are deemed to have arisen from features of primate behavior and
cognition, generally speaking.
A list of points for a continuist script for the emergence of the human lan-
guaging network can now be generated to contrast with the list presented in
Chapter 4:
(i) languaging shares much in common with the social behaviors of other
species; however, languaging distinguishes itself by the extra loops of
coordinations that occur in languaging interactions and that have cogni-
tive consequences, such that over time these coordinations of coordina-
tions have increased in complexity, and the cognitive domains in which
humans can be oriented have increased in diversity;
(ii) languaging does not exist as a reference to another reality existing out-
side of languaging;
(iii) languaging is fully embodied in the individual and completely embed-
ded in sociocultural relationships;
(iv) languaging is constitutive of the kind of cognition often called cultural
thinking;
(v) languaging is constructed in ontogeny;
(vi) languaging is an activity that is constitutive of the lineage we call
human;
(vii) as has been stated before, although languaging is not co-extensive with
the whole of human cognition, it is difficult to turn many corners in the
brain without bumping into some kind of process or architecture that
supports and/or is affected by languaging; and
(viii) languaging as an orienting behavior derives from behavior of non-
linguistic origin that has developed into a system of cooperative consen-
sual interactions often called natural language or human language.
The second neo-Darwinian synthesis 135
What is interesting to note, furthermore, is that a significant chunk of evolu-

tionary theory as a whole rests on how we theorize language/ing.
The second neo-Darwinian synthesis

As work in the (first) neo-Darwinian synthesis progressed throughout the sec-
ond half of the twentieth century, it became increasingly clear that whatever is
going on genetically, it is happening at the level of large gene networks, not of
individual genes. Thus, interest in the importance of developmental factors has
increased. Accordingly, in the last ten years, and in response to the originally
reductionist, gene-centered impetus of the first synthesis, a move has begun
towards a second synthesis. It was presaged decades ago by Waddington and
is currently evidenced by such a recent title as Evolution in Four Dimensions:
Genetic, Epigenetic, Behavioral, and Symbolic Variation in the History of Life
(2005) by Eva Jablonka and Marion Lamb. In a similar vein, and as much an
extension of Darwin’s evolutionary theory as these just-mentioned titles, is the
work of Humberto Maturana, in particular the monograph he co-authored with
Jorge Mpodozis entitled The Origin of Species by Means of Natural Drift or
Lineage Diversification Through the Conservation and Change of Ontogenic
Phenotypes (1992).
Not surprisingly, Maturana and Mpodozis build their evolutionary script out
from the concept of autopoiesis, and they take as a biological phenomenon
any phenomenon that involves the autopoiesis of at least one living being.
Their goal is to restore the role played by the behavior of the living being
as a guide to the ongoing existence and to the becoming of all living beings.
Maturana and Mpodozis’s work, along with the newer approaches pointing
toward a second synthesis, do not require the abandonment of neo-Darwinism
but rather an expansion of it, along with a reorientation. What they all share is
a renewed attention to the encounter of the phenotype with the environment as
an important site of both phylogenic and ontogenic explanation. In recogniz-
ing the importance of the encounter of the phenotype with the environment,
Maturana and his various co-authors over the years have been the most theoret-
ically advanced, both in terms of being several decades ahead of the curve and
in terms of theoretical daring. That is, while Maturana and Varela reached deep
by redefining the living in terms of autopoiesis back in the 1970s, Jablonka and
Lamb, for instance, more recently chose to sidestep the issue of “the defini-
tion of life,” declaring it to be “a really messy subject” (2005: 40). Because of
these factors, then, this (second) neo-Darwinian synthesis will henceforth be
referred to as the self-realizationist script, again to highlight what is axiomatic
about the understanding of an organism in the phenomenological framework
offered by Maturana and his co-authors.
The self-realizationist script

In the epistemological framework of the self-realizationist script, the under-
standing of the living being is organized around the concepts of self-realization
and autopoiesis. In this framework, a living being is: (i) living insofar as the
network of processes that are constitutive of it continues to realize itself in the
ongoing operation of the living being; otherwise it dies; and (ii) a being insofar
as it is a unity existing in space that may be distinguished by an observer. That
is, a living being is a living being insofar as it is a unity that continues to realize
itself in its autopoietic organization, i.e., there is uninterrupted operation in the
continuous realization of its self-realizing organization.
The self-realizationist framework has a relatively short history, relatively
little presence in the literature of evolutionary biology, but relatively wide and
growing influence across a variety of disciplines. This framework is associ-
ated almost exclusively with the work of Maturana; Maturana and Varela; and
Maturana and Mpodozis. The most important work on the subject is Origen de
las Especies por medio de la Deriva natural o La diversificatión de los linajes
através de la conservación y cambio de los fenotipos ontogénicos (1992) by
Maturana and Mpodozis, mentioned above in its English translation. The
unpublished edition referred to in this chapter was translated into English by
Cristina Magro of the Universidade Federal de Minas Gerais (Belo Horizonte,
Brazil) and by the present author. Admittedly, Jablonka and Lamb would count
as a more mainstream approach to the second neo-Darwinian synthesis, that
is, if we take the term mainstream to mean work published in English at a
major university press. In any case, the self-realizationist script of Maturana
and Varela is generally compatible with the aforementioned works and with
developmental systems theory, and since their account of languaging is central
to the attempt to incorporate languaging as an orienting behavior into a twenty-
first-century linguistics, their evolutionary account is preferred here.
The conceptual basis of the evolutionary framework of Maturana and
Mpodozis – just like the conceptual basis of the work of Maturana and Varela –
derives from the observation that everything is said by an observer whose
particular way of looking determines what that observer says. An observer
observing a living being in its environment, for instance, can consider the
operation of the living being either: (i) from the point of view of the domain of
operation of the living being, in which case the observer may wish to describe
the living being’s internal functioning and structural changes (physiology); or
(ii) from the point of view of the domain of the living being’s interactions with
its environment, in which case the observer may wish to describe the history
of the living being’s interactions with its environment (behavior). To repeat the
point made in Chapter 3: (i) and (ii) are two separate domains, and as long as
they are kept separate in the observer’s descriptions, conceptual confusions do
The self-realizationist script 137
not arise. Recall that the word autopoiesis was coined by Maturana specifically
because it had no history and could unequivocally mean “what takes place in
the dynamics of the autonomy proper to the living systems” (Maturana and
Varela 1980 [1972]: xvii).
Certainly there is a long tradition of thinking in which the two domains of
description are confused and conflated, namely the tradition of classical empir-
icism and realist representationalism to which self-reproductionist evolution-
ary biology belongs. Again, the introduction of autopoiesis into the descriptive
domain of the observer facilitates the maintenance of the distinction between
the two observational domains. Now there is a term to connote what we want
to call the internal dynamics proper to the autonomy of the living being in
terms other than understanding these internal dynamics as “taking the outside
(the environment in which those internal dynamics can be seen to be realizing
themselves) in.” Other attempts to break from the classical tradition and to
distinguish the two domains – that of the internal organization of the organism
under observation and that of the organism’s environment – have been made,
notably by William James. Nevertheless, the particular extensions of autopoi-
esis and the elaboration of the self-realizationist framework seem confined to
that of Maturana and his co-authors.
In this self-realizationist framework, then, the operation of a living being is
understood to be a structurally determined (and not genetically determined)
system which exists insofar as the living being maintains, i.e., conserves,
its organization and congruence with its circumstances. The conservation of
organization, that is, adaptation, is a systemic phenomenon and not strictly a
biological one. As such, it depends on the dynamics of the living being within
the environment in which it realizes itself. That is, adaptation is a relation of
dynamic operational concordance between the living being and the environ-
ment in which it realizes itself. Defined as such, adaptation is an invariant,
that is, adaptation is a condition of existence, is constitutive of it, and does not
require another explanation. In other words, adaptation is not a variable, and
therefore it is not meaningful to talk about organisms that are better adapted
than others, even in relative terms. Nor is it meaningful to talk about adaptive
strategies or processes, even in relative terms. To the extent that the living
being exists (or existed), it is (or was) by definition adapted to its medium.
Clearly, then, the term adaptation has a different meaning in the self-realiza-
tionist framework than it does in the self-reproductive framework. In the self-
realizationist framework, only self-realization is constitutive of a living being.
Although certainly many (but not all) living beings engage in sexual behavior
that leads to reproduction, self-reproduction is not constitutive of a living being
in the self-realizationist framework.
In short, there is, in effect, no adaptation of the living being to the envi-
ronment in the self-reproductionist sense of (relatively more or less good)
adaptation to the conditions of a preexisting environment. Considered from

the internal dynamics of the living being, either the living being maintains its
organization or it dies, which means that from the point of view of the internal
operation of the living being, the environment either does not exist or is irrel-
evant. Similarly, there is also no selection by the environment or changes that
the living being undergoes. From the point of view of the observer who cor-
relates the living being’s way of living to its environment, the observer can say
that the structure of the environment triggers changes in the living being, but it
is only the self-realizing organization of the living being that determines, i.e.,
specifies, which configurations of the environment can trigger changes in it.
The observer can say that what counts as the environment for the organism
is determined by the structure of that kind of organism and that the environ-
ment is maintained and/or altered through the repeated organism–environ-
ment interactions in that organism’s ongoing self-realization, i.e., during
the course of its epigenesis or ontogeny. At the same time, the observer
may be interested to state how or why certain features of the environment
are able to trigger particular changes in the organism, in which case the
observer will wish to make statements about the organism’s lineage, or phy-
logeny. In both cases, the observer will say that organism and environment
are involved in a dynamic of relations and interactions through which they
change together in a congruent manner, or co-evolve. However, in neither
case can the observer say that the environment selects the organism, nor that
the organism’s environment preexists the organism, nor that the organism
is adapting itself to the environment. Thus it is more accurate to say that
organism and e nvironment co-drift.
The description of the dynamics of the ongoing realization of the organism
in relationship and interaction with the environment, then, is of central interest
to the observer, in terms of both the organism’s ontogeny (= the organism’s
ongoing self-realization) and lineage phylogeny (= the history of what the
observer classes as lineages of such ontogenies). Since the encounter between
organism and environment is through the organism’s ongoing ontogenic reali-
zation, those who pursue descriptions in a self-realizationist framework will
view the organism’s behavior as the fundamental agent in establishing the path
followed both by the organism’s ontogeny and, then, by the lineage phylogeny.
Since, for those who pursue descriptions in a self-realizationist framework,
there is no adaptation of the organism to the environment and no selection by
the environment of the organism, the organism’s ontogenic realization will be
seen as spontaneous (i.e., undesigned) and called ontogenic drift. Similarly, its
phylogenic history will be seen not as evolution, but as phylogenic drift. If it
is the case that certain lineages of organisms form part of the environment of
another lineage of organisms, they will not be said to be in competition with
one another but rather, and once again, in co-drift. Phylogenic co-drift, which
is a process of continuous speciation, is also a process of continuous conserva-

tion of the various systems that constitute lineages.
For instance, in the ongoing realization of any organism classed as a homo
sapiens sapiens are realized simultaneously the systems that also constitute
primates, mammals and vertebrates. This point is sometimes lost in the tra-
ditional representation of historical relationships among species in terms of
the branching-tree model. But there are alternatives to showing historical rela-
tionships both among languages and among species in what may be called a
flower-petal model of speciation. The flower-petal model reminds us that we
are at every moment vertebrates and mammals as much as we are primates.
We realize our vertebrate selves at every moment in our ontogenic drift and in
no more and no less representative a fashion of the category vertebrate than
do, say, snakes, and we realize our mammalian selves at every moment in our
ontogenic drift in no more and no less representative a fashion of the category
mammal than do, say, cows. The same goes for our primate selves, although
there is much in the evolutionary literature that wants to stress human unique-
ness, thus making the traditional tree a convenient way for humans to end up
on top. A typical ideogram appears in Jablonka and Lamb’s Evolution in Four
Dimensions, where they depict an evolution mountain that begins at the bot-
tom with genetic and epigenetic inheritance systems, goes up to behavioral
systems, and culminates at the top with symbolic systems (2005: 341). Among
the many interests of Maturana and his co-authors’ work is their penchant for
redirecting perception. In The Tree of Knowledge, Maturana and Varela repre-
sent the natural drift of living beings in terms of distances of complexity with
respect to their common origin, as if they are looking down on the evolution of
organism. They picture this drift as a large drop of water that has plopped on
top of a hill and then splits into different courses on its way down, some drop-
lets spidering out and away very far from the origin, while others stay quite
close and/or stop moving (1992 [1987]: 113) (Figure 5.1).
Here we encounter again the point raised in the previous chapter about the
insufficiencies of the tree model. It would be counterproductive to say that the
tree does not capture important features of much of evolution. But it is also the
case that it does not provide imaginative space for other important features of
evolutionary history, e.g., the fact that “the majority of genes in most genomes,
because they are patchily distributed within their respective species, phyla, or
domains, perforce have complex (and non-‘vertical’) evolutionary histories”
(Doolittle and Bapteste 2007: 2046). Pattern pluralism – the idea that different
models and representations can capture different features of the relationships
among organisms – opens the way to new models without advocating abandon-
ing the old, e.g., the tree. The choice of representation depends on the perspec-
tive one is seeking to illustrate. My preference in this chapter is for the water
droplet and flower petal.
Figure 5.1 The natural drift of living beings as distances of

complexity with respect to their common origin
The perspective advanced here puts at issue how we represent and then,
often unconsciously or out of habit, privilege exemplars of the taxonomic cat-
egories we create. Now, just as we could agree that no species would qualify
as the best representative of any of the categories vertebrate, mammal, etc. – is
a fish a better or worse example of a vertebrate than a kangaroo? is a mouse
a better or worse example of a mammal than a chimpanzee? – we might also
agree that no species is the least representative, i.e., the most unique, example
of any of these categories either. Humans are a particular kind of primate, to
be sure – that is what a species classification is, after all, a classification by
particular kind – but there is no reason that humans are the most particular kind
of primate. It is sometimes said that humans are the most generalized primate,
since we have come to inhabit the widest range of niches.1 However, the view
of humans as the most unique species is, precisely, part of the millennia-long
heritage of a rationalized view of our subject matter, with the capacity for lan-
guage offered as the crown jewel.
1
It would be interesting to consider how Western philosophical views of human beings and
their place in the world might have been different if Europe has been a habitat for non-human
primates.
It follows from all of the above that genotype and phenotype occupy different
places in the network of ideas that give rise to the self-realizationist framework
than they do in the self-reproductionist framework. In the self-realizationist
framework, there is only structural determinism and no genetic determinism.
Although the genotype is a starting point that defines or establishes a particu-
lar domain of possibilities, it is only the ontogenic phenotype that encounters
the environment and, therefore, it is only the ontogenic phenotype that sets
boundaries to genetic drift. Evolutionary history is not, then, the history of
genetic change but rather a history of conservation and change of lineages of
organisms. This includes the history of conservation and change of ways of liv-
ing of those lineages in which ontogenic phenotypes classed in those lineages
realize their ongoing operation.
Sexuality certainly modifies the phenomenon of reproduction by establish-
ing closed nets of genetic flux, but sexuality is as much a systemic phenom-
enon as any other that is maintained in the dynamics of the living being as
it realizes itself in its environment. Reproductive isolation, which provided
Mayr with his definition of a species, is a particular kind of conservation of an
ontogenic phenotype, and in the self-realizationist framework, sexual lineages
are understood to be constituted and conserved in the same way in phylogenic
drift as asexual ones. Any act or product of human reproduction, involving
whatever specific characteristics one wishes to attribute to it, does not present
an exemption from the conservation of the ontogenic phenotype as understood
in the foregoing discussion.
To repeat: sexuality certainly modifies the phenomenon of reproduction by
establishing closed nets of genetic flux, but it is as much a systemic phenom-
enon as any other that is maintained in the dynamics of the living being as it
realizes itself in its environment. Any act or product of human reproduction,
involving whatever specific characteristics one wishes to attribute to it, does not
present an exemption from the conservation of the ontogenic phenotype …
You might be scratching your head at this point, because it may sound to
you like the self-realizationist framework is demoting something as impor-
tant as sexuality. Sex is a big deal, right? Social conventions (try to) establish
and regulate who has sex when and with whom. Certain religious practices
and some sporting practices prohibit it. Books, magazines, movies, YouTube,
you-name-it are preoccupied by it. Which is to say that people tend to think
about sex frequently – okay, some of us think about it all the time. Now, the
self-realizationist framework is not saying that sexuality is not important to
evolution, but it is giving reproduction a different explanatory emphasis in the
framework. Sexuality certainly modifies the phenomenon of evolution, but it
is not the motor driving it. There is no motor; there is only the (differential)
conservation of ontogenic phenotypes. Ontogenic phenotypes may engage in
lots and lots of sex or none at all, whether they belong to lineages that sexually
reproduce or not. (The dependence of the concept of biological species on

reproductive isolation among sexually reproducing organisms means that it
cannot easily be applied, for example, to bacteria and organisms that reproduce
clonally or parthenogenically.) When ontogenic phenotypes that are classed as
a certain species fail to continue to realize themselves, i.e., maintain their con-
servation, for whatever reasons, that lineage is said to die. All this is to say that
the self-realizationist framework unhooks its definition of the term species (the
conservation of a way of living) from the traditional mechanism of speciation
(sexual reproduction).
Notably in the self-realizationist framework, there is no problem of
accounting for non-adaptive features or the effects of so-called deleterious
alleles or even of so-called salubrious alleles, since the observer is not con-
cerned with assigning a function or a biological reason for the existence of a
trait that justifies its presence in a selective history. Both ontogenic drift and
phylogenic drift are processes constituted in the conservation of organization
of living beings in their ongoing existence. These processes conserve every-
thing that belongs to their realization as totalities no matter how any part of
the “everything” participates at any particular moment in the realization of
the living being. In this respect, Futuyma is quite right when he says that:
“[deleterious] alleles are not deleterious in the new environment in which
we now live. Unaided vision is an adaptation to an environment that the
readers of this book no longer know; myopia and astigmatism are no longer
detrimental” (1986: 524–26). This is to say that organisms are continuously
realizing themselves in their environments, and those organisms classed as
human living beings have been realizing themselves in co-drift with their
environment since the moment they can be said to have begun realizing
themselves as human beings as such.
One aspect of the way of living of human beings significant to the present
discussion is the description of languaging. In the self-realizationist frame-
work, languaging is not understood as a unique and genetically determined
advantage conferred on human beings through the competitive struggle for
existence by their superior ability to accurately perceive a preexisting reality
and to share that perceptual knowledge through propositional structures with
another of their kin or kind. Rather, languaging is a particular aspect of a way
of living that has been conserved in phylogenic drift and is continuously real-
ized in the ongoing operation of ontogenic phenotypes classed as languaging
living beings, which we are now prepared to define.
Languaging is a behavior that emerged and is maintained through the con-
sensual coordinations of other languaging living beings and not as an adapta-
tion to a preexisting reality. The significant environment for languaging living
beings is other languaging living beings, which makes languaging a form
of consensual coordination among those living beings that participate in the
languaging way of living. As such, languaging does not exist as a reference to

another reality existing outside of languaging but rather shapes the so-called
reality in which the languaging living beings maintain their ongoing realiza-
tion. We could just as easily call this reality human ecology, the sociocultural
world, or a historically elaborated harmony of illusions.
There is no time after which languaging began that other human activities
could then thereby begin, and there is, in effect, no other human activity that
happens without languaging. Languaging is thus constitutive of living beings
usually referred to as humans. The lineage of languaging living beings begins
whenever we, namely the subset of us languaging living beings who take an
interest in such things, determine that our lineage begins. The circularity is
deliberate and, indeed, inevitable.
In light of the above description of the self-realizationist framework, you
may see that Maturana and Mpodozis have taken issue with the following
four tenets of the mainstream of evolutionary theory: (i) the medium preex-
ists the living being that occupies it; (ii) adaptation is a variable, so that it
is meaningful to talk about organisms that are more or less adapted, as well
as about adaptive processes and strategies; (iii) the evolutionary process is a
process of genetic change, the evolutionary unit is the population, and there-
fore evolutionary change happens as a change in the genetic composition of
populations; and (iv) every change requires the application of some sort of
force to be produced. In the biological process, selective pressure constitutes
this force, which, under the form of competition, leads continuously to the
survival of the (relatively) best adapted. From this conceptual point of view,
the differential survival of living beings, i.e., natural selection, is the mecha-
nism that generates the change, and not a result of any other more basic
biological process.
At the same time, Maturana and Mpodozis see their work as continuous
with Darwin’s evolutionary theory and not as a revolutionary departure from
it. This is, by the way, a change in style from those who claimed a Chomskyan
revolution that has subsequently been mythologized for many decades. As for
Maturana and Mpodozis, they acknowledge their continuity with Darwin’s
evolutionary theory by addressing the questions they believe gave rise to evo-
lutionary thought in the first place: (i) how can we explain the diversity and
similarity we observe among living beings? (ii) how can we explain that the
different types of living beings currently existing are living in their natural
domain in such a way that is totally congruent with their circumstances and
that, when this congruence is broken, they die? (iii) how can we explain that
taxonomists, who frequently classify living beings considering only a few
dimensions of their ongoing existences, can classify them in systematic cat-
egories that order and relate them in such a way that the result is biologically
significant?; and (iv) what is it that the taxonomist distinguishes when he or
she classifies a living being and formulates a taxonomic category with bio-
logical meaning?
Although Maturana and Mpodozis endorse the importance of these ques-
tions, they believe that the questions require a reformulation in terms of the
conceptual approach that considers living beings as autopoietic systems. Most
notably, their reformulation recognizes the operation of living beings as struc-
turally determined systems that exist insofar as they maintain their organiza-
tion and congruence with their circumstances. As such and has been said,
Maturana and Mpodozis reassert the role played by behavior as a guide to the
becoming and to the ongoing existence of all living beings. In a very concrete
way, then, Maturana and Mpodozis have turned the modern synthesis inside
out and made the encounter of the living being with its environment – the way
of living of the living being – the most important locus of ontogenic activity
and, therefore, the most important locus of investigation. Instead of evolu-
tion being driven by genes and genetic mutation in a population, as in the
self-reproductionist framework, evolution in the self-realizationist framework
is guided by individual phenotypes that conserve a particular way of living
such that they can be classed in one particular lineage or another. The cart of
genetic change is pulled by the horse of phenotypic activity, rather than the
other way around.
For Maturana and Mpodozis, then, the total genotype, and then the genome
as well, may change, as they are pulled along by the drift of the ontogenic phe-
notype, and not the other way around. When the shifting or phylogenic drift of
the total genotype and genome is produced in such a way that it is delimited by
the conservation or change of the ontogenic phenotype, the gene variations that
do not directly affect the realization of the epigenesis of this ontogenic phe-
notype remain protected (are conserved) in the genetic system. This protected
genetic variability allows for the epigenetic variability that makes possible the
shifting of the ontogenic phenotype in phylogenic drift. This is so because
genetic determination in a strict sense does not occur, since the living being
realizes itself in an epigenetic dynamics. Due to gene protection or hiding,
neutral genetic change can and does occur.
The theoretical innovation of Maturana and Mpodozis receives both support
and important modifications in the work of Jablonka and Lamb. The latter two
also suggest that changes at the genetic level can be seen to stabilize adaptive
modifications that are initiated through phenotypic plasticity and inheritance
mechanisms that go beyond the genetic and that include epigenetic control
mechanisms, as well as behavior-influencing substances, e.g., pheromones, and
symbolic manipulation (languaging). The important modification of Maturana
and Mpodozis’s work involves the investigation of epigenetical inheritance
systems, which are not a part of Maturana and Mpodozis’s account, for it has
become increasingly clear that it is not individual genes but rather large gene
A continuist script for languaging 145
networks that play major roles in development. Jablonka and Lamb fill in, in
a sense, a large part of the biological story that Maturana and Mpodozis leave
out. Like Maturana and Mpodozis, Jablonka and Lamb see their work not as
abandoning Darwinism but rather expanding it.
A continuist script for languaging

Languaging has much in common with other human social behavior and with
the social interactions of other types of organisms, most notably non-human
primates, and with the vocal interactions of other types of organisms, most
notably birds and sea mammals such as dolphins and whales. In other words,
many groups of organisms exhibit behaviors that we languaging living beings
describe as belonging to a linguistic domain. That is, we can observe patterns
of recurrent behaviors among groups of individuals whose ontogenic interac-
tions we observe to be widely coincident in ontogeny and whose particular way
of living we observe to be preserved in phylogeny. Accordingly, we class these
organisms as belonging to the same species.
Particular to the types of recurrent behavior characteristic of those organ-
isms we class as humans are the extra loops, so to speak, that have arisen in
the recurrent coordinations of our behaviors in the course of (the successive
ontogenies of organisms belonging to) our lineage. These extra loops have
become a stable and, indeed, pervasive part of our niche. They coordinate the
coordination of our behaviors and have given rise to a new domain, the domain
of languaging. Here the notion of orienting is stressed in order to maintain the
distinction between our domains of description: that of the observer and that of
the individuals involved in the interaction under observation.
The script: the behavior of one human modified the behavior of another
human by engaging in a behavior that oriented the second human in her cog-
nitive domain to an interaction that was independent of the features of the
orienting interaction itself. That is, it was not the orienter’s behavior but rather
some effect of it that prompted a response in the orientee. Her response, in
turn, oriented the first human in his cognitive domain which, in turn, elicited
a response that once again oriented her in her domain, and so on, such that all
subsequent conduct depended on the outcome of their independent, although
coordinated, interactions. It does not matter which particular objects or events
the first outcomes brought forth – i.e., it does not matter which dependent vari-
ables were created from the independent variables – only that the outcomes
could be seen (by an observer who, in this case, is now one or both or all of
participants in the interaction) to coordinate the coordinations. In other words,
we say that languaging began when human behaviors began to organize them-
selves around observed effects of coordinations, and the behaviors that stabi-
lized around these observed effects (signs, words, phrases, gestures, marks)
constituted, in the history of lineages, a new domain of interaction: the coordi-

nation of coordinations.
One venerable candidate for the first word, which goes back at least to
Jespersen (1922) and which is taken up by MacNeilage, “is that the female
parent, in the presence of the nasal demand vocalization made directly to her
[by a nursing infant] decided that ‘This sound stands for me’” (2008: 150).
The worldwide relationship between nursery terms for mothers and even milk
which include a nasal is well known. Such a possibility is not objectionable. It
even gains support from studies of the anterior cingulated cortex, which plays a
role in animal communication and which suggests the possibility that “the iso-
lation call or ‘isolation peep,’ which young animals make when out of contact
with their parents, may have been the first vocalization to evolve in mammals”
(MacNeilage 2008: 172). Humans, good mammals that we are, need not be
expected to behave differently.
In any case, the foregoing script makes no demands of an increase in intel-
ligence for these coordinations of coordinations to have occurred. Rather, what
we have come to call human intelligence is a result of the long history of these
coordinations of coordinations. Coordinations of coordinations begin when
one individual’s – or a group of individuals’ – response to an otherwise routine
behavioral coordination, e.g., a vocalization or a gesture, ceases to be pre-
dictable. Perhaps the vocal signal is ambiguous or even mistaken (the mother
misinterprets the infant’s sucking sound). At the very least, the vocalization
or the gesture must be open to interpretation. It is in the interpretive moment
that languaging begins; and it is the response that flows from the (mis)inter-
pretation that determines what happens next, for that response will affect the
subsequent behavior of the beginning languager who must now orient himself
to the lack of the predictable response. The interpretive moment might also be
called taking a perspective on a situation, and it crystallizes into languaging
when it has some kind of uptake by the original vocalizer that the interpreter
made available. Perhaps Skinner was on to something when he noted that “the
‘listener’ must be responding in ways that have been conditioned precisely in
order to reinforce the behavior of the speaker” (1957: 225). As discussed in
Chapter 1, listening is not a throw-away category in a discussion of languaging
behaviors, and surely human listening abilities have been shaped evolutionar-
ily just as much as human speaking abilities have.
But now how does a response to an otherwise routine behavioral coordina-
tion that includes a vocalization come to cease to be predictable and how do
routine vocalizations come to be open to interpretation? Another way to ask the
question is to wonder how the ability to co-opt action orchestration by means
other than by the action itself got instantiated in human brains. A first, direct
response to the question comes from several recent studies that suggest that the
pathway was through gesture. They begin with the fact that the modern human
hand developed before the modern human vocal tract, and they ultimately link
the development of language to the existence of mirror neurons (Arbib and
Rizzolatti 1997; Rizzolatti and Arbib 1998; Corballis 2002; McNeill 2005,
2012; Armstrong 2008). As mentioned briefly in the previous chapter, mirror
neurons – such as those identified in the macaque – respond to actions inde-
pendently of whether the monkey sees, hears or performs those actions. These
neurons were so dubbed by Rizzolatti because they seem to provide a mirror
between action and perception, between what you do and what you see. That
is, there were discovered to be cells that responded to production and observa-
tion of the same action – whether the monkey grasped a raisin from a tray or
watched another individual grasp a raisin from the same tray – meaning that
the cells responsive to sensory stimulation and motor production were congru-
ent. This congruence is the property for which they are named (Venezia and
Hickok 2009: 1404).
The idea here is that a gestural protolanguage led the way into what we now
have in speech through the mirror neurons, where “meaningfulness depends
on simulating a social response – the social response of another in yourself,
a kind of auto-socialization,” as McNeill has phrased it. Given what was said
at the end of the last chapter about us having always inhabited a noisy world,
McNeill’s and MacNeilage’s positions are most consistent with the framework
outlined in the present study because they do not serialize a gestural protolan-
guage to precede speech (see also Fitch 2010: 433–65 who similarly rejects
gestural protolanguage theories). McNeill, for his part, speculates that the
critical evolutionary event might have been a new self-response by humans’
mirror neurons, and he sees the two kinds of sequential actions – speech and
gesture – converging, “with significances other than motor actions themselves
as the integrating component. Such an evolution brought meanings into areas
44 and 45, enabling significances other than instrumental actions themselves
to orchestrate complex, sequential movements. These movements were both
manual and oral/pharyngeal/respiratory, and were organized in the too nar-
rowly labeled ‘language area’” (2005: 251). More recently, McNeill has pos-
ited a new adaptation in the evolution of humans whereby mirror neurons were
“twisted” to respond to one’s own gestures, distinct from the “straight” mirror
neurons, which respond to the actions of others. According to McNeill, this
twisting accomplished two things: (i) “it gave to one’s own gesture the sense
of being social and public”; and (ii) “by mirroring one’s own gestures and
their significance, the new ‘twisted’ mirror neurons made the gesture and its
imagery available in Broca’s area” (2012: 65).
MacNeilage, for his part, given his attention to the mandibular cycle as the
place to look for the origin of speech, also does not give priority to gesture but
does not rule out its importance. As for the mirror neurons, their importance
is clear:
In animals that evolved to move in order to acquire food, independence of direct stim-
ulus-response relationship develops, but even here the causal sequence input-to-output
remains important. But mirror neurons have enabled a radical reversal of this causal
sequence, a development which can hardly be overemphasized. Now the organism’s
reflexive (turned back on itself) use of its own output representation capabilities serves
as a guide for more highly adaptive processing of input. Speech learnability is only
one of a panoply of important emergent functions that this evolutionary development
allows. (MacNeilage 2008: 324–25)
Although mirror neurons – which had not been identified when, for instance,
Maturana and Varela were developing their ideas about autopoeisis and lan-
guaging – whether “twisted” or not, may play an important role in anyone’s
theory of the instantiation, development and maintenance of language/ing, they
do account for why humans are able to coordinate their coordinations so effec-
tively. In addition, a recent study – one not concerned with mirror neurons as
such – has found widespread speaker–listener neural coupling, extending well
beyond the low-level auditory areas, when communication occurs. Existing
neurolinguistic studies have been mostly concerned with either speech pro-
duction or speech comprehension. The interest of this new study is that the
listener’s brain was discovered to exhibit predictive anticipatory responses
when hearing a story told, with the greater the anticipatory speaker–listener
coupling, the greater the understanding (Stephens et al. 2010).
A second, indirect strategy for answering the question of how the ability
to co-opt action orchestration by means other than by the action itself got
instantiated is to review the range of scenarios that have been proposed for
the origin of language in the hope that a glimmer of understanding may fall
out from the mass, and this review is undertaken in the last section of this
chapter. Although this strategy is not wholly satisfactory, tracing the history
of such contingencies in a lineage is the equivalent of writing the evolution-
ary script for that lineage in the self-realizationist framework. The course
of what could be seen as accidents constitutes drift and does not contrast
itself to any other explanation of change, e.g., invoking a mechanism such as
natural selection.
Languaging, as a particular type of coordination (of coordination) of behav-
ior, has enabled those of us who operate in it to describe ourselves and our
circumstances. Our descriptions, then, are the medium in which we realize
ourselves as human beings as such. One of our descriptions is of ourselves
as observers, and observing occurs in languaging, just as languaging occurs
in observing, as the above script illustrates. Another of our descriptions con-
cerns our subject matter itself, and one description of it that has had particular
presence in the historical record is code theorism. The reason to shift to an
orientation toward our subject matter as languaging as an orienting behavior is

not because this latter description is now somehow right. Rather, we adopt it, in
part, because it fits better (is more continuous) with neurobiological discourse
that accepts the understanding of the operational closure of the nervous system
and that in the last hundred-plus years has shifted away from realist repre-
sentationalism. These developments, in turn, have flowed out of a Darwinian
understanding of speciation that is still being developed and can be, loop-like,
affected by the description of languaging that has yet to be integrated into the
discipline of linguistics.
Let us apply this (new) orientation to the New Chimp Language Debates.
These have taken up where the Chimp Language Debates of the 1970s left
off, that is, when Herbert Terrance, who had trained Nim Chimpsky and was
disappointed with Nim’s progress, declared Noam Chomsky and his linguis-
tic theory the winner. Since the mid-to-late 1980s, a team of primatologists
at the Language Research Center at Georgia State headed by Sue Savage-
Rumbaugh have proposed that the language abilities of their pygmy chimps
Kanzi, Mulika, Sherman and Austin went beyond the “limitations of instru-
mentally conditioned response patterns” and into the realm of communica-
tions that are “representational,” that is, of the type presupposed as central
to human communications. Savage-Rumbaugh and her colleagues identified
the chimps’ language abilities to be representational when they grouped not
just objects but also lexigrams into proper superordinate categories even when
the referents were absent. Seidenberg and Petitto (1987) then took issue with
Savage-Rumbaugh’s analysis and argued rather that non-human primates are
not similar enough to human primates for their respective behaviors to be
described in similar terms.
It is easy to be sympathetic to the position taken by Savage-Rumbaugh et al.
that in whatever terms we want to describe what humans do when they lan-
guage, we can describe what chimps do similarly. That is, we observe behavior
in chimps that we recognize as being (more or less widely) coincident with
ours, which is entirely to be expected since we have classed our currently sepa-
rate lineages in a prior unified lineage. So it is not surprising to us to see that
chimps are able to follow our lead and to organize, to a certain extent, their
interactions with us by means of coordinations of coordinations and even, on
occasion, to so organize some of their interactions with their fellow chimps. At
the same time, it is easy to be sympathetic to Seidenberg and Petitto’s position
which could be glossed in terms of Pinker’s observation that what is so interest-
ing about chimpanzee signing is that “fundamentally, deep down, chimps just
don’t ‘get it’” (1994: 340). The chimp way of living conserved in the chimp
lineage is not the same as the human way of living conserved in the human
lineage. Their drift has not included the instantiation and development of the
domain of coordinations of coordinations of behavior.
However, both positions must be rejected for their shared reliance on the
descriptive terminology that includes mental representations, because to con-
tinue to use the term mental representations to describe both human and non-
human primate cognition collapses any distinctions we may wish to make
among the various cognitions of the species of the world. In other words, there
is no one independent reality “out there” that all organisms bring inside their
heads and that gets somehow represented; rather, different organisms realize
different behavioral and cognitive domains. The benefit of a developmental
systems approach is that it allows us both to acknowledge similarities and to
respect difference among organisms, and this without resorting to polarized
descriptions whereby either “all organisms exist on some kind of behavio-
ral cognitive continuum that can be measured against their relationship to an
external reality” or “chimps just don’t ‘get it’.”
And the debate just does not seem to want to go away. Recent work by com-
parative cognitive scientist Heidi Lyn analyzing the 1,497 errors that Kanzi
and Panbanisha, a female bonobo, made while matching lexigrams over 12,157
trials shows evidence that the lexigrams are organized in their heads in a com-
plex, hierarchical manner. That is, when the chimps erred, their mistakes were
not random. For instance, if the lexigram stood for ‘blackberry,’ the error was
likely to sound like blackberry, be edible, be a fruit or be physically similar.
This led Lyn to think that the chimps worked with these terms not as one-to-
one associations but in more complex mental arrangements and “that bonobos
and other apes can acquire systems of meaning that are closer than anyone has
thought to what humans do, and that some aspects of language acquisition are
not unique to humans” (Erard 2007: 1675). At the same time, primatologist
Robert Seyfarth notes that the apes confused fewer target-error pairs that were
either both nouns or both verbs, implying that they do not distinguish parts
of speech. He took this result to argue “against the claims made elsewhere
that Kanzi has spontaneously developed an elementary grammar” (Erard 2007:
1676). Again, I can only agree with both and disagree with the terms of their
disagreement.
The debate about the mental organization in a non-human primate’s head
compared to that in a human head is similar to the debate about mamma-
lian auditory systems played out both in Hauser, Chomsky and Fitch and in
Pinker and Jackendoff. For instance, Pinker and Jackendoff write that Hauser,
Chomsky and Fitch “suggest that there have been no evolutionary changes
to the mammalian auditory system for the function of speech perception in
humans” (Pinker and Jackendoff 2005: 206ff.) and then proceed to list all
the ways that non-human primates “fail” to perceive speech in the ways that
humans do. Pinker and Jackendoff are right to question the idea that there
have been no evolutionary changes to the mammalian auditory system for the
function of speech perception. The human auditory domain cannot not have
Living our cognitive domains 151
been significantly shaped by the emergence, development and maintenance of

languaging in our lineage which has occurred over the past two million years,
with the more or less current form of it becoming recognizable (to our cur-
rent observers’ eyes and ears) around 100,000+ years ago. By the same token,
Pinker and Jackendoff would readily acknowledge that humans would “fail”
to perceive the species’ calls of non-human primates in the way that the non-
human primates do.
Living our cognitive domains

Just as different organisms realize different cognitive and auditory domains, so
they also realize different visual domains. We return to the so-called problem
of vision that arose at the end of the last chapter.
Certainly all primates can be seen to share certain sensory/motor/cognitive
structural features (as do all mammals, all vertebrates, etc.), which is precisely
why, as has just been said, we are able to infer a shared phylogeny and label
them primates (mammals, vertebrates, etc.). However, from the point of view
of the self-realizationist framework, chimpanzee vision, by definition, is not
the same as human vision, since it is chimpanzee vision. And there is no sense
in which any organism’s vision and any lineage’s visual evolution can be cali-
brated along an accuracy meter. That is, a lineage’s visual evolution is not
toward more accuracy (accurate with respect to what?) and, as such, no organ-
ism’s vision is (more or less) accurate than any other’s.
Certainly, we can reconstruct an evolutionary story for the visual mecha-
nisms characteristic of mammals, and we can say that we see an increase in
complexity, in the sense that we can identify more parts (light receptors, lens,
iris, etc.) as well as more connections among those parts. However, when we
embrace the concept of autopoiesis, we see that the organisms that we postu-
late to have existed early on in that lineage did not have less accurate vision
than the organisms that came later. All we can say is that the organisms that
came early on in the lineage lived their visual domains, just as the organisms
that have come later on in the lineage have lived/are living theirs.
The usual story told about vision and evolution is that every stage that we
can reconstruct for it signals a step toward an improvement of vision. The result
of such a story is the belief that we (humans) now enjoy vision that is accu-
rate, even astonishingly accurate, according to some. However, this particular
evolutionary story does not produce the belief in the accuracy of our vision;
it is rather the result of our belief in the accuracy of our vision. Of course, we
think our vision is accurate. We succeed in our environment with it. We have
organized our entire world around the way it functions, and the way our world
functions supports our visual domain. If we have problems in our world that
we diagnose to be related to our vision, we go to the optometrist to correct it
by means of glasses, contact lenses or laser surgery, that is, to make it conform
to the species norm that allows us to function effectively in the world that we
have organized for ourselves. However, at no moment are we seeing the world
the way it really is, somehow independent of our visual apparatus. Rather, we
are living our visual domain.
Similarly, with respect to languaging, the course of evolution has not been to
make our communication more accurate or more precise with respect to trans-
ferring the accurately perceived picture of something or another from the head
of a speaker into the head of a listener by means of coding thoughts into the
formal containers of language and sending them through the serial channel of
speech or signs. As for languaging, it is surely the case that over the course of
the millennia, we languaging living beings have increased the range and com-
plexity of our consensual coordinations. However, to say that we have improved
them in any way would require establishing what would count as an improve-
ment. As for the reasons for the increase in range and complexity, it should be
pointed out, first, that there are more of us now (seven billion and counting) to
elaborate our languaging together than there were in the Pleistocene, creating
effects similar to those in large-scale capitalist economies in producing and
spreading wealth: more individuals create more of everything. Second, there
are also more languaging living beings in our lineage now than there were
for any group of languaging living beings in the Pleistocene. This means that
languaging living beings today have inherited more languaging possibilities
from more forebears than languaging living beings living 100,000+ years ago
could have. In other words, there are more languages and cultures today than
there were 100,000+ years ago and more time behind them in which possi-
bilities could loop, expand, crisscross and tangle. For instance, the activity of
languaging about linguistic historiography was not a possibility available in
the Pleistocene because it requires, at the very least, the support of a developed
writing system, print culture and university structure for it to be elaborated.
On complexity
The notion of complexity suggests a dual conception of nature: on one side
so-called simple phenomena, such as the motion of a pendulum or the equation
of state of a perfect gas; on the other, so-called complex phenomena involved
in life or human behavior. However, in recent decades this duality seems to be
vanishing. Who would have thought that particles thought to be elementary
would have such an intricate structure or that the motion of an elastic pendu-
lum would be comparable in complexity to that of fully developed turbulence?
A range of contemporary sciences – from non-equilibrium physics to econom-
ics to linguistics – seem to have rediscovered the importance of time. From the
origin of our universe to the origin of life, researchers have become aware of
In summary 153
how general is the phenomenon that highly unstable, dynamic systems – such
as cosmological or biological or linguistic systems – become more and more
complicated over time and that the complexity is irreversible.
Evolutionary theorists have long recognized a dual dimension to complexity
in evolution: that not only do the possible products of evolution become more
complex over time, so can the genetic system itself be subject to increasing
complexity. Clearly, the system of sexual reproduction found in a great many
organisms, based on two sexes whose gametes unite to produce offspring, is a
very complex mechanism and provides a way of recombining hereditary fac-
tors into a large number of new combinations. It itself is a product of evolu-
tion, since the most primitive living things, such as bacteria, do not possess
it, though some of them have less complex, so-called parasexual mechanisms
which make some degree of recombination possible.
Waddington noted as much and illustrated the phenomenon with an analogy
to the so-called evolution of factories as a result of the Industrial Revolution.
The development of factories (here Waddington invokes the forces of competi-
tion, which for the purposes of this analogy, he compares to natural selection)
not only brought about an evolution of the factories’ products (which corre-
spond to organisms) into more elaborately fabricated articles, but over time the
organization of the factories themselves along with the supply networks, i.e.,
the mechanisms by which the articles are produced, have become more elabo-
rate. “To take another example,” Waddington continues, “if a group of begin-
ners take up the practice of playing card games with one another, they would
not only become more skillful at playing the game they first start on, but are
likely to pass on to playing subtler and more complicated games. Thus, this, as
it were, two-tier evolution – an evolution of the end-product itself and also an
evolution of the mechanism by which the end-product comes into being – is
quite a normal sort of happening” (1975: 275). It is in the sense of complex-
ity as intricacy, complication and subtlety that I see the trajectory for human
languaging behaviors.
In summary
Languaging as an ongoing activity is linked: (i) to the nervous system as a
whole and not just to the area above the neck and/or cerebrum in the cortex,
and although operational localizations can be established in the nervous sys-
tem, these localizations can be usefully viewed in terms of areas where certain
neural patternings converge, and not as autonomous isolations of faculties or
functions; and (ii) to the phenotype, as opposed to the genotype. In this frame-
work then, languaging is doubly embodied.
To repeat: languaging is an orienting activity whereby languaging living
beings coordinate themselves through their mutually orienting behaviors. The
degree of effectiveness of these behavioral orientations will depend on the his-

tory of the structural coupling of two (or more) such languaging living beings
with respect both to their ontogenic and to their phylogenic histories. Humans
will, in general, have more effective and sustained coordinations of actions
with other humans, with whom they share a phylogenic history, than they will
with, say, non-human primates, cats, dogs, birds or wolves, but there may well
be exceptions that occur ontogenically, e.g., certain primatologists and certain
chimps may have highly effective consensual coordinations.
It follows that the conditions allowing the consensual coordinations we
observe operating today can be described as being roughly those that have
always shaped such practices. This is not to say that what we might iden-
tify as the human manner of living in, say, the Pleistocene, was the same as
that of today or that the cognitive abilities of languaging living beings in the
Pleistocene were the same as those of today. This is to say that: (i) the orient-
ing behavior of today can be seen to have arisen out of the orienting behavior
of our phylogenic past and erases a need to distinguish between evolutionary
time and cultural time; (ii) although our languaging did not arise as a result
of cognitive developments, this activity has had a profound impact on human
cognitive capacities, as languaging as a manner of living has been looped
in and through successive generations of languaging living beings since the
Pleistocene; this point is crucial to a developmental systems linguistics that is
interested to understand how languaging integrates cognitive capacities over
extended thematic passages and gives rise to phenomenological domains such
as an autobiographical self and narrative in general; (iii) differences in cogni-
tive style, personal history and/or momentary interests – i.e., interpretation –
among languaging living beings continue to crucially affect the coordination
of actions, just as they have from the very beginning; and (iv) the effects of
the differences in cognitive style, personal history and/or momentary interests
in the coordination of actions over time play themselves out in different ways
today than they did in the Pleistocene, given that the range and complexity of
our consensual coordinations have increased.
A range of evolutionary scenarios

The work of Chomsky, that of Chomsky et al., that of Pinker, that of Pinker
et al., that of Maturana, and that of Maturana et al. hardly exhausts the possi-
bilities of evolutionary scripts for our subject matter. They do, however, delimit
the possibilities in this respect: explanations congenial to a Chomskyan model
of language do not explain the appearance in the hominid lineage of the so-
called faculty of language as emerging from non-linguistic abilities or behav-
iors (in fact, they do not see the faculty of language as any kind of behavior at
all); explanations congenial to languaging as an orienting behavior do explain
A range of evolutionary scenarios 155
the appearance in the hominid lineage of languaging behavior as emerging

from non-linguistic abilities and behaviors. One of two logical possibilities
must be true: the features of whatever we choose to ascribe to the linguis-
tic domain either did or did not arise from (= can or cannot be explained in
terms of a development from) behaviors and/or capacities of non-linguistic
origin. Maturana mapped out the two positions almost forty years ago in “The
Biology of Cognition” (1970). Needless to say, in these past forty years, many
more scenarios have filled in the delimited space and have come from a wide
range of disciplines. Beginning at the beginning, such scenarios include those
discussed next.
The nineteenth-century linguistic heritage

An originating scenario could go something like this: two or more of our ances-
tors are walking down a path together (“coordinated” behavior, or behavior
that is oriented in the same direction); one stubs his toe on a stone and utters
“Grh”; another or others of our ancestors associate(s) the sound sequence with
the stone itself; henceforth the sound sequence [grh] serves to coordinate their
coordinations with respect to the stone. Or it could go something like this:
two or more of our ancestors are in grooming relationships with one another
(coordinated behavior); one of our ancestors has a habit of picking nits from
the back of another while uttering “Ng”; perhaps that individual is concen-
trating very hard on the task or sighing in satisfaction at finding another one;
over time, another or others of our ancestors associate(s) the sound sequence
[ŋ], not with the internal dispositional state of the utterer, but with the nit that
is removed; henceforth the sound sequence [ŋ] coordinates their coordinations
with respect to nits.
The above scenarios are versions of theories proposed in the nineteenth cen-
tury and labeled, that is to say, stigmatized by Max Müller as the “pooh-pooh”
theory, whereby speech arose from grunts of emotion. In the early twentieth
century, Jespersen took up the term and argued against such a theory on the
grounds that interjections make up only a very small part of any repertoire.
Also, as Terrence Deacon and others, including Skinner and Tomasello, and
Pinker and Jackendoff, have pointed out, human verbalizations, unlike other
non-human primate vocalizations, are mediated through the cortex and are
produced in a very different way than cries of emotion such as laughing or
sobbing.
Or an originating scenario could go something like this: our ancestors, being
social creatures, likely engaged in many group activities other than grooming.
They might have organized themselves around food gathering forays such as
hunting or searching for nuts and berries (coordinated behavior). Whatever
sounds emitted for whatever reasons and accompanied by whatever bodily
activities and interactions stabilized and came to be conventionalized for those

activities.
Jespersen, who extended Müller’s categories, would have filed this scenario
under the “yo-he-ho” theory of speech, whose origin lies in group efforts, typi-
cally physical group efforts that produced some kind of vocalizations. In the
nineteenth-century version of this scenario, these vocalizations were grunts
that had rhythmic elements – thus, Jespersen’s disparagement of them in terms
of yo-he-ho – and these rhythmic elements were thought to lie at the basis of
the universal use of prosodic features. Origin scenarios that begin with physi-
cal exertion (chucking a spear, reaching or stooping for berries out of easy
reach) are yo-he-ho.
Then there is the “bow-wow” theory, whereby speech arose from imita-
tion of the sounds of the environment, evidenced by onomatopoetic words in
all languages, and the “ding-dong” theory, which posits that human speech
reflects the so-called natural correspondence between articulatory gestures
and meaning that is evidenced in sound symbolism. In The Roots of Thinking
(1990), Maxine Sheets-Johnstone offers just such a sensory-kinetic model of
the origin of language. She focuses her attention not on the larynx, as does
Deacon, but on the tongue as the mediating center of sound and as a source of
touch as well. She proposes that at the origins of language/ing articulatory ges-
tures, sound and meaning were congruent, suggesting an original non-arbitrary
language perhaps not dissimilar to the one proposed in the eighteenth century
by Charles de Brosses, who thought that there was a resemblance between the
movement of the organ of articulation and the meaning that the sound con-
veyed (see Andresen 1981).
By the way, Müller (of “crossing the Rubicon” fame) was the one who
coined the phrase “bow-wow” to disparage the theory of his American rival
William Dwight Whitney, who did believe that the origin of language lay in
imitating the sounds of nature. Whitney, for his part, dubbed Müller’s posi-
tion the “ding-dong” theory to suggest Müller’s belief in an original per-
cussive resonance between the mind and its surroundings, such that at the
beginning man was like a kind of bell: when an idea struck him, he naturally
rang. Whitney argued against this position on uniformitarian grounds, that
is, Whitney did not know why an original percussive resonance should have
operated only at the beginnings of language and be nowhere in evidence
today. Now, Whitney and Müller both: (i) expected to find clues to the ori-
gins puzzle in Franz Bopp’s theory of monosyllabic roots; and (ii) agreed
that the first things named by the earliest humans would have been indi-
vidual objects – “cave,” “tree,” “fountain” – not groups of objects. In other
words, for both nineteenth-century linguists, language was a naming process.
Where they differed was whether a speaker would need to specify a general
conception verbally before s/he could use it to create a noun. Whitney said
Yes, Müller No (Alter 2005). Although Whitney was operating out of a com-
monsense realism, where abstractions and general terms reflected similarities
that actually existed in nature, while Müller had a more Kantian cast to his
thinking, they both associated language with thought processes and not, say,
with the dynamics of social control.
Returning to Jespersen, his own preference was for the “la-la” theory, which
shares a bit with the Scheherazade effect mentioned below. “La-la” stresses
love and play and the warm fuzzies. It is, in a word, romantic, and the purpose
of language is to secure a mate.
Stripping away the romance, we find the study on a captive group of Guinea
baboons at the Brookfield Zoo in Chicago. A team of researchers recently dis-
covered that “unlike lots of other animals that call, sing, or whistle to advertise
fertility but fall silent after the deed is done, female baboons give loud stac-
cato grunts after mating. The grunting tends to be more intense after sex with
a higher-ranking, dominant male.” Before this study, researchers had thought
that the grunts were signals encouraging more males to compete for sex. But
this study showed that the female’s grunts actually put off other mates. The
idea here is that once a female baboon mates with a preferred male, it is in her
interest to give that male’s sperm the best chance to fertilize her eggs by avoid-
ing further copulation. Males comply by guarding a female that has grunted. In
other words, the grunts are intended (or at least have the effect) of making her
partner stick around (Maestripieri et al. 2005).
Anthropology
Taking a different tack with respect to love and language is Dean Falk’s
“putting the baby down” hypothesis for the origin of language. Here the focus
is on infant-directed communications that arose of necessity when hominin
mothers adopted new foraging strategies that required maternal controlling of
physically removed infants by means of reassurances and the like. The central
idea is that “bipedal mothers had to put their babies down next to them periodi-
cally in order to go about their business, and that prosodic vocalizations would
have replaced cradling arms as a means for keeping the little ones content”
(2004: 503). Over time, and as mothers increasingly used “prosodic and ges-
tural markings to encourage juveniles to behave and to follow, the meanings
of certain utterances (words) became conventionalized” (2004: 491). Falk is
willing to give credence to the ideas: (i) that the prosodic features of contem-
porary motherese have roots in the late australopithecines/early homo line of
the species; (ii) that language could have emerged from an earlier primate com-
munication system; and (iii) that language is a naming process.
As far as the naming process goes, she observes that the word mama is
produced in infants as early as two months old, usually as part of a cry. Falk’s
conclusion: “It does not seem unreasonable to suggest that the equivalent of
the English word ‘Mama’ may well have been one of the first conventional
words developed by early hominins. After all, wouldn’t maturing prelinguistic
infants, then as now, be inclined to put a name to the face that provided their
initial experiences of warmth, love, and reassuring melody?” (2004: 503). John
Newman, one of the many respondents to Falk’s proposal, backs her up by not-
ing the depth of infant-directed vocalizations in non-hominin mammals:
I have argued elsewhere … that the brain circuits underlying mother-infant communi-
cation arose early in mammalian evolution and have had a conservative evolutionary
history. In primates, a region of the rostral midline (limbic) cortex is important for the
expression of ‘crying’ sounds (isolation calls) … Recently, evidence has emerged indi-
cating that this same region is activated in human mothers upon hearing infant cries.
(2004: 520)
The general idea here is that there appears to be brain circuitry dedicated to
mother–infant communication in mammalian brain organization that human
language piggy-backed on and out of. Such a scenario has already been out-
lined, above, in the context of MacNeilage’s proposal.
Evolutionary psychology
Evolutionary psychologists have taken up the question of the origins of lan-
guage, partly in response to Pinker. In the chapter “Language” in the intro-
ductory textbook Human Evolutionary Psychology (2002) by Louise Barrett,
Robin Dunbar and John Lycett, the authors subscribe to (one version of) the
continuist position in their question: “How did hominids get from essentially
monkey-like vocalisations to semantically meaningful words?” (2002: 328).
They endorse the idea that “language evolved out of conventional primate
vocal communication by a simple elaboration of the same kinds of sound pat-
terns” and believe that “speech arose by elaboration from natural primate vocal
communication” (2002: 329). This is a position that was rejected by Chomsky,
Fitch, Hauser, Jackendoff, Pinker, Skinner and Müller. For Barrett, Dunbar and
Lycett, in a nutshell, “language is a form of grooming-at-a-distance designed to
facilitate the bonding of very large groups” (2002: 334). In other words, unlike
those who subscribe to a Chomskyan model of language, Barrett, Dunbar and
Lycett are interested to fill in the gap in evolutionary explanations by providing
a precursor function/behavior out of which the human version of bonding-at-
a-distance, i.e., language, emerged. They entertain three hypotheses about how
this occurred:
(i) The Social Gossip Hypothesis. Our social world is in a state of constant flux
as individuals make and break relationships. To operate in such a world, we
need to have an accurate model in our minds of who is currently friends
with whom, and this model needs to be constantly updated as relation-

ships come and go;
(ii) The Scheherazade Effect. One of the functions of language may be to
enable humans to attract and maintain the interest of mates. This ver-
sion of the social function hypothesis gives language a more exotic role,
because it suggests that one of its important functions is to dazzle or enter-
tain, rather than merely to facilitate the exchange of information; and
(iii) The Social Contract Hypothesis. In human social systems, marital units
(monogamous or polygamous families) live in close association within
a larger grouping (band or village). The problem is that, in the classic
hunter-gatherer setup in which we evolved, males who went hunting to
provide meat for their females left their mates at risk of being mated by
rivals; equally, their own mates were exposed to the risk that their hus-
bands might themselves mate with other females in neighboring groups.
This placed intolerable strains on the relationship involved. Marriage
contracts, whereby mates declared to each other – and perhaps publicly
to the rest of the group – their marital obligations and agreed not to mate
with other individuals, were the solution to the problem. The social con-
tract hypothesis is highly identified with Terrence Deacon’s work (see
below). Historiographically speaking, social contract hypotheses go back
to a kind of Lockean understanding of language as an agreement and usu-
ally contrast with theories that posit the origin of language in emotional
states and cries.
A hint of the social contract hypothesis is apparent in Karmiloff and

Karmiloff-Smith’s Pathways to Language: From Fetus to Adolescent when
they write: “Not only are words arbitrary, they are also conventional. By this
we mean that people learning English accept that ‘car’ refers to that kind of
vehicle, rather than each person choosing his own word to represent this mean-
ing. We would find it very difficult to communicate with one another if we did
not have such a tacit agreement” (2001: 58). Terms such as accept and tacit
agreement have a liberal democratic feel about language learning and using
processes that can also be detected in Tomasello. So does the description that:
“As the infant brain gradually learns to detect which sound combinations are
the most frequent in his language, he also becomes sensitive to which com-
binations of sounds are legal and which are not … In every language, some
sound combinations are illegal” (2001: 50, emphasis mine). In another pas-
sage, they write that: “words can act as an invitation to form a category” (2001:
68, emphasis mine). Overall, the social contract hypothesis posits independent
agents that enter uncoercively into exchanges that benefit both (or all) parties.
As a critique, Skinner noted that the idea of the supposed origin of a culture in
terms of a social contract or of social practices as commandments presupposes
a prior act of creation (that of the contract, the commandment) as the explan-
ation of the phenomenon to be explained (1981: 502).
Barrett, Dunbar and Lycett see the originating factor for getting-language-
off-the-ground to be the function of what they call grooming-at-a-distance.
Once simple language evolved to support the servicing of relationships within
the group, then the gossip hypothesis plus the social contract hypothesis, with
which the Scheherazade effect is compatible, provided the context in which the
development of more complex symbolic forms evolved.
Evolutionary neurobiology
Evolutionary neurobiologist Terrence Deacon’s The Symbolic Species: The
Co-evolution of Language and the Brain is a major contribution to the discus-
sion of the evolutionary origins of our subject matter. He guides the reader
through a detailed description of the development of human vocal skills in
the species that exceed the capabilities of any living non-human primate and
that took place over a very long period of time. From that point of view, it is
“unlikely that human speech suddenly burst on the scene at some point in our
evolution. The ability to manipulate vocal sounds appears to have been in proc-
ess of continual development for over 1 million years” (1997: 252). He agrees
with Chomsky when he notes that the possibility that a Chomskyan model of
language could not yield an evolutionary story other than by happy accident,
and then argues persuasively against him when he writes: “The most universal
attributes of language structure are by their nature the most variable in surface
representation, variably mapped to processing tasks, and poorly localizable
within the brain between individuals or even within individuals. Therefore,
they are the least likely features of language to have evolved specific neural
supports” (1997: 333). He denies that innate rules, innate general principles
and innate symbolic categories can be built in by evolution.
Instead, Deacon proposes that the ability of humans to flexibly learn what-
ever language they are exposed to as children derives from the development of
a prefrontal cortex ability that is present in all mammal brains and particularly
well developed in primates. This is the ability to maintain attention on some-
thing in short-term memory in order to do something opposite or something
complementary, e.g., foraging for a resource like fruit or operating in a social
context where learning complicated conditional associations is necessary.
A pronounced predisposition in humans toward this dual attention is at the
basis of the infant’s ability to learn a particular language, which means that a
human infant is biased not to rely on dealing with the world through imme-
diate experience but rather through the patterns of associations organized by
the symbols provided by the language it is born into. Other mammals, includ-
ing primates, do very well interacting with the world on an immediate basis,
according to their own experiences, making co-relational (indexical) learning

common. A human infant, by contrast, easily “forgets” its own, immediate
experiences with objects and events and instead orients (my term) its associ-
ations in and through the orienting behaviors of the languagers who consti-
tute its social world. Deacon believes that, over time, the “indirect mode” of
symbol use itself was the prime mover for the strong prefrontalization of the
brain in hominids.
Although Barrett, Dunbar and Lycett cite Deacon with approval, Deacon
does not quite advocate the idea that human language emerged from primate
vocalizations, such as vervet monkey alarm calls. Some of Deacon’s most
interesting passages concern the comparison of the muscular-skeletal sys-
tems characteristic of human vocalizations and the visceral systems activated
in non-human primate vocalizations. The location where the two systems –
the visceral and the muscular-skeletal – intersect in humans is the larynx.
Speech is a great disruption of the respiratory system, but it is one that causes
humans few problems. Humans have a great deal of cortical control over
laryngeal muscles and the muscle movements of the tongue, lips and jaw
that need to be coordinated with breathing in order to speak. Other char-
acteristic disruptions of breathing include laughing and sobbing. However,
these involve rhythmic sound production and very stereotypic breathing
patterns that segregate the sound into repeating units – laughs vocalized on
the outbreath and sobs vocalized on the in-breath. “Both are associated with
relatively stable mouth-tongue positions and no articulation of the lips, or of
the tongue against other oral structures. The fixed stereotypic positioning of
the mouth and tongue during these human calls often makes it impossible
to speak at the same time.” In laughing and sobbing, neurologically speak-
ing, the visceral motor system is foregrounded against a stable facial/skel-
etal background. Speech inverts the relationship in that the skeletal motor
features of the vocal tract take the foreground against a comparatively more
stable vocal/respiratory background (1997: 246).
Deacon notes that human laughter and sobbing are like primate calls which
are bottom-up viscerally driven patterns, based on alternation of the pres-
ence and absence of vocal sounds and superimposed on relatively more stable
mouth postures. In speech, the situation is the reverse: “relatively slower tonal
changes and exhalation patterns become precisely timed and subordinated to
match the rapidly fluctuating articulatory movements of the mouth and tongue.
Speech is thus comparatively top-down control. The skeletal motor systems
driving mouth and tongue movements dictate breathing and laryngeal patterns”
(1997: 250–51).
What is at issue here, then, is the shift from subconscious, automatic, sub-
cortical, i.e., limbic, arousal to (semi-)conscious, voluntary, intentional, cor-
tical control over vocalizations. In other words, it seems that Deacon would
agree with all those who say that whatever we want to identify today as human
language is sufficiently different from primate calls in terms of neurological
instantiation and social functioning and should be classed in a different lineage
of phenomena. The point is that it is the observer who determines where the
branching points in our evolutionary stories occur, based on what the obser-
ver identifies as a significant similarity or dissimilarity in the development of
whatever organism, trait or behavior she is tracing.
Deacon’s story about the pressures that brought the symbolic system into
being concern the introduction of a novel food source, namely meat, into the
social group and the sex-role changes it brought about in terms of hunting/
scavenging and male provisioning of mates and offspring. The long and short
of Deacon’s argument is that:
without symbols that refer publicly and unambiguously to certain abstract social rela-
tionships and their future extension, including reciprocal obligations and prohibitions,
hominids could not have taken advantage of the critical resource available to habitual
hunters [meat] … Symbolic culture was a response to a reproductive problem that only
symbols could solve: the imperative of representing a social contract. (1997: 401)
This is Deacon’s hypothesis for the social contract. It is bound up with the shift
from polygyny to mostly pair bonding in the hominid species, change in canine
teeth, stone tool technology and a general pattern of social-sexual organization
that is rare in the animal world.
Those of us who have read far into evolutionary theory are aware of what
are called just-so stories and the ease with which one can invent plausible sce-
narios. Deacon is as aware of the problem as anyone. I do not actually have
a problem with his story, except to agree with the evolutionary psychologists
that the introduction of the novel food source seems more like Step 2 than Step
1 in getting language/ing off the ground. In other words, the social contract
theory sounds more like an effect than a cause. However, to me, even so-called
grooming-at-a-distance seems like more of an effect than a cause.
As may be evident, many of the current attempts to explain an origin-
ating evolutionary scenario have long histories. Those researchers who do
not propose scenarios as such suggest instead associations or correlations
between the development of certain behaviors in early humans and the emer-
gence of language. This list of associations includes: the shift from arboreal
to ground-dwelling life that required cooperation for defense, hunting, and/
or the gathering of fruits and vegetables; the formation of coalitions or alli-
ances for deceptive purposes; bipedalism in general; canine tooth reduction;
tool use; more accurate use of weapons (ability to aim). Bipedal locomotion,
for instance, plays a critical and perhaps surprising role in vocal evolution,
in that bipedalism frees the thorax of its support role during quadrupedal
locomotion, and this freedom permits the uncoupling of breathing and strid-
ing necessary for the subsequent selection for vocal virtuosity and speech. It
also plays a role in the expansion of the brain, for as humans lost body hair
and moved to life in the savannahs of East Africa, they moved more in the hot
sun and evolved more sweat glands. More sweat glands allowed the brain to
cool and grow, with the result that humans are now the sweatiest of all mam-
mals and have one of the largest brain-to-body ratios. In addition, the earliest
fossil records of bipedalism in the hominid line are always accompanied by
the reduction of the projecting canine teeth. In certain primates decreased
tooth use is associated with increased stick use for finding food. That is to
say that a stick used for food foraging could double as a defensive weapon
as well, one even more effective than canine teeth. So, it is possible that “the
need for carrying such a dual-purpose tool may well have been the selective
force that led to the development of habitual bipedalism” (Brace 2004: 507).
Lieberman associates syntactic abilities with our ability to put one foot in
front of the next (2006). MacNeilage associates speech production with the
mandibular oscillations of ingestion (2008).
There is no one originating scenario at one particular moment that will
account for the current range and flexibility of human languaging. We cannot
unwind from current practices to imagine accurately a particular condition that
permitted sound-sequences to be emitted for whatever reasons and for sound-
sequences to be interpreted in whatever ways the interpreters found appropriate.
However, there is no reason for us to confine ourselves to imagining a particu-
lar condition, but rather we can see the emergence of languaging in the species
as an interlocked set of circumstances that involve the human body from the
soles of the feet (bipedalism) to the top of the head (the prefrontal cortex) and
many places in between (the larynx, the lungs, the sweat glands, the hands)
and in continuous interactions with fellow humans: from the moment of birth
in the way that babies are born (the birth canal–baby head ratio is far narrower
in humans than any other primate’s, thus suggesting that humans have required
assisted births for several hundred thousand years), to the way babies are held
and interacted with, to the way they are conceived (face-to-face sexuality,
along with a shift in the female from estral cycles to non-seasonal sexuality).
The particulars of the human way of living that have been conserved over many
hundreds of thousands of years have necessarily produced and reflect human
particulars in the cognitive domain, with the result that any human living being
born and raised in the human world will be (more rather than less) ready to
become a languaging living being. In addition to inheriting many primate-wide
cognitive skills such as pattern recognition, categorization and social relation-
ships, they also come ready to be oriented. Tomasello (1999, 2003) has called
this the capacity to understand conspecifics as intentional/mental agents such
as the self, and he argues that this biological adaptation is the only one neces-
sary to account for all the species-unique aspects of human cognition. The
mirror neurons seem to have a part to play in this understanding.
We turn next to look at the ways that living beings become languaging living
beings in ontogeny.
6 The ontogenic script begins
The previous two chapters were devoted to comparing and contrasting alterna-
tive evolutionary accounts of language and then producing a developmental-
systems-theory-inspired evolutionary script for language/ing. The next two
chapters will be devoted to producing a developmental-systems-theory-inspired
ontogenic script for language/ing. Building upon Oyama’s idea that evolution
is a succession of developmental systems, the chapter opens with a review of
some aspects of the human developmental system that might reflect our gen-
eral primate heritage. The chapter continues with a look at what happens when
the developmental system produces an abnormal result, as is the case with
Williams’ syndrome. The chapter ends with consideration of what cognitive
abilities we wish to attribute to an infant, which topic necessarily engages us in
the debate about the relationship between thought and language. It is the goal
of the chapter to exemplify a developmental-systems-theory approach to that
age-old debate.
Getting off the ground and into the languaging loop

Living beings who are born into this world with a high probability of becoming
languaging living beings belong to a lineage of organisms grouped according
to the species sapiens, the genus homo, the family hominidae, the order pri-
mate, the class mammalia and so forth. The order primate usually garners the
most attention from those interested in understanding humans and our place in
the world, and discussions tend to revolve around identifying the traits, abil-
ities, behaviors and tendencies that humans do and do not share with our pri-
mate cousins. We are currently in a research moment at a far remove from the
mid nineteenth century when Max Müller could so confidently declare that
“Language is our Rubicon, and no brute will dare to cross it.” New studies
seem to be coming out daily to suggest that some previously thought uniquely
human ability can either be observed in other primates or taught to them.
The behavioral economist Keith Chen at the Yale-New Haven Hospital
laboratory has trained capuchin monkeys to use money in the form of actual
coins. He discovered that when they were taught to use the coins, the group of
165
seven capuchins “responded quite rationally to simple incentives; responded

irrationally to risky gambles; failed to save; stole when they could; used money
for food, and on occasion, sex” (Dubner and Levitt 2005: 32). Returning to the
subject of emotion, addressed more fully at the end of Chapter 2, I here remind
readers that Paul Griffiths refers to a 1988 study suggesting that humans have
evolved a sense of fairness that causes them to bear costs rather than accept
grossly unfair divisions of the economic product of a transaction (1997: 157).
However, not more than two years ago primatologists Sarah Brosnan and Frans
de Waal at the Yerkes National Primate Research Center at Emory University
performed a series of experiments that led them to conclude that a sensitiv-
ity to fairness is not exclusive to humans and may have emerged early in the
primate lineage. Using capuchins once again, Brosnan gave monkeys pebbles
that they were supposed to return to her in exchange for a slice of cucumber.
They were satisfied to trade the pebble for a cucumber until they observed a
neighbor performing the same task for a grape, a more highly valued treat. The
monkeys subsequently refused the cucumber, even though the alternative was
no reward at all. They preferred to receive nothing rather than a lesser treat
when their fellow was being given unfair preferential treatment (Brosnan and
de Waal 2003: 297–99). By the way, in Chen’s experiments, the monkey who
was paid a coin for sex turned around and traded it in for a grape, evidently the
true coin of the realm.
Recently chimpanzees in the West African savannah have been observed
making spears from sticks and using them to kill small mammals, as reported
by Jill D. Pruetz of Iowa State University and Paco Bertolani of the University
of Cambridge. Until this observation, it was believed that only people crafted
tools expressly to kill prey. The very latest uniquely human trait to be called
into question is bipedalism. Primatologist Robin Crompton of the University
of Liverpool has observed orangutans walking upright in trees on thin, flexible
branches. What is of interest is that their knees are straight, unlike those of
chimpanzees, who are sometimes upright during foraging but with bent knees,
and that the surface the orangutans are walking on is springy, much like spongy
ground would be. The observation led Crompton to a comparison with human
runners, who tend to keep their knees and hip joints extended when moving
on springy indoor athletics tracks. It likewise dovetails with a thirty-year-old
hypothesis that upright walking first evolved in the trees, but it upsets the dec-
ades-old assumption that walking upright is an innovation of human locomo-
tion (O’Higgins and Elton 2007).
It should not be surprising, then, that much of human cognition, learning
strategies and behaviors would be grounded in our primate heritage and that
if we actively seek similarities between human ways of being in the world
and those of our fellow primates, those similarities will be found. Take, for
instance, the well-known semantic congruity effect. When asked to compare
Getting off the ground & into the languaging loop 167
two large animals, such as a cow and an elephant, adult humans are much
quicker to respond when the question is “Which is larger?” rather than “Which
is smaller?” When asked to compare two small animals, such as an ant and a
rat, they are much quicker to respond when the question is “Which is smaller?”
rather than “Which is larger?” Recently it has been shown that there is a simi-
lar semantic congruity effect that appears when monkeys are asked to make
numerical judgments. It thus seems to be that this effect is a consequence of
the comparison process rather than stimulus encoding, that is, in the latter case,
whether or not the being in question possesses symbols to represent numbers
precisely (Cantlon and Brannon 2005, 2006). In humans, the semantic congru-
ity effect appears when we language but it is not caused by our languaging.
One of the goals of a developmental systems linguistics is to tease out the
nature of this hybrid cognition.
Now, the perception of similarities between humans and non-human pri-
mates works both ways. While sorting pictures of humans from those of other
animals, language-trained apes will put their own portrait on the human pile.
If they do not actually come to think of themselves as humans, de Waal has
observed, they at least obviously sympathize with the humans that surround
them: they want to fit in, be like them (2001: 223–24).
And yet no amount of enculturation – raising a non-human primate like a
human child – will turn a chimpanzee or an ape into a human. What we can
say is that the enculturated ape is epigenetically human, but it remains genetic-
ally ape. And conversely: a human infant raised by wolves will be genetically
human and epigenetically wolf. The questions then are: “what is characteristic-
ally human about being human?” or “what is characteristically ape about being
ape?” I am not sufficiently informed or qualified to advance an idea about what
makes an ape an ape. However, I do say and have suggested at various points
throughout the present study and in different ways that what makes a human
a human is not possessing language as such but rather being the inheritor of
the wide-ranging effects of belonging to a lineage that has been engaging in
languaging behavior for several hundred thousand years, with another million
or two years of languaging-like behavior before that.
The Canadian psychologist Merlin Donald, in his interesting and influen-
tial Origins of the Modern Mind, takes the position that cognitive evolution
had to precede the emergence of language, that is, language is a reflection of a
speaker’s intelligence. Given this, it follows that he would argue that the gulf
he perceives between apes and humans cannot be bridged by language alone –
or, rather, his particular conception of it. It is clear that apes cannot carry on
a conversation or read or master other symbolic tasks such as mathematics,
but the cognitive differences, Donald writes, go well beyond activities that are
specifically linguistic in nature. Apes are not adept at human athletic feats or
human play and games which, for Merlin, count as non-linguistic in that “most
informal games have little or no verbal component and deaf-mute children have
no difficulty mastering them. Yet these games are already beyond the reach of
apes” (1991: 120–21). To support Donald’s point, Pinker and Jackendoff state
that the rhythmic properties of language and music may well be unique to
humans, since “informal observations suggest that no other primate can easily
be trained to move to an auditory beat, as in marching, dancing, tapping the
feet, or clapping the hands. This is surely one of the most elementary character-
istics of the human rhythmic response, and one that is displayed spontaneously
by young children” (2005: 211). Once again, we turn to de Waal, who quotes
the Gestalt psychologist and primatologist Wolfgang Köhler, who offered, in
turn, this 1925 account of what looked to be a chimpanzee game:
One after another, the rest of the group would approach … and finally they march in
orderly fashion in single file … The character of their movements changes; they no
longer walk, they trot, and as a rule with special emphasis on one foot, while the other
steps lightly; thus a rough approximate rhythm develops, and they tend to “keep time”
with one another. They wag their heads in time to the steps of their “dance” and appear
full of eager enjoyment of their primitive game. (2001: 226–27)
Given this description, the question whether the chimps can march, dance and
move to a rhythmic beat is, as de Waal might say, the same as asking whether
chickens can fly.
Whether or not apes can or cannot be trained to move to a (human) audi-
tory beat (since they seem to march to their own), the point to be made here is
one that has been made before in Linguistics and Evolution: deaf-mute chil-
dren who have no difficulty mastering informal or even, for that matter, formal
games belong to the lineage of languaging living beings. Humans who are
born deaf (and not exposed to sign language) belong to a lineage whose cogni-
tion and behaviors have been significantly shaped by the sense of hearing and
by the activity of languaging, just as humans who are born blind belong to a
lineage whose cognition and behaviors have been significantly shaped by the
sense of sight. We would not have the body morphology and locomotion we do
if we were not visual creatures; congenitally blind humans who have otherwise
normal ontogenies walk upright and forward. Individuals blocked from access-
ing the modality of hearing for whatever reason in ontogeny are not blocked
from living the effects of the modality of hearing that have shaped the lineage
in phylogeny.
Does this mean that everything that is characteristically human about humans
is a result of our languaging behaviors? The short answer is Yes.
The longer answer requires a more generally phrased discussion of the pos-
sible range of instantiating events and the ramifications of the development of
the ability of humans to orient fellow humans in their cognitive domains by
means of coordinations of coordinations. This takes us back to the evolutionary
scenarios played out at the end of the last chapter. Did the instantiating events
take place in pooh-pooh-type grooming activities, in rhythmic activities aris-
ing in the course of game playing, or in yo-he-ho-type working activities such
as lifting heavy objects and food foraging? The next question is: are these
mutually exclusive types of activities? In imagining origins, it is not necessary
to require a distinction between hygiene (grooming), play (games) and work
(food foraging) as might seem reasonable to us now. Nor is it necessary to
imagine that marching, dancing, foot stomping and hand clapping must have
been confined to particular types of social interactions 100,000 or 200,000+
years ago, although these activities now have quite particular histories and
social settings where they are more and less appropriate.
As long as the activities and the behaviors of those involved in those activ-
ities led to the result that all involved in those activities had their attention
oriented similarly enough in a domain of interactions (vocal/gestural) that were
independent of the orienting interactions themselves, the domain of languaging
is able to come into existence. The rhythmic responses so common to humans
and the prosodic features found in the world’s languages could and, in my
description of things, do share a common history. They are homologues. The
ability of humans to coordinate all parts of their bodies – from the feet, to the
hands, to the larynx, to the tongue, to the visual and auditory domains as well
as to much of cognition, then back down through the body loop – with their
fellow humans through extra loops specified in the new phenomenal domain
of coordinations of coordinations is what languaging is in its widest sense. The
fact that languaging is so woven into the human body is reflected in the way
that it is woven into brain structures, particularly subcortical brain structures
such as the basal ganglia that “are involved in regulating walking, manual ges-
tures, speaking, syntax, thought processes as well as regulating mood and per-
sonality” (Lieberman 2005: 233). The story of all human bodily dispositions
and behaviors do not reduce to the story of languaging, but very little of what
humans are now can be separated from it.
We see the phylogenic effects of these bodily coordinations of coordina-
tions in the ontogeny of the human infant who comes into the world ready and
able to be coordinated in such a fashion or, to put it another way, defenseless
against being so coordinated. A number of non-human primates, particularly
chimpanzees, can be brought, to a certain extent, into the domain of human
languaging, but they do not come as softened up to do so as do human infants
for two reasons. First, the predominance of cortical projections to the visceral
motor systems in humans produces an oddity of human vocalization: babbling.
Almost all babies babble, even deaf babies (but there are cases of non-babbling
human babies who, not surprisingly, do not become proficient languagers).
Babbling in humans begins within a few months of birth, and no other mam-
mal species’ babies produce anything like the vocal play of human babies.
Unlike the calls of other mammals that occur when they are excited or upset,
human babies tend to babble only when they are calm. When they get upset,
we know what they do: they cry. “It is the first sign,” Deacon writes, “that
the human vocal motor output is at least in part under the control of the cor-
tical motor system. Even the timing of the first appearance and maturation
of babbling and vocal mimickry corresponds to the maturation of the cortical
motor output pathways.” This has to do with myelination, the laying down of
a fatty substance that acts as a sort of neural insulation, promoting fidelity of
the signal and more rapid signal transmission. By the time a child produces
its first words and its first steps, usually around age one, “the projections that
carry voluntary movement information to the brain stem and spinal cord have
nearly reached adult levels of myelin” (Deacon 1997: 251). The next step for
the infant, according to MacNeilage (2008), is to disrupt the preferred pat-
terns found in babbling and coordinate articulatory activity around the patterns
found in the ambient language.
Second, primate cognition is a particular kind of social cognition. The loop-
ing of languaging through the lineage called human has caused primate cogni-
tion to turn a corner. How to characterize this corner has sparked a debate that
de Waal has succinctly framed as “Do apes ape?” De Waal correctly takes the
mere out of “mere imitation” when he says that non-human primates can and
do imitate behaviors of conspecifics and non-conspecifics, e.g., their primat-
ologist trainers, and that this mimicking behavior requires complex process-
ing and production (2001: 216–32). The question is whether this imitation is
somehow the same – presumably in function and cognitive processing – as the
kind of imitation displayed by human infants’ social learning. Tomasello says
that it is not and then makes the distinction between emulation learning (what
non-human primates do) and imitation learning (what humans do), although he
may have more recently revised that position somewhat.
In The Cultural Origins of Human Cognition, Tomasello describes emulation
learning as focusing on the environmental events that occur. The non-human
primate is interested to attend to what happened and how it happened and can
reproduce the change of state on its own. Tomasello’s example is that of a non-
human primate mother turning over a log and finding insects underneath to eat
and the non-human primate child thereby learning that by turning over logs,
food can be found. In Tomasello’s assessment of the situation, it does not matter
how the log got turned over (the wind could have done it) in order for the non-
human primate child to learn about a new food site. This, he argues, is different
from the situation where the human child is disposed to so-called purely imita-
tive learning, namely following the procedure of the adult for the sake of follow-
ing the procedure of the adult, even if the procedure is not necessarily efficient
or easy. Tomasello’s point is that “imitative learning is not … a ‘higher’ or
‘more intelligent’ learning strategy than emulation learning; it is simply a more
social strategy” (1999: 30). One could say, then, that the human child comes
into the world open to imitating, in the sense of being disposed to being oriented
in one’s own cognitive domain as if one were in the cognitive domain of the
other. It is only through the developmental process that the human infant learns,
more or less, to differentiate her personal world from that of those around her.
Coming into languaging facilitates that process or, rather, builds on and hones a
process that was already part of the primate lineage, since some non-human pri-
mates can be demonstrated to have a concept of self separate from others and to
recognize others’ intentional states, although they do not, perhaps, discriminate
them as finely as do humans. The emulation/imitation distinction proposed by
Tomasello caused quite a stir and prompted teams of researchers to see whether
non-human apes could be seen to imitate. Recently, a team including Tomasello
came to the conclusion that some apes are able to understand others’ intentions
as rational choices of action plans and then to use that understanding in per-
forming the same tasks (Buttelmann et al. 2007).
As for humans, the infant gets into the languaging loop, according to
Tomasello following psychologist Jerome Bruner, at around nine to twelve
months by means of three new behaviors: (i) an appreciation of the joint atten-
tional frame; (ii) the understanding of communicative intentions; and (iii) cul-
tural learning that takes the form of role reversal imitation (Tomasello 2003:
21). These behaviors could just as easily be described as a breakdown of the
components of orientability. Tomasello is devoted to the idea that human cog-
nition is a particular version of primate cognition that results from “a small
but important twist” in the way humans have evolved in the social domain to
comprehend others as intentional agents (1999: 23–24). Tomasello’s twist cor-
responds to my loops.
But now, is it really and truly the case that everything characteristically
human about humans is a result of our languaging behaviors? What about dis-
tinctive aspects of human physiology, e.g., the fact that we have finer motor
skills, less body hair and, quite literally, thinner skin than other primates? What
do these features have to do with our languaging? Well, fine motor skills in the
fingers and hands could be viewed as developments homologous to the refine-
ment of the control of the musculature in the lower half of the face related to
changes in the FOXP2 gene. These refinements may even be related to the feet
in the so-called syntax of walking. To push the point even further, I have no
difficulty imagining that our skin has thinned as our languaging has developed.
What would make me think this? Let us return to Frans de Waal, in particu-
lar the first dream he had about chimpanzees when he was starting out at the
Arnhem Zoo in the Netherlands. De Waal recounts that, in the dream:
my preoccupation with the distance between them and me was apparent. During this
dream the large door to their quarters was opened for me from the inside. The apes were
pushing each other aside in order to get a good look at me. Yeroen, the oldest male,
stepped forward and shook my hand. Rather impatiently he listened to my request to

come in. He refused point blank. That was out of the question, he said, and besides, their
society would not suit me: it was much too harsh for a human being. (1982: 54)
I had the opportunity to ask de Waal about this dream and why he thought
ape society would be too harsh for a human being. He gave me the answer I
had anticipated: “Because we have language, and we tend to work things out
through language” (personal communication, February 2005). But, still, what
does this have to do with skin? The answer is that in mammalian development,
the ectoderm exists as a single layer of cells covering the embryo surface. As
the embryo grows, stem cells differentiate into either epithelial skin cells or
neurons, thereby creating a seamless developmental relationship between the
inside of the organism and its outside. This is yet another dimension of the
unbroken coincidence of an organism’s being, knowing and doing, where the
ongoing sensory contact of the organism with its outside world – conspecifics
first and foremost for the primate lineage – directly affects its neural organiza-
tion, and vice versa.
In the ape lineage, nerve endings end up many layers below the surface of
skin. Ape society is not too harsh for apes. The way they communicate with one
another, bodily, gesturally and vocally, corresponds to the level of intensity that
is appropriate for them. In the human lineage, nerve endings end up very close
to the surface of the skin. If our skins were any thicker, we would literally not
be in touch with the (to us) far less harsh conditions of human society. These
conditions would include, from an ape point of view, the impossibly nuanced
bodily postures, bodily gestures, facial expressions, vocal gestures and vocal
modulations that characterize our social interactions. The softening up of the
human infant, rendering it vulnerable to and defenseless against the orienting
behaviors of the social world into which it is born, along with the thinning of
our skin to receive all that we need to receive through skin-to-world and skin-
to-skin contact does not, of course, mean that we do not have the potential to be
a dangerous and deadly species. It also does not mean that every living being
who belongs to the lineage of languaging living beings develops into what is
recognized to fall in the norm of languaging living beings.
When things go wrong

Oyama has pointed out that the relative reliability of the entire developmental
system has led some researchers to imagine that it is assured by a genetic code
or blueprint. It is comforting, or perhaps only customary, to think that there
must be something deep inside – call it Eternal Form – that secures the produc-
tion of good outcomes time after time, something that buffers the production
against the vagaries of an unpredictable world. When we abandon the inter-
twined notions that there is any so-called information “in” the genes and that
When things go wrong 173
the world is wholly unpredictable, we are able to see an entire developmental

system unfold in all its regularity, and we are better able to understand not only
how things go right but also how and where they go wrong. The exquisite intri-
cacies of the developmental system – the whole of the interaction of genetic,
biological and social factors in play both before and after birth – allows for the
modulation of human behavior with great subtlety and nuance. This complex-
ity also allows dysfunction to occur at a large number of sites and, at times, to
exert effects in areas at a far remove, both physically and temporally, from the
source of the original insult.
In the case of an activity as wide ranging as languaging, problems can and
do occur at various moments and places in the developmental process, begin-
ning with problems at the genetic level. The problems understood here are not
those that arise from the sensory deprivation of blindness or deafness that may
affect and interfere with the living being’s coming into languaging,1 but rather
problems such as Williams’ syndrome (WS), dyslexia, schizophrenia and aut-
ism. Certainly, a complete understanding of the causes and pathways of these
problems could lead to improved therapies for the sufferers. However, they are
of interest in the context of Linguistics and Evolution because they not only
shed a refracted light on how we want to understand the biosocial organization
of our subject matter but also, in some cases, provide further clues to the evo-
lutionary story that we want to tell about it.
Karmiloff-Smith’s study of WS provides arguments against the non-devel-
opmental modularist and nativist view of language. Her neuroconstructivist
approach “highlights how tiny variations in the initial state could give rise
to domain-specific differences in endstates” (1998: 390). WS children, for
instance, exhibit particular proficiencies in face processing and language, but
the way they process them is different from normal controls, and this is because
their brains are different, in that:
• the WS brain is 80 percent of normal volume;
• the total cerebral gray matter is significantly reduced;
• there is abnormal layering, orientation, clustering and size of neurons;
• the anterior regions are smaller than in normal controls but larger than in
Down syndrome brains;
• the dorsal hemispheres show cortical malformation;
• the cerebrum is particularly small;
1
Because languaging is not solely an auditory phenomenon, it is not to be assumed that blind chil-
dren come into languaging in the same way as sighted children. For instance, it has been found
that blind children use significantly fewer auxiliaries like can and do and will because “mothers
of blind children tend to ask far fewer questions, like ‘can you give me that?’ or ‘do you want
some milk?’, and to use far more directives, like ‘that this’ or ‘come here.’ The blind child [is]
less exposed to auxiliary verbs in daily dialogue and … takes longer to learn them” (Karmiloff
and Karmiloff-Smith 2001: 188).
• the limbic structures of the temporal lobe are small but proportionally similar
to normal control;
• the frontal cortex displays a near normal proportional relation with the pos-
terior cortex, although both are reduced in size. (1998: 393)
It is not surprising, then, to learn that claims of WS proficiency in language
are not standing up to scrutiny. For instance, the processing of some aspects
of complex grammar is beyond the ability of WS individuals, and both chil-
dren and adults with WS tend to perform poorly on tests of structures with
embedded sentences that normal controls as young as six are able to handle
(Karmiloff and Karmiloff-Smith 2001: 203). But the case remains that even in
tests where WS subjects achieve results that fall in normal ranges, this parity
does not mean that their processing pathways are the same as those of normal
language learners. There is evidence, even for face recognition, that when the
faces are inverted, WS individuals do not perform as well as normal controls,
thereby providing evidence against something like a face-recognition module.
As Karmiloff and Karmiloff-Smith point out:
From the moment of conception and throughout embryogenesis and postnatal brain
development, the brains of infants with WS develop differently from those of other indi-
viduals. The deletions on chromosome 7 are missing from every cell of the developing
WS embryo. So right from the outset, the WS brain is growing atypically. (2001: 201)
The point is that the WS brain is not just a typical brain with some parts affili-
ated with language learning affected and some not affected. The whole brain is
different. Therefore the processing will be different.
Another syndrome that has been much studied for its interesting developmen-
tal dynamics is developmental dyslexia, which is distinguished from acquired
dyslexia that results from brain injury. The manifestation of developmental
dyslexia crucially depends on the type of writing system or systems in use in
the languaging living being’s environment and the prevalence of orthographic
regularities or irregularities found in that writing system. Because dyslexia – a
reading difficulty found with written systems based on an alphabet – seems to
involve deficiencies in phonemic processing, a greater incidence of developmen-
tal dyslexia in English-speaking and French-speaking countries is reported than
in, say, Spanish-, Italian- or German-speaking countries, because the orthog-
raphies of the last three languages are far more regular, i.e., typically with one
graph to one sound, than are English and French, which are notoriously irregu-
lar. One study estimates that between ten to thirty percent of the population of
the United States is affected by developmental dyslexia (Kandel et al. 1995:
646–47). There is, furthermore, a correlation between dyslexia and handedness,
in that a disproportionate percentage of dyslexics are left-handers, suggesting
that the disorder might involve a deficit in the development of dominance by the
left hemisphere. Abnormalities in the cell layers of certain regions of the left
hemisphere have also been discovered in the cases of dyslexics who have been
studied anatomically, and these observations suggest that “migration of neurons
to the left cortex during development is slowed in dyslexic patients, a defect
that might be attributable to injury of the fetus” (Kandel et al. 1995: 647). The
particular area of interest for alphabet-based written systems is the left temporo-
parietal cortex, which shows hypoactivation in both children and adults with
dyslexia, when compared with typically developing children and adult readers
when performing tasks that demand phonological awareness (Gabrieli 2009:
281). By way of contrast, for Chinese, which is logographic and not alphabetic,
learners with reading disorders show atypical activation in the left prefrontal
cortex and not in the left temporo-parietal regions (Gabrieli 2009: 281).
A long line of thinking exists that correlates handedness, brain lateralization
and the emergence of language in the species, although it could be that footed-
ness is more directly pertinent. In any case, a new dimension has been added to
this mix by the British psychiatrist Timothy Crow’s work on schizophrenia as
it relates to abnormalities in the ways the two hemispheres of the brain relate
to one another (1997, 2005). He notes, as many other brain researchers have
noted, that what distinguishes the human brain from the brains of other spe-
cies is not its size or basic structure but rather the way the parts of the cerebral
cortex are connected. This connectivity both supports and presumably resulted
from the development of languaging (my term, not his) in the species.
Crow focuses specifically on the lateralization of functions and locates the
site of genetic interest corresponding to this lateralization on the sex chromo-
somes. Individuals with sex chromosome abnormalities – Turner’s syndrome
individuals who lack one chromosome (XO) and individuals who have an extra
X chromosome (XXY in Klinefelter’s syndrome and XXX syndrome) – have
characteristic spatial or verbal deficits in development that can be attributed to
the non-dominant (typically right) or dominant (typically left) hemispheres,
respectively. Individuals with an extra Y chromosome (XYY syndrome) also
have delays in verbal development that are similar to those seen in the extra X
syndromes. The case can then be made for a genetic component of asymmetry
that is located on the X chromosome with a gene present in homologous form
on the Y chromosome, because males who have only one X chromosome do
not have deficits comparable to those seen in Turner’s syndrome. Crow links
the genetic stabilization of hemispheric asymmetry, i.e., lateralization, to the
evolutionary period when what would probably be recognized today as human
language could be said to have emerged – roughly the time period that has been
asserted throughout the present book, i.e., 100,000 to 250,000 years ago.
Crow also links the particular way human cerebral asymmetry functions,
called torque, to the syndrome of schizophrenia. Torque marks a difference
between the functioning of the non-human primate brain, basically a two-
chambered organ, to that of the human brain, which is more properly viewed
as consisting of four chambers: left and right anterior-motor and left and
right posterior-sensory. Torque refers to the way that the distribution of tis-
sues to the two hemispheres is subtly different along the antero-posterior axis
and to the bias that is thereby imposed on inter-hemispheric connections. In
normal human brain functioning, there is a right-to-left connectivity in the
anterior (motor) and a left-to-right connectivity in the posterior (sensory)
parts of the brain.
Throughout the present book, it has been suggested that both cortical hemi-
spheres and many subcortical structures subserve the functioning of languag-
ing. Now we add the effect of their connectivity. We know that asymmetric
hemispheric connections were introduced at some point in hominid evolution.
Crow hypothesizes that a further shift in the connections, a torque, occurred
along with the development of languaging. For the record, Crow is arguing for
a macro-mutation (Pinker-like) speciation event whereby language evolved all
at once, whereas the perspective taken in Linguistics and Evolution is that the
genetic changes pulled along by ontogenic phenotypes engaging in long-stand-
ing behaviors help to stabilize those behaviors. The deep weaving of languaging
into the brain – even into the thalamus, which is involved in focusing attention
and may provide a site of the interaction between languaging and memory
systems – argues against a macro-mutation speciation event. In any case, the
nuclear symptoms of schizophrenia – those defined as constant and character-
istic of the disorder – are, indeed, language/languaging-oriented and include
the inability to properly distinguish between internally generated thoughts and
sound patterns from those of an external source (hallucinations) along with
the breakdown of the indexical function of language (pronoun usage). Thus a
strong connection exists between the disorder and certain aspects of language
production and perception.
Furthermore, Crow directs attention to chromosome 21, particularly to the
Xq21.3/Yp11.2 homology block that was established by an X to Y duplication
six million years ago, i.e., at the time of origin of the hominid lineage. Within
this block a gene pair has been subject to change in the hominid lineage, with
deletions on the Y-segment. These Y-segment deletions are interesting because
this particular deleted segment belongs to a group of cell-surface adhesion
molecules that govern cell-to-cell interactions and are axonal guidance fac-
tors – both functions being key to development of proper asymmetry in the
brain. Crow speculates that the X to Y duplication and its subsequent modifica-
tion (deletions) on the Y may have played a central role in hominid speciation,
with the most recent change at around 160,000 years representing the transi-
tion to language and modern homo sapiens – what Crow called the ‘big bang’.
He argues that this mechanism generates epigenetic diversity relating to the
species’ capacity for language and proposes it as the basis of the genetic pre-
disposition to psychosis (2008).
Three pertinent points to make here are that: (i) Crow is proposing the basis of
a genetic predisposition to psychosis; and (ii) other studies suggest that the over-
whelming evidence argues against the possibility that most cases of schizophre-
nia are caused by a common gene but rather that psychoses are best accounted
for by polygenic models; and (iii) schizophrenia might best be thought of as a
neurodevelopmental disorder, where a person’s liability to develop the disorder
is abetted by psychosocial causes as wide ranging as suffering complications
in utero, growing up in environments where caregivers show elevated levels of
‘communication deviance,’ and misusing amphetamines. From a DST perspec-
tive, it is significant to note that certain brain abnormalities in schizophrenics,
who may exhibit their first symptoms only in their teenage years, are consist-
ent with second- and third-trimester pregnancy abnormalities. The point here
is that, for creatures with complicated nervous systems and complicated social
environments, problems that manifest themselves somewhere along the devel-
opmental path might have quite distant causes.
Another kind of developmental problem involving neural connectivity – or,
rather, lack of connectivity – between and among various parts of the brain
manifests itself in the condition of autism. Recent imaging studies have shown
that the autistic brain, in addition to abnormalities in the volume and the dis-
tribution of the white matter that insulates neuronal signals, exhibits a lack
of cooperation between different brain areas. Autism is most often associated
with deficits that arise in social interactions, but serious problems in reasoning
tasks that require integrating different types of information are now getting
the attention of autism researchers. It has long been noted that autistic people
tend to obsess over details but find broader meanings elusive; they might mem-
orize facts easily but fail to grasp complex concepts. With the new imaging
studies, a hypothesis is forming that the prefrontal cortex, which is devoted to
complex processing, may have overly developed internal communications but
poor connectivity with more distant brain regions in autistic people, especially
connections between hemispheres – and, thus, the lack of the autistic person’s
ability to integrate words, facts or visual details in order to see what is usually
called the big picture.
Of particular interest to us is the possibility that part of the communica-
tion failure could be with the so-called mirror neurons discussed in the previ-
ous chapter. In the autistic brain it appears that the mirror-neuron system is
impaired. If this is so, it would explain the failure to imitate spoken words
and might account for some of the language delays in autism. In the terms of
a developmental linguistics, autism is a severe inability to be oriented in one’s
own cognitive domain through the conventional symbols in use in a particular
community, and it is likely that the autistic child at age twelve months does
not exhibit the joint attentional awareness that Tomasello says emerges at the
end of the first year. In development, timing is all. Studies coming from the lab
of the neuroscientist Eric Courchesne suggest that most of the abnormal brain
growth in autism occurs from birth to age three (Wickelgren 2005: 1858), and
this early abnormality produces long-ranging problems in negotiating social
interactions and social meanings generally. By way of contrast, schizophrenia,
which also is characterized by an abnormality in brain connectivity, usually has
an onset post-puberty and produces quite different symptoms and effects.
Given the full-bodied approach to languaging undertaken in Linguistics
and Evolution, the topic of autism and gesture is of interest, and the exact
manner and extent to which those with autism gesture is currently the topic
of considerable debate. The typical view has been that those with autism can
understand all types of gesture, although understanding abstract gesture is a
good deal more problematic. However, autistic individuals seem to use instru-
mental rather than expressive gestures (instrumental gestures are defined as
more concrete, such as “goodbye” or “go away,” and expressive gestures are
defined as more abstract, such as the thumbs-up or blowing someone a kiss
(Attwood et al. 1988)). Questions have now arisen concerning the instrumen-
tal–expressive distinction, as well as how to understand the very type of ges-
ture that defines autism, the “stereotype” – the arm flapping, hand waving,
finger tapping. Once seen as pathological, these gestures now may be seen
to have interactional and communicative undertones or even be seen as self-
expressive or self-comforting (Dickerson et al. 2007). Most important is the
possibility that, for those with autism, the addition of gesture makes speech
more difficult to understand. One study tracked patterns of eye movements in
autistic children who listened to descriptions of different shapes, e.g., a tri-
angle, and found that their eyes would track more slowly when iconic gestures
of those shapes were paired with the vocal description, the slow eye move-
ment indicating slower information processing (Silverman et al. 2007). The
opposite is the case for typically developing children. The point here is that
the language–gesture loop can be disrupted under certain non-typical condi-
tions (autism) and preserved under other non-typical conditions (the case of
IW mentioned in Chapter 1).
Ontogeny: the very beginnings

The first step in becoming a languaging living being is to have been born into
the lineage of languaging living beings. In order to continue with our story,
however, the biggest challenge is to bring together two disparate subdisciplines
that are known traditionally as language acquisition and discourse analysis. In
the subdiscipline of language acquisition, the researcher attempts to answer the
question (here radically oversimplified): how does the child acquire language?
This question seems to attract interdisciplinary attention from psycholinguists,
cognitive neuroscientists and neuroscientists. (Cultural anthropologists have
Ontogeny: the very beginnings 179
interest in this issue as well but would likely phrase the organizing question
differently, namely without the term acquisition and with a term more like
inculcation.) Most often, researchers who study language acquisition have
interpreted the how of the central question ontogenically and have allied their
accounts with theories of cognition and learning. More recently, some research-
ers in this subdiscipline have tried to incorporate the phylogenic axis, such that
the how includes an evolutionary script to support the ontogenic one.
In the subdiscipline of discourse analysis, the researcher attempts to answer
the question (here again, radically oversimplified): how is meaning made in
social and cultural contexts? Discourse analysts, as the very name suggests,
find alliances with researchers whose interests cover (already-established)
discourses. These define entire fields in the human and social sciences, the
ones that study social structures, social relations, social change and any and all
meaning-signaling systems known as semiotics.
In trying to bring these two subdisciplines together, a disjunction becomes
apparent: discourse analysts have tended not to stick their oars into either
phylogenic or ontogenic discussions of languaging, while those researchers
interested in phylogenic and ontogenic issues have typically not strayed far
from models of language/ing that focus on sentence-level analyses.
Bruner is the first theorist, it seems, who put “pragmatics into the middle of
things” (1983: 37) in understanding how the living being becomes a languag-
ing living being (my terms, not his). For a developmental systems linguis-
tics, an utterance cannot be analyzed out of the context of its use, and its use
must include the intention of the utterer along with the interpretation of that
intention by the addressee. Individuals may make requests by any number of
linguistic means, e.g., the utterance “It’s hot in here” can, depending on the
circumstances, function perfectly well as a request to open a window. Bruner
reminds us of a point that “should be so obvious as to need no comment,”
namely that the engine that drives the enterprise of learning a language is not
the acquisition of language per se, but rather the need to get on with whatever
activities are going on in the child’s social environment. However, unless one
keeps one’s eye on the pragmatics of language, Bruner notes, this point can
be easily overlooked (1983: 103). For Bruner, and for a developmental sys-
tems linguistics, coming into languaging is a consequence of being a living
being belonging to a certain lineage and born into a certain kind of community,
namely a languaging community; it is not an end in itself. Languaging is a
means by which we languaging living beings coordinate our actions. It is a
way, the way, to get our business – whatever it is – done.
By putting pragmatics into the middle of the living being coming into lan-
guaging, Bruner not only neatly brings together language acquisition studies
and discourse analysis, he also places the last level of analysis explored by
linguists in the twentieth century, namely the study of language in all of its
situatedness, into the first place of understanding the child’s coming into lan-
guaging. Bruner writes:
the learning of speech acts may be easier and less mysterious than the learning either
of syntax or semantics. For the child’s syntactic errors are rarely followed by corrective
feedback, and semantic feedback is often lax. But speech acts, on the contrary, get not
only immediate feedback but also correction. Not surprising, then, that prelinguistic
communicative acts precede lexico-grammatical speech in their appearance. Not sur-
prising, then, that such primitive “speech act” patterns may serve as a kind of matrix in
which lexico-grammatical achievements can be substituted for earlier gestural or vocal
procedures. (1983: 37–38)
Bruner’s move also gives us plenty of room to loop languaging as an orienting

behavior into the linguistic discourse. Living beings that belong to the lineage
of languaging living beings come into the world orientable and are treated as
such from day one by those around them. By Bruner’s account, it takes a year
of such treatment for the child to produce behaviors that to an observer are
understood to be demonstrating an understanding of others’ intentions. That
is, it takes a good year of experience with the cause–effect social world in
which the child is immersed along with his fellow living beings for the child
to develop into the kind of living being about which it can be said that “it takes
one to know one.” In knowing what one of us is up to, or in behaving as if one
knows such that one is treated as if one knows, the living being becomes a
languaging living being.
Bruner observes, for instance, that even very young children are sensitive
to the circumstances in which adults may use alternative linguistic means to
say whatever it is they have to say and that they use those circumstances
interpretively. For instance, they do not confuse a request for action (“Why
don’t you play with your ball now?”) with a request for information (“Why
are you playing with your ball now?”), although the two questions have the
same grammatical form. For Bruner, this means that what children learn when
they are learning a particular language is speech acts rather than grammatical
forms as such.
How far have language acquisition studies come in the twenty years since
Bruner wrote Child’s Talk? To hear Tomasello tell the story: not very far. In a
recent book, he writes that:
The reason performatives are important in the current context is that most theories of
language acquisition basically ignore them. But they are frequently used communica-
tive symbols, and they have a very similar intentional structure to expressions with a
more clearly referential component. (2003: 24)
To imagine that specialists in the early stages have been basically ignoring
performatives and expressives such as Hi or Thank you is surprising, to say the
least. It is also symptomatic of a linguistics that does not begin with the real-
ization that every utterance is some kind of speech act.
In addition to having been born into the lineage of languaging living beings,
the question “How does a living being become a languaging living being?”
entails answers to two other questions: (i) What is presupposed by the term
languaging living being? and (ii) What is the length of the developmental lin-
guistic system in a languaging living being’s life?
What is presupposed by the term languaging living being?

A developmental linguistics moves away from the rationalist sameness inform-
ing a model of the idealized speaker/listener and toward the awareness of lan-
guaging differences among individuals that have been richly investigated along
ethnic, economic, gender, generational and religious lines, among others.
Recent work in formalist linguistics has stepped away from what might be called
the strong essentialist position of early-mid-and-late Chomsky. Jackendoff,
for one, acknowledges that people do not have identical cognitive structures
associated with their linguistic communication and that there are reasons why
communication cannot be perfect. “For one thing,” he writes, “people are in
different states of knowledge: they have different histories and different goals.
In addition, one can never be absolutely certain, from the other’s behavior, that
one’s message has been understood the way one intends” (2002: 36). Still, he
retains what might be called a weak essentialist position, when he writes: “I
would imagine that differences among individuals’ brains are more or less like
differences among their faces: a bit more here, a bit less there, slightly differ-
ent placement here, slightly different proportions there, but everything works
essentially the same” (2002: 91–92). Here sameness and difference are in the
eye of the beholder. The real question is: “For what reason and/or to what pur-
pose are X and X¹ (or, possibly, Y) seen to be the same or different?” The same
goes for Jackendoff’s passage where he compares a language and a species:
I imagine, then, that to speak of a language in linguistics is a bit like speaking of a
species in biology: one acknowledges that members of a species are not genetically
identical; and cases sometimes arise where what is apparently one species shades off
imperceptibly over some geographical range into another. Does that mean there are no
species? Some biologists think so. But as long as we regard the term as a convenient
first approximation, there seems no harm in it. (2002: 35–6)
Again, we observers, we languaging living beings, create the categories species

and language to serve our purposes, be they scientific, philosophical, artful,
playful or otherwise. Now, it is not, as far as I know, that some biologists deny
the concept species. It is rather that they do not accept the classical empiricist
concept that Jackendoff is evidently subscribing to, because species, for them
(and for me) is not a first approximation of a category that exists out there and
which we aspire to getting right, i.e., to match with the facts. Rather, the con-
cept species is a result of a process of distinguishing among living beings and
of classing them either as same or as different according to the criteria that we
observe to be significant.
Chomsky valued the principle of essentialized sameness for the way that
it provides a foundation for universal human rights. A similar good can be
derived from the assertion by Tooby and Cosmides of the “psychic unity of
humankind – that is, a universal and uniform human nature” (1992: 79). To
reinvoke the distinction between monomorphic and polymorphic characteris-
tics made in What Emotions Really Are and mentioned in Chapter 2, Griffiths
critiqued the assertion of “the psychic unity of humankind” in terms of “the
doctrine of the monomorphic mind.” The reason for the popularity of the doc-
trine, Griffiths argues, is that the assertion of a monomorphic mind – one that
postulates no variations between and among people – acts as a bulwark against
racism. Our humanity, in this way of thinking, is secured in our genes, and
if our genes are all the same, then phenotypic racial differences are superfi-
cial and modifiable. However, there are other ways to avoid racism, one of
which is to distinguish pancultural traits from universal traits. “The fact that
racial differences in various cognitive traits are insignificant,” Griffiths writes,
“does not show that individual differences in those traits are insignificant.” He
continues:
If human phenotypic variation follows the pattern observed for genetic variation then
within-group differences should dwarf between-group differences … This suggests that
individual differences [in any given ability] may be substantial and important despite
the lack of between-group differences. Furthermore, between-group differences need
not be of a kind that would support anything like a race concept. (1997: 125)
To underscore the point that what is known as race is not a particularly useful
system of genetic classification for human beings, the population geneticist
Luigi Cavalli-Sforza has noted that “regardless of the type of genetic mark-
ers used (selected from a very wide range), the variation between two ran-
dom individuals within any one population is eighty-five percent as large as
that between two individuals randomly selected from the world’s population.
It seems wise to me, therefore, to abandon any attempt at racial classification
along the traditional lines” (2000: 29). A biological species is, however, a use-
ful enough classification, and although it is based on a series of criteria that the
classifier judges to be the same, a species is not, nor is it intended to imply, a
homogeneous classification. Again, it is a question of dimension of assessment.
Compared to other species, homo sapiens are all alike – which is why we class
ourselves as such. Compared to one another, however, we are all different.
What is the length of the developmental linguistic system in

the individual’s life?
Bruner uses the phrase developmental linguistics, but he uses it to refer only to
the very early years. In the present study, it is meant to refer to the ontogenic
phenotype’s entire existence, such that there is no moment in a languaging liv-
ing being’s ontogeny before languaging development starts from the moment
of conception and ends at the moment of death. However, it still makes sense
to review how we interpret some of the patterns evident in the child’s encounter
with language.
For instance, much is often made of the so-called childhood vocabulary
explosion during the second postnatal year. Often specialized mechanisms are
posited that build on the first words to better acquire new words. A recent study
suggests that specialized processes are unnecessary to explain the vocabulary
explosion. Instead, “acceleration is guaranteed in any system in which (i) words
are acquired in parallel, that is, the system builds representations for multiple
words simultaneously, and (ii) the difficulty of learning words is distributed
such that there are few words that can be acquired quickly and a great number
that take longer” (McMurray 2007: 631). The distribution of difficulty derives
from many factors, according to McMurray, including context, frequency,
phonological and syntactic factors, and even the child’s capabilities. Again,
even here, a developmental systems approach – which emphasizes the struc-
tural features of the organism–environment complex, not specialized learning
processes – accounts for the form of growth. “Acceleration,” McMurray writes,
“is an unavoidable by-product of variation in difficulty.”
In addition, many studies of so-called language acquisition focus on the crit-
ical period that is generally thought to extend until puberty but tend to cut off
investigation of children’s speech around four or five years of age. It does,
indeed, seem to be the case that for many, but not all, individuals the task
of, e.g., producing a native-speaker phonology in a second language becomes
more difficult after puberty. Understanding the relationship between ease of
learning (certain features of a) language and the complex neuro-social factors
involved in puberty is certainly part of the developmental linguistic system and
worthy of much study. However, puberty is but one phase in the languaging
living being’s ontogenic drift, just as the moment of birth is but one, albeit
important, moment that distinguishes what we might call, for consistency’s
sake, the difference between pre- and postnatal embryology. In addition, a
recent wide-scale study casts doubt upon the critical-period hypothesis for
second-language acquisition and suggests that in ontogeny there is no precipi-
tous drop in ability to learn a second language but rather a gradual decline over
the lifetime of the individual (Hakuta et al. 2003). Learning to language does
not come all at once but takes place over an entire lifetime. Roger Brown has
a beautiful image that grammatical morphemes “like an intricate sort of ivy,

begin to grow up between and upon the major construction blocks, the nouns
and the verbs” (1973: 249). One could add that the ivy shoots out cognitive
tendrils as both the child’s world and her conception of it expand and then,
necessarily, change.
Take, for instance, the topic of pronoun acquisition and use. It is well docu-
mented that children begin to produce pronouns at around two or three years of
age. However, the next thing to note is that they may not use them if the object
or person referred to is not in the same room with them until after seven years
of age. And to this can be added the work of psychologist James Pennebaker
on the oft-overlooked fascinations of his study of American English adult-use
of pronouns in general and the series I-me-mine in particular: that women tend
to use more pronouns and references to other people, including I-me-mine,
than men; that the older the person, the less they use first-person singular pro-
nouns (and – intriguingly – the more they use future tense verbs and the less
they use past tense verbs); that the higher the status, the lower the use of first-
person singular pronouns; that liars make less use of first-person singular pro-
nouns than truth-tellers; and that public figures speaking in press conferences
and published poets in their poetry use more first-person singular pronouns
when they are depressed or prone to suicide (Pennebaker 1997; Campbell and
Pennebaker 2003; Chung and Pennebaker 2007). The point here is not to turn
linguists into psychologists but rather to use psychology to make green and
grow the thin sticks of observations about pronouns – e.g., that like this, that,
here, there, today and yesterday, they belong to the class of words or expres-
sions called deictic/deixis – to which linguists seem content to limit them-
selves when our textbooks routinely announce that pronouns are acquired by
age two or three and that is the end of the story. The important classificatory
practices of the twentieth century deserve now to be infused with life by the
introduction of a psychologically real individual inhabiting from day one a
socially real world.
In addition, the learning of spatial frames of reference takes place slowly and
right through middle childhood (Levinson 2003). Functional neuroanatomical
differences can be shown to exist between adults and school-age children in
the processing of single words (Schlaggar et al. 2002), suggesting not only
brain differences but also different experiences with single words. (My experi-
ence of the word book, which includes the writing of this one, is necessarily
different from a ten-year-old’s.) Additionally, the relationship between speech
and gesture changes over a languager’s lifetime: in the earliest months of life,
speech and gesture are quite separate, becoming more unified between 14 and
23 months. Between the ages of 12 and 14 months, when speech and gesture
may be in conflict, infants attend to gesture. The onset of two-word utterances
is reliably related to the onset of production of gesture–speech mismatches, the
kind that signal the child’s readiness to learn something new (Goldin-Meadow
2003: 50), while abstract pointing might not emerge until around age twelve
(McNeill 2005: 193). Additionally, a study comparing young and elderly female
adults with regard to their use of gesture in an object-description task showed
a marked decrease of imagery-driven gestures in elderly females, while their
ability to describe the objects accurately was not impaired (Cohen and Borsoi
1996). The suggestion is that elderly women appear to be less involved with
visual imagery than young adult females. The overall point is that as we grow
and age, our languaging changes.
The kind of languaging I do now is obviously continuous with my languag-
ing of twenty and forty years ago, but it is not the same. Every time I return to
France, for instance, I notice some slight but perceptible syntactic change in
and among the more obvious lexical and semantic changes, with argot chan-
ging at blinding speed. Neither a particular language in a generation nor the
languaging behaviors in an individual are stable, although linguists sometimes
treat them that way. As a complement to Brown’s image of ivy, languaging for
me is like a neuro-social lace that gets tatted as individuals are woven into and
weave themselves into the community. The pattern in an individual’s lifetime,
although it changes, may remain recognizable. However, down through the
generations, the lace-like patterns of successive generations are quite distinct.
Tomasello may be distinguished as the current researcher of children and
language with the most to offer the ontogenic script in a developmental lin-
guistics. This is so because he argues forcefully against the need for anything
resembling a universal grammar and, at the same time, he is aware that the
alternative Bates-MacWhinney-type competition model of connectionism is
more of a model of language comprehension and not a model of language
learning (2003: 190). He is furthermore sympathetic to construction grammar,
and the aim of his Constructing a Language is to establish “the kinds of cog-
nitive and social-cognitive processes by means of which children acquire dif-
ferent kinds of constructions and organize them into some kind of structured
inventory and use them to produce creative yet canonical utterances” (2003:
192). He identifies the existence of general cognitive processes, which fall
under two headings, namely intention-reading and pattern-finding, that can be
grouped in four basic sets of processes:
• intention-reading and cultural learning, which account for how children

learn linguistic symbols in the first place;
• schematization and analogy, which account for how children create abstract
syntactic constructions out of the concrete pieces of language they have
heard;
• entrenchment and competition, which account for how children constrain their
abstractions to those that are conventional in their linguistic community;
• functionally based distributional analysis, which accounts for how children

form paradigmatic categories of various kinds of linguistics constituents.
(2003: 295)
In other words, Tomasello provides a guide to the psychology of construc-
tion grammar. For him, the most fundamental unit of intentional action is “the
utterance as a relatively complete and coherent expression of a communica-
tive intention, and so the most fundamental unit of language learning is stored
exemplars of utterances” (2003: 296). It follows from this that language learn-
ing involves both the building up and extending of constructions at the same
time that it involves the breaking down and analyzing of the many fixed so-
called holophrastic expressions the child (and all of us) routinely encounter:
I wanna X, Gimme X, Where’s-the-X? Can you X?, along with all the other
constructions previously mentioned. All this is to say that Tomasello gives us
a named psychological space that supports languaging and offers increased
room in which the linguist can move around.
A final historiographic review: thought and language

We are poised to push the ontogenic story further and are now armed with a
variety of cognitive processes that can be said to account for how the child
comes into languaging, those processes that the encounter with the environment
and languaging living beings can be said to put into motion. Nevertheless, we
are still faced with the problem of how to describe what is or is not inside the
infant/child as it first confronts the languaging environment. This problem takes
us straight to the highly problematic and long-studied relationship between the
constructs language and thought, a territory that requires a preliminary historio-
graphic weeding. As usual, we must confront the historical plaque covering the
issue in order to move into new territory. The current state of the issue can be
easily exemplified by a review of a fairly recent debate concerning what the
infant/child brings to the understanding of spatial concepts.
Three possible relationships between language and thought have tradition-
ally been advanced in the last several hundred years. They differ by the order
of precedence that is posited to exist between the two terms, that is, between
the two faculties and/or activities. The possibilities are, logically enough, that
(i) thought precedes language, (ii) language precedes thought, or (iii) thought
and language are contemporaneous.
Thought precedes language

This is the position usually associated with the Enlightenment, in particu-
lar with the British and French philosophers, the earliest and most influen-
tial expression of it issuing from John Locke. The thought-precedes-language
Thought and language 187
position is part and parcel of the rationalist empiricist philosophy we have

encountered several times already and whose features of interest to us in this
context can be summarized thusly: (i) there is a faculty of Reason; its appear-
ance on the human scene necessarily preceded the appearance of language,
since this faculty was the condition of the latter’s appearance; (ii) the purpose
of language is the communication of thought, given that language is primar-
ily a social activity; (iii) perception is passive and knowledge results from the
impressions left by the stream of sensations on the tabula rasa of the mind;
(iv) words are the labels attached to the single or multiple sensations after they
have left their marks on the mind; and (v) it is possible to conceive of making
language a better vehicle of communication; it can be improved by individual
and/or collective effort.
Note that, in this framework, language is primarily a process of naming, and
remember that the origins of language are often cast in terms of how the nam-
ing process got started and are often shaded in terms of a social contract.
Language precedes thought

For the sake of brevity and with no nuance, this position can be said to be
that of the later German Romantics. The name of Wilhelm von Humboldt,
a complex figure in this story, comes to mind. Here the order of precedence
between the two terms is obviously reversed, along with the attendant features
of the philosophical framework that supports the language-precedes-thought
position and that can be summarized thusly: (i) without language there would
be no thought; to think is to use symbols; thinking is internalized language;
(ii) as useful as language is for social intercourse, the forms of language serve
primarily the purposes of thought; (iii) perception is structured by the active
application of the framework of language to the flux of sensations; (iv) gram-
matical structures are the most characteristic and inward part of a language; (v)
there is no sense in which language can be improved or even altered by indi-
vidual and/or collective effort; language is an entity unto itself and is, to some
extent, imposed upon individuals.
Note that, in this framework, if what may be thought is dependent upon and
posterior to language, it follows that a particular language sets limits to what
is thought. The exact extent of the imposition and of the limits is open for dis-
cussion. Leaving room for the individual freedom of creative action, Humboldt
wrote famously that language is not ergon (work) but energeia (activity). In
any case, here comes into existence the idea that the frameworks of different
languages may differ, which leads to the possibility that individuals in differ-
ent language communities may adjust to so-called reality in different ways.
As for the origins of language, the German Romantics came to look on lan-
guage increasingly as a social fact, the possession of a nation rather than of an
individual, and this view fostered the idea that the origin of language was to
be found in collective activity. Humboldt advanced such a view when he wrote
that: “The existence of language proves, however, that there are also spiritual
creations which by no means originate with any individual, to be handed on to
other individuals, but which come forth out of the simultaneous, spontaneous
activity of all” (quoted in Brown 1967: 83). Since Humboldt does not (and,
in fact, cannot) spell out how this “simultaneous, spontaneous activity” came
about, his view is the collective equivalent of Müller’s individualistic ding-
dong theory, namely it endows language with a mystical glow all its own, one
that pulsates with collective soul.
Thought and language are contemporaneous

This position is, again briefly and without nuance, that of the early German
Romantics and has been described as the case where “language (exemplified by
a sequence of actual spoken words, say) is the outward manifestation of an inner
process that accompanies and matches it at every point” (Brown 1967: 54).
Historically, it marked a transition between the (heavily) French to the (heavily)
German thought–language positions. The central idea is that we do not have
two processes here, but only one. It proves to be a very unstable position in the
history of ideas. A contemporary version of the position may well be found in
Fodor’s and then Pinker’s construct mentalese, i.e., thought that is already lin-
guified, so to speak. Language, for Fodor and Pinker, is the outward manifest-
ation of an inner process of mentalese that accompanies and matches it at every
point. No behavioral or cognitive precursor of this mentalese is posited.
The French and German traditions come together, interestingly enough, in
the German-trained French-speaking Swiss linguist, Ferdinand de Saussure. In
his Cours de linguistique générale, he praises the eighteenth-century French
Port Royal grammarians for their method, i.e., synchrony, and writes: “clas-
sical grammar has been criticized as unscientific; still, its basis is less open to
criticism and its data are better defined than is true of the linguistics started by
Bopp” (1985 [1916]: 82). We furthermore saw in the visual evidence from the
ideogram of the talking heads in Chapter 1 that Saussure belongs to the line
of Lockean code theorists. At the same time and in direct opposition to Locke
and the Port Royal grammarians, Saussure aligns himself with the language-
precedes-thought camp when he states that:
Philosophers and linguists have always agreed in recognizing that without the help
of signs we would be unable to make a clear-cut, consistent distinction between two
ideas. Without langue, thought is a vague, uncharted nebula. There are no pre-exist-
ing ideas, and nothing is distinct before the appearance of langue. Against the float-
ing realm of thought, would sounds by themselves yield predelimited entities? No.
(1959 [1916]: 111–12)
Thought and language 189
Saussure is in the langue-precedes-thought camp, because without langue

“thought is a vague, uncharted nebula.” Here we can tease out another con-
trast between the two major thought–language positions: Saussure thinks of
thinking (without langue) as something messy, while for those who are in the
thought-precedes-language camp, thinking seems to be thought of as a fairly
organized activity. These days, the term cognition is often preferred to the term
thinking, as in the phrase cognitive science, perhaps because: (i) thinking has
often been imagined, as Saussure phrases it, as “a vague, uncharted nebula”;
(ii) thinking is often considered to be anchored by language; (iii) because of
its link to language, thinking is a less general term than cognition, since much
cognitive activity is non-linguistic; and/or (iv) because cognition carries noth-
ing of the contradictory ways the term thinking has been conceptualized.
In the eighteenth and nineteenth centuries those who engaged with the
thought–language problem were concerned to determine or to find universals
in every language that would be evidence of either: (i) the existence of uni-
versal Reason, a priori and innate, that precedes language; or (ii) the exist-
ence of universal archetypes that might not, in all languages, yet be manifested
(if those languages happen to be not quite “developed”), but will eventually
unfold. Although in the latter instance, it would seem that time would be a fac-
tor, it actually has no independent effect; time is necessary only as the dimen-
sion that allows for the appearances of the a posteriori inevitabilities.
A developmental systems linguistics has little to gain from traditional dis-
cussions of the thought–language problem because the theorists on all sides
have tended to spin out their speculations in an undifferentiated space that
confuses, collapses and compresses the differing temporal dimensions that are
required of developmental descriptions. They speak of “X preceding Y” in dis-
satisfyingly atemporal generalities that do not distinguish clearly or consist-
ently among the possibilities of whether X precedes Y:
• in the history of the species;
• in the development of the individual, i.e., the infant coming into language/
ing;
• in the production of any particular instance of speech.
As far as the last point goes, Dan Slobin has coined the useful phrase think-
ing for speaking to refer to the ways that different languages require children
to attend to different elements of a situation, depending on the grammatical and
morphological categories of their native languages (1996). His work is thus
not so much devoted to determining the possible overall cognitive effects of a
particular language but rather on what is required to be expressed in any given
situation. Similarly, McNeill (2005) wonderfully mobilizes the Vygotskyan
notion of Growth Point to argue, among other things, that: (i) context is not
an add-on in any given utterance but instead the context is part of the Growth
Point; and (ii) that the Growth Point is the moment when the two incompatible
and therefore unstable forms of communication, speech and gesture, get sorted
out, such that what is finally produced could now be called thinking for both
gesturing and speaking.
In general, the compression of time is to be expected for pre-Darwinian
theorists and yields vicious circles from which there is no escape, making the
carryover of these pre-Darwinian concerns and discourses to post-Darwinian
reflections unbeneficial. This is to say that DST linguistics has little interest
in questions about language universals formulated in terms such as: “are all
languages cut from the same cloth?” and/or “are there universal thought forms
or universal language categories?” since these questions are typically answered
in frameworks in which the effects and the play of the varying temporal dimen-
sions have no place. A developmental systems linguistics has interest in rec-
ognizing similarities and differences among languages. The similarities will
mostly be seen as cases of recurrent emergence: similar cognitive abilities in
similar circumstances will yield similar solutions. The differences will arise
from the fact that languages are historical products.
A case in point: the tight-fit/loose-fit distinction

With this background, we can evaluate the debate concerning the learning of
spatial categories in children engaged by Paul Bloom (2004) and Susan Hespos
and Elizabeth Spelke (2004), on the one hand, and that of Soonja Choi, Laraine
McDonough, Melissa Bowerman and Jean Mandler (1999), on the other. Hespos
and Spelke are building on Choi et al.’s earlier studies that investigate the spatial
distinctions between English in/on (containment vs. surface attachment) and
the Korean verb kkita which refers to actions resulting in a tight-fit relation
regardless of containment or support. In their article, Hespos and Spelke report
on experiments they performed on five-month-old infants raised in English-
speaking environments that showed the infants to be sensitive to the Korean cat-
egories of meaning. They conclude that “language learning therefore seems to
develop by linking linguistic forms to universal, pre-existing representations of
sound and meaning” (2004: 453). For my part, I do not see how a responsiveness
to certain conditions is evidence of anything but an ability to be responsive.
Nevertheless, Bloom applauds their results.2 He evidently means for thought
to precede language in both the ontogenic sense – since Hespos and Spelke are
2
Bloom, in fact, declares that “Saint Augustine was right.” Now, the fourth-century Saint Augustine
was made famous in academic circles in the twentieth century by Ludwig Wittgenstein, who
opened his Philosophical Investigations (1953) with Augustine’s autobiographical account of
how he learned to talk. Wittgenstein offered it as an example of the so-called picture theory
of language and the referential conception of meaning. He used it in order to argue against
it and to propose a conception of meaning-as-use. Bloom invokes Augustine to support the
Case in point: the tight-fit/loose-fit distinction 191
working with very young infants – and the phylogenic sense, since Hespos and
Spelke point out that non-human primates interact with the mechanical prop-
erties of objects in terms of the arrangements and motions of surfaces, which
differ for tight-fitting and loose-fitting objects. Thus, the human infant’s ability
to be responsive to the tight–loose distinction shows that this capacity does not
depend on language experience.
How does developmental systems linguistics respond to these results? It cer-
tainly has no difficulty in recognizing that all animals experience and display
evidence of cognitive activity for if they did not, they would be dead; and it
comes as no surprise that humans, in belonging to the primate lineage, share
cognitive activities with other non-human primates. At the same time, devel-
opmental systems linguistics does not reject out of hand the possibility that
language precedes thought in some sense. This is the position that Blooms
concludes is that of strong relativism. Based on what was just said about DST
linguistics embracing the notion that all living beings experience and exhibit
evidence of cognition, DST linguistics does not fall under this definition of
strong relativism, if one takes thought to mean “cognitive activity generally.”
Bloom opens his commentary with the idea that those who subscribe to linguis-
tic relativism maintain that “learning a language has a profound influence on
a child’s mental life. If so, then speakers of different languages might think in
very different ways.” So we could acknowledge, as did the German Romantics,
that there may be degrees in which language influences thought, with one of
those degrees being that language profoundly influences thought, such that
speakers of different languages might think in very different ways.
Although developmental systems linguistics does not embrace the idea
that language precedes all thought, it is nevertheless very comfortable with
the position that language profoundly influences thought and would even add
the rider: such that even speakers of the same language might think in very
different ways. Tomasello has stressed the importance of disagreements and
misunderstandings to understanding the relationship between one’s own men-
tal states and those of others (1999: 177ff.). Tomasello’s focus, of course, is on
children learning a language, but all of us throughout our daily lives find our-
selves in situations where we misinterpret and are misinterpreted, need or give
language-precedes-thought position, saying that it is possible that “the strong Augustinian view
is correct: language learning might really be the act of learning to express ideas that already
exist” (2004: 411). Now, Bruner, for his part, also invokes Augustine in order to argue against
him, consistent with Wittgenstein’s critique of language-as-a-naming-process. However, instead
of interpreting Augustine’s position on language learning as promoting a version of innatism,
as does Bloom, Bruner is unhappy with it rather because it allows a “mindless behavioristic
version” of language learning, one in which the learner is presumed to have “no predispositions
toward or knowledge of the material to be learned” (1983: 32). Although taking diametrically
opposite positions, Bloom’s and Bruner’s interpretations of Augustine’s account of language
learning agree to the extent that they both see in it an absence of any developmental story. Bloom
approves; Bruner disapproves.
clarification, find that our views do or do not coincide with those around us. A
similar point is made by Alton Becker in his “A Short Essay on Languaging,”
a part of which was quoted in Chapter 1. In fact, it is only through our lan-
guaging together that children and adults alike can realize that someone else’s
experience of a situation is different from their own. DST linguistics begins
with the premise that we are all different one from the other, and so it is not
surprising to find those differences manifesting themselves even under the so-
called same conditions of belonging to a community labeled, for instance, as
an English-speaking community.
Of course, Bloom is interested to determine whether Language A affects the
thought of a particular group of speakers, first, all in the same way and, second,
in a way that is different from the effects of Language B on another particular
group of speakers. Here the answer is undoubtedly Yes, as Hespos and Spelke
themselves suggest in their comment that “language experience reduces sen-
sitivity to conceptual distinctions not marked by the native language” (2004:
454–55). They go on, of course, to say that a native language does not produce
the concept, and although we could argue for a long time about whether con-
cepts are factory-installed, the three of us might agree that a reduction of sen-
sitivity of conceptual distinctions counts as a difference. The question remains:
how much of one? Do Koreans who regularly make the tight–loose distinction
think differently from English adults who do not make the tight–loose distinc-
tion but who have no trouble recognizing it once it is pointed out to them or
who are able to discover it on their own? The answer could be arrived at by:
(i) determining what dimension of assessment one is using to measure diffe-

rence. If one assumes that words (or symbols) are labels for already-existing
concepts, then there is, by definition, no difference to measure. Language
has no effect on thought. Language may then be conceived, as it was in the
eighteenth century and still is by many researchers today, as a dress cover-
ing thought. Just as there are many dress styles that may cover one physical
body, so there are many languages that can cover one body of thought, as
many theorists in the eighteenth century might have said and did say. Since
this seems to be the position of Hespos and Spelke, I speculate that, for
them, a reduction of sensitivity of conceptual distinctions on the preexist-
ing body of thought could at most, over time, chafe the skin and make it
unreceptive to a renewal of a conceptual distinction that it once had. Thus,
any reduction of sensitivity, if it made any difference, would only make a
superficial difference;
(ii) deciding, in the Hespos-Spelke model, how many reductions of sensitivity
of conceptual distinctions one is going to consider at once. Taking only the
tight–loose distinction, we might not want to say that Korean and English
thoughts are terribly different. If we added ten more conceptual distinctions,
thereby chafing the skin a lot, we might be inclined to observe more of a

difference. And if we were to throw in Korean honorifics, we would def-
initely have complicated the picture. However, if honorifics were to count
as so-called cultural distinctions that are a part of social, learned nurture
rather than so-called conceptual distinctions that are a part of biological,
innate nature then the three of us would have another disagreement – or,
rather, a different version of the disagreement about human infants com-
ing into the world loaded for concepts. Cross-disciplinary work on what
is called ethnosyntax, as well as culture–grammar interactions including
honorifics, are broadly accepted by linguists (see, for instance, Nevins
et al. 2009: 358). The question here is how the nativist is going to integrate
these studies. For the developmentalist, there is no problem.
What inspired Choi et al. to investigate the English–Korean differences in
the first place was their interest in contesting the view that “children’s early
spatial conceptualizations are tightly constrained to a limited set of highly sali-
ent notions” (1999: 243). They note that recent research has shown that spatial
classification is “more varied across languages than had previously been recog-
nized” (1999: 262) and that in/on is hardly a universal and can be broken down
and crosscut by other properties of spatial configurations, as in Korean. Now,
there is a way in which the two studies agree: Choi et al. say that “crosslin-
guistic variation in spatial classification suggests a corresponding flexibility in
human spatial cognition” – a statement Hespos and Spelke would presumably
not dispute – but there is a deeper way in which the two studies do not agree,
because in their next sentence Choi et al. go on to say that:
perhaps, then, infants do not approach the language-learning task with strong pre-estab-
lished biases toward just a few basic spatial concepts, but instead are able to appreciate
a wide variety of similarities and differences among spatial situations. In this case they
might be able to home in on language-specific principles of spatial categorization from
very early on in the acquisition process. (1999: 262)
The Hespos and Spelke experiments on preverbal five-month-olds were
designed precisely to determine whether sensitivity to spatial concepts pre-
cedes significant language inputs, and they determined that it did. (All they
had to do was load the infant for more spatial concepts than had previously
been thought to exist in the world’s languages, as Choi et al. point out.) By way
of contrast, Choi et al. are claiming that children learning the two languages
begin to acquire sensitivity to these language-specific categories by eighteen to
twenty-three months (1999: 261). The difference seems to boil down to what
counts as and how to understand preverbal vs. already-verbal sensitivity.
Choi et al. speak of the infant “acquiring sensitivity,” “appreciating a wide
variety of similarities and differences,” having the “ability to home in on
language-specific principles,” “attending systematically,” “concentrating on
tight fit and disregarding containment.” They speak of target words as “draw-
ing the attention of/guiding the attention to.” And, of course, they propose:
“Another possibility is that children do not necessarily develop all the needed
concepts nonlinguistically, but can construct them on the basis of the lin-
guistic input.” Their final description is a completely useful one for DST
linguistics:
hearing the same word across many different situations may lead children to identify
properties that are shared by those situations, and that distinguish them from situations
to which the word is not applied. This construction process presumably draws on chil-
dren’s perceptual ability to recognize many different kinds of similarities and differ-
ences across spatial situations, and on their ability to conceptualize what the situations
picked out by a word have in common … Linguistically informed work on this problem
will be crucial for determining how language-specific input interacts with children’s
nonlinguistic understanding of space to provide knowledge of the spatial semantic cat-
egories of a particular language. (1999: 264–65)
The care that Choi et al. take with their descriptions is a good model to follow
for how the ontogenic script should be written.
There is, now, a fourth possibility in the language–thought relationship that
can be mentioned, one that is associated with the twentieth-century psycholo-
gist Lev Vygotsky, who incorporates the effect of time and experience and then
language in developing thought. This development does not occur in a vacuum.
Quoting, once again, from Tomasello:
The point is not just that linguistic symbols provide handy tags for human concepts or
even that they influence or determine the shape of those concepts, though they do both
of these things. The point is that the intersubjectivity of human linguistic symbols – and
their perspectival nature as one offshoot of this intersubjectivity – means that linguistic
symbols do not represent the world more or less directly, in the manner of perceptual
or sensory-motor representations, but rather are used by people to induce others to con-
strue certain perceptual/conceptual situations – to attend to them – in one way rather
than another. (1999: 128)
This, of course, summarizes the tradition of languaging as orienting behav-
ior, one that has inspired Linguistics and Evolution and that enjoys a discur-
sive tradition extending to William James, at least, when he states: “All human
thinking gets discursified; we exchange ideas; we lend and borrow verifica-
tions, get them from one another by means of social intercourse … Agreement
thus turns out to be essentially an affair of leading – leading that is useful
because it is into quarters that contain objects that are important” (1981 [1907]:
97). Fleck says something similar when he notes: “Cognition is … not an indi-
vidual process of any theoretical ‘particular consciousness.’ Rather it is the
result of a social activity, since the existing stock of knowledge exceeds the
range available to any one individual” (1979 [1935]: 38).
Proceeding with caution, Choi et al. write that:

The crosslinguistic difference in looking behavior on the split pairs [where two images
shown simultaneously were distinctly different with respect to whether they were
actions representing ‘in’ or ‘kkita’] suggests that when the target word drew the atten-
tion of the two groups of children to the same scene in the conflated pairs [where the
two images could both be interpreted as action representing ‘in’ and ‘kkita’] it did so
for different reasons: For the English learners it was because the word drew their atten-
tion to the containment relation, while for the Korean learners it was because the word
pointed to the tight fit relation. (1999: 263)
Or, again, Choi et al. write: “it was the children’s understanding of the target
word that guided their attention to particular scenes during the test trials, not
simply a nonlinguistic preference for these scenes” (1999: 263). All this is to
say that in our languaging together, our discursifying, we are constantly lead-
ing and being led, constantly choosing this perspective on a situation rather
than that perspective, constantly privileging, neglecting, downplaying, fore-
grounding, distinguishing and presuming.
Languaging, all languaging, is a turning in this direction or that. Learning
a language means learning the history of the turnings that a group has estab-
lished in its languaging together. The problem with the Bloom-Hespos-Spelke
approach is not that they might think that those who language in Korean and
those who language in English might really, deep down in evolutionary history
and current biology, think alike, while others may perceive the two groups as
turning the attention of their fellows in albeit subtly different ways, which, for
them, counts as a difference. It is rather that Bloom-Hespos-Spelke imagine
that once they have decided the issue – thought precedes language, conceptual
distinctions are everywhere the same – the story is over. A developmental sys-
tems linguistics rejects the idea that the development of languaging behaviors
has no effect on the ontogenic phenotype over time and that the languaging
behavior said to be acquired by age three or five (or exhibited at five months)
is the same as the languaging behavior of the seven-year-old, the twelve-year-
old, the twenty-year-old, etc. As has already been said, many aspects of a com-
munity’s language may take years to acquire. The point here is that languaging
induces conceptual development in a number of domains and that particular
languages may induce particular, non-commensurate developments.
7 The ontogenic script continues
The question posed at the end of the previous chapter remains: what influ-
ence – if any – does a particular language have on the being, doing and know-
ing of any given languaging living being or group of languaging living beings?
The answer will take us into the intriguing realm of linguistic relativity, where
we will need to provide a developmental-systems-theory-inspired account of
how particular languages may affect particular languagers in the dimensions of
number, color, time and space. As usual, such an account will require “a more
generally phrased discussion,” as Waddington recommended for evolutionary
studies in Chapter 2.
Virtuous circles
In their lovely turn of phrase, Maturana and Varela speak of the “unbroken
coincidence of our being, our doing, and our knowing” (1992: 24). The poetry
of the unbroken coincidence invites a bit of explanatory prose. Maturana and
Varela favor circles. For them, going in circles is not a vice but a virtue. What
we do is bound to our experience of the world. What we know does not differ
from who we are. Who we are determines what we do. We are dynamic feed-
back systems of being, of doing, and of knowing. If you finish Linguistics and
Evolution with one idea, let it be this: our being, our doing, and our knowing
are inseparable. If you want a second take-home idea, it is this: we are our
being, we do our doing, and we know our knowing through our languaging
together – as this very sentence, paragraph, chapter and book attest.
How can we tell a vicious circle from a virtuous circle? Waddington shows
us the way. He considers two terms, A and B. A vicious circle occurs when we
attribute some property to A or claim that some subsystems within A produce
an attribute that belongs to A as a whole, and from these premises attempt to
deduce some character of B. The viciousness arises from the fact that B has
not been independently referred to in the premises. Waddington identifies such
a vicious circle at the center of Herbert Spencer’s work. The same one can
be found in the work of Julian Huxley, who claimed that evolutionary pro-
gress is good and then defined the good by means of evolutionary progress.
196
Virtuous circles 197
“The second clause, that good can be defined from evolutionary progress,”
Waddington writes, “adds nothing whatever to the first, which states that evo-
lutionary progress is good” (1960: 80). By the way, Herbert Spencer is perhaps
best known for coining the phrase survival of the fittest, which has its own cir-
cularity in that those who survive are, by definition, fit, and those who are fit
are the ones who survive. Now, if, for instance, we state that the complexity of
language is evidence that cognitive development must have attained a certain
level before language could appear (which would correspond to A), we cannot
use this to make deductions about anything other than the level of cognitive
development that is necessary to support language as we know it (which would
correspond to B). We have a vicious circle.
On the other hand, a valid circular statement of causal organization,
Waddington continues, is of the form that A conditions the appearance of an
attribute B and that B produces an attribute of A. In this case, there is no logical
fallacy. Instead, we have a model of evolutionary development and change if
one says, for instance, “that nature through the processes of evolution has pro-
duced in the human race a certain perceptive-intellectual apparatus (influence
of A on B), that this apparatus sees nature in a particular way (influence of B
on A)” (1960: 81). If at any moment A influences B, then it is likely to change
the manner in which B acts upon A, so that the system as a whole will become
modified as time passes. The context of Waddington’s remarks is, perhaps sur-
prisingly, the subject of ethics: “The processes of evolution have produced the
phenomenon that the human race entertains ethical beliefs. Man can then, not
so much through experiment but rather by taking account of its results, use
evolution to guide the way in which those beliefs will develop in the future”
(1960: 81). Waddington is concerned to understand how any characteristic of
a living organism must be thought of not only as something that has a func-
tional role in the life of the organism, and not only as something which has
been evolved through a considerable course of history, but also as something
that undergoes a process of development during the individual lifetime of the
organism. Accordingly, Waddington locates the pursuit of ethical discussion
within the social context of our particular beliefs as they have been built up
across the layered time frames that have structured the discussions throughout
the present study: the evolutionary, the socio-historical and the ontogenic.
The important point to make here is that organized causal systems – of the
type A influences B and B influences A – are not necessarily, or even usually,
stable in time. In fact, the temporal instability of such systems is one of their
most important properties. In this chapter, we are interested to apply this model
to the relationship between (some conception of) language/ing and (some con-
ception of) thought/thinking. To take the first temporal dimension – the evo-
lutionary – I have argued that the emergence, development and maintenance
of languaging behaviors in the species have significantly shaped, for instance,
the auditory and articulatory systems characteristic of humans, including neu-

romotor behaviors such as respiration. In addition, Givón (1998, 2002) has
proposed that the neural machinery that supports human language processing
evolved out of various components of the primate visual information-process-
ing system (influence of A on B), which would imply that, over the long term,
the human visual information-processing system has also been affected by lan-
guaging behaviors (influence of B on A – the McGurk effect would serve as
an example).
Furthermore, the emergence, development and maintenance of languaging
behaviors in the species have significantly shaped neurological connectivity
in general, including: inter-hemispheric connections and torque; cortical con-
nections with the larynx, tongue, lips and jaw; the connections between and
among such cortical areas as Broca’s, Wernicke’s, Heschl’s, etc., and the mir-
ror neurons in the premotor cortex, as well as the connections that come to be
established between cortical areas and subcortical structures, such as the basal
ganglia and the thalamus. As mentioned previously, it has been shown that the
thalamus undergoes an unusual ontological development in humans in terms of
cellular growth. Since the thalamus plays a central role in attentional process-
ing (filtering distracting information) and integrating numerous neurological
functions and neuronal circuitries, it has come to be recognized as an import-
ant structure supporting high-level cognitive activity such as languaging. In
addition, both the thalamus and the neocortex are disproportionately larger in
humans than in other primates. Given this, certain researchers have hypoth-
esized that the expansion of the neocortex may well have co-evolved with the
expansion of the thalamus (Li 2002: 100–01).
It is also the case that the emergence, development and maintenance of lan-
guaging in the species has affected the social nature of human cognition in
general, evidenced in the newborn’s earliest behaviors, which can be described
as a readiness to respond to all aspects of the languaging environment and
through which a languaging living being develops. Although languaging is not
co-extensive with the whole of human cognition, as has been said several times
already, the view taken in Linguistics and Evolution is that one cannot turn too
many corners in the brain without running into structures or connections that
support languaging (influence of A on B) and that, therefore, have also been
affected by languaging (influence of B on A).
As far as the socio-historical dimension with respect to languaging behav-
iors is concerned, it was established in the early chapters that the discipline of
linguistics came into existence when researchers began to organize and under-
stand the nature of temporal instabilities in the languages of the world, i.e.,
began to develop philology, aka historical and comparative linguistics. (Let us
note that philology concerns itself not only with the instabilities but also with
the stabilities that can be observed over time.) The American linguist Edward
The Sapir-Whorf hypothesis 199
Sapir summed up a particular view on the causes of temporal instabilities when

he aphorized that “All grammars leak” (1921: 38). Today we would say that it
is not grammars that leak but rather individuals, who are differently oriented
but coincident in their consensual coordinations as they language in particular
situations, who unweave and reweave the linguistic materials that have been
handed down to them. Another way of saying this is that grammaticization is an
inevitable process of languagers working with those materials, also known by
Givón’s well-known maxim: yesterday’s syntax is tomorrow’s morphology.
A full understanding of that maxim requires an understanding of the social
dynamics at work in the morphologizing process. Japanese pronouns and hon-
orifics offer a vivid example. The most formal form of ‘I,’ watakusi, origin-
ally meant ‘slave’ or ‘servant,’ while ‘you’ in the medium formal form kimi
meant ‘king.’ The copula haberu is self-humbling and derives, unsurprisingly,
from han ‘to crawl’ and iru ‘to be.’ Penelope Brown and Steven Levinson
understand the situation thusly: “Of course such etymological origins are quite
irrelevant to the modern speaker of Japanese; they are simply honorific forms
that have to be learned. Our point is that they are not arbitrary forms: they ori-
ginate as productive outputs of face-preserving strategies which then become
stabilized and change their meaning” (1987 [1978]: 278–79).1 Learning a lan-
guage means, in effect, learning the shaken-down effects of the social history
of a particular group of people.
First we grant that languaging behaviors have significantly shaped the human
brain/body. Next we want to determine the systematic ways that a particular
language may affect the being, doing and knowing – the cognition, broadly
conceived – of those who live their lives in and through it.
Edward Sapir and Benjamin Lee Whorf and

the Sapir-Whorf hypothesis
Future linguists are sometimes drawn to the field when they hear that the
Eskimo have a hundred words for snow or that the Chinese character for the
word crisis is made up of the characters danger and opportunity. These exam-
ples lend a certain mystique to lives lived through languages very different from
our own. The Eskimo can be imagined to perceive an exotic winter wonderland
of unimaginably fine distinctions, while the rest of us slog around in undiffer-
entiated snow. The wise Chinese can be admired for being able to view a dif-
ficult situation as an opportunity for positive change, while Westerners tend to
resort to crisis-mode behavior when faced with steep challenges. Interest and
even astonishment at the diversity of the ways that humans beings lives their
lives through their languages only increases when students learn, for instance,
1
There is not room here to explore Brown and Levinson’s concept of face.
that the Dugum Dani language of the New Guinea West Highlands has only
two colors terms: modla for light, bright colors and mili for dark, dull colors
(sometimes translated misleadingly as ‘white’ and ‘black,’ as in Berlin and
Kay 1969: 46). In addition, the Pirahã, a hunter-gatherer tribe of about two
hundred people who live in small villages on a tributary of the Amazon, are
essentially innumerate in that they have only two number terms: hói = ‘one’ (or
‘small’) and hoí = ‘two,’ while all other groups of objects are designated baagi
(or aibai) = ‘many.’ Their neighbors, the Mundurukú, lack words for numbers
beyond five. If we embrace the concept of linguistic relativity and entertain the
idea that language influences thought, where thinking is broadly conceived in
terms of perception and/or cognitive activity, we might be led to ask whether
the Dugum Dani view the world monochromatically and whether the Pirahã
and the Mundurukú are incapable of conceiving of large numbers.
Upon investigating matters, students may be disappointed (or relieved) to
learn that: (i) the original reference to Eskimos and snow, made by Franz Boas
in the early part of the twentieth century, was aimed at showing structural dif-
ferences among languages, given that Eskimo has four lexically unrelated
words for snow: aput, ‘snow on the ground,’ qana, ‘falling snow,’ piqsirpoq,
‘drifting snow’ and qimuqsuq, ‘snow drift’; and this example originally carried
no cultural or cognition implications (see Martin 1986);2 (ii) the Chinese word
in question, weiji, made up of the characters wei2 and ji1, is more correctly read
as “an opportunity for danger,” just as the English word photo-op is an oppor-
tunity to take photographs; whether it presents any kind of positive opportunity
for the person being photographed is incidental;3 (iii) human beings the world
over – despite whatever color terms are available in their native languages
and despite the differing boundaries that can be drawn among the various
color terms in different languages – are able to discriminate the full range of
hues, and they will generally agree on the focal points for those hues; and (iv)
human as well as non-human animals can conceptualize numbers, including
small quantities and large approximate estimates, without language; and there
is evidence that numerical approximation is a basic competence, available to
2
But see also Oozera et al. (2004) for a description of the nuanced terminology the Yupik have
devised to describe the complex snow and ice formations in their environment based on changes
in the wind. In this way, the “Eskimo words for snow” example does, indeed, illustrate the notion
of linguistic relativity outlined in the chapter, but not through the way that Boas’s example was
used and then popularly misused.
3
Speaking of Chinese and relative linguistic effects, there is evidence that musicians who speak
tone languages such as Mandarin, Cantonese, Vietnamese and Thai have a greater incidence
of perfect pitch than Western musicians. This is so because speakers of tone languages tend to
speak with greater pitch stability than speakers of intonation languages. Thus, when musicians
are learning the notes of the musical scale, they approach it as a verbal labeling task. It is as if
“music is acquired by speakers of tone languages as though it were a feature of a second lan-
guage” (Deutsch and Henthorn 2004: 351). And like many aspects of language acquisition, the
earlier the musical training, the more likely the incidence of perfect pitch.
preverbal infants and many animal species. These kinds of examples are used
to make the case for the position that human cognition is everywhere the same,
independent of the language in use in the community.
We have now engaged with one of the most intriguing and contentious issues
to be entertained in twentieth-century linguistics – namely, the relationship (or
lack thereof) between the features (lexical, syntactic) of particular languages
and the cognitive abilities/tendencies of those who live their lives through
them. The so-called founding statement of what has come to be known as the
doctrine of linguistic relativity was articulated by the linguistic anthropologist
Edward Sapir more than eighty years ago:
Human beings do not live in the objective world alone, nor alone in the world of social
activity as ordinarily understood, but are very much at the mercy of the particular lan-
guage which has become the medium of expression for their society. It is quite an illu-
sion to imagine that one adjusts to reality essentially without the use of language and
that language is merely an incidental means of solving specific problems of commu-
nication or reflection. The fact of the matter is that the ‘real world’ is to a large extent
unconsciously built up on the language habits of the group. No two languages are ever
sufficiently similar to be considered as representing the same social reality. The worlds
in which different societies live are distinct worlds, not merely the same world with dif-
ferent labels attached. (1949 [1929]: 162)
The phrase that human beings are “very much at the mercy” of a particular
language is quite striking in this passage, as is Sapir’s rejection of the idea that
language is any kind of labeling of so-called reality. He opened his earlier book
Language by contrasting what he called the “inherent, biological function” of
walking with the “social” activity of talking: “Eliminate society and there is
every reason to believe that [the individual] will learn to walk, if, indeed, he
survives at all. But it is just as certain that he will never learn to talk, that is, to
communicate ideas according to the traditional system of a particular society.”
Although there is evidence that a human being may not learn to walk upright if,
say, raised by wolves, it is surely the case that s/he will not learn to talk if not
raised in a human environment. In any case, Sapir continues a series of reflec-
tions that lead him to the statement that: “Speech is a human activity that varies
without assignable limit as we pass from social group to social group, because
it is a purely historical heritage of the group, the product of long-continued
social usage” (1921: 4). The phrase “varies without assignable limit” was to
be attacked some decades later by Chomsky, who was concerned, precisely,
to establish those limits within a conception of a generative grammar as an
inborn capacity unique to humans that operates entirely independently of any
particular social group.
Benjamin Lee Whorf was a chemical engineer by training, a fire prevention
specialist at the Hartford Insurance Company by profession and a reputable
scholar of linguistics and anthropology. Whorf’s great interest in American
Indian languages led him to the International Congress of Americanists

in 1928, where he met Sapir for the first time. In 1931, when Sapir left the
University of Chicago to take a position at Yale as professor of anthropology,
Whorf enrolled in Sapir’s course in American Indian linguistics. Whorf’s asso-
ciation with Sapir brought him further into the field of American Indian stud-
ies. He has become best known for his work on Hopi, especially his description
of the Hopi conception of time. Although there are disputes about how strongly
or weakly Whorf advocated linguistic relativism, the following oft-quoted pas-
sage from Whorf gives a good flavor of his position:
Formulation of ideas is not an independent process, strictly rational in the old sense,
but is part of a particular grammar, and differs, from slightly to greatly, between dif-
ferent grammars. We dissect nature along lines laid down by our native languages. The
categories and types that we isolate from the world of phenomena we do not find there
because they stare every observer in the face; on the contrary, the world is presented in
a kaleidoscopic flux of impressions which has to be organized by our minds – and this
means largely by the linguistic systems in our minds. (1956: 213)
Here we get more of a gradient sense of relativism, in that particular grammars
can differ from slightly to greatly. To whatever degree we want to say that
Whorf advocated linguistic relativism, it is fair to say that he believed that: (i)
all higher levels of thinking are dependent on language, in which respect he can
be aligned with Vygotsky; and (ii) the language people regularly use influences
the way they understand and interact with their environment.
Although Sapir and Whorf never co-published any papers, their names are
forever linked, mostly probably through the passage in Carroll’s introduction
to Language, Thought, and Reality, a collection of Whorf’s articles, where
Carroll refers to “Whorf’s principle of linguistic relativity, or, more strictly,
the Sapir-Whorf hypothesis (since Sapir most certainly shared in the devel-
opment of the idea)” (1956: 27). Not incidentally, Carroll notes that Whorf’s
work was not a rationalization for a failure of communication between cul-
tures and between nations. Rather, Carroll writes that Whorf would have
hoped that “a full awareness of linguistic relativity might lead to humbler
attitudes about the supposed superiority of standard average European lan-
guages and to a greater disposition to accept a ‘brotherhood of thought’
among men” (1956: 27). Similarly, Chomsky anchored his version of gram-
mar in the human genome, since a “concern for intrinsic human nature poses
moral barriers in the way of manipulation and control” (1988: 166). The
point here, once again, is that no one position can lay exclusive claim to
the ethical high ground. Whorf’s nurture-based and Chomsky’s nature-based
orientations issue from and to the same ethical imperative: recognition and
respect for all peoples of the earth.
After slumbering for some years in disrepute (for the very reasons that
the Chomskyan model of language and the mind prevailed in linguistics and
the cognitive sciences), the Sapir-Whorf hypothesis has become a subject of

renewed interest. Studies devoted to it have flourished over the past decade
and more, and these have investigated differences concerning the linguistic
expression of such categories, among others, as number, color, time and space,
and systemic differences – be they perceptual, cognitive and/or behavioral –
to be found among languagers from different languaging communities where
these categories are organized differently. The strategies for getting at these
perceptual, cognitive and/or behavioral differences are various and require,
first and foremost, that the researcher have good listening and observing skills.
Strategies may take the form of something as simple as asking subjects to
draw a picture of some feature of their mental experience. Or they may involve
measuring or analyzing:
• gestures, which are of interest because they are “a less conscious and moni-
tored track than language” (Núñez and Sweetser 2006: 403);
• linguistic and non-linguistic sorting tasks;
• linguistic and non-linguistic matching tasks;
• preferential looking and head-turning tasks, as well as sucking rates (often
used in tests with preverbal infants);
• short- and long-term memory tasks;
• the rate, ease and/or accuracy of cognitive functioning and/or processing;
• error profiles (slips of the tongue/hand) that occur when speaking or signing
and that provide clues to mental organization.
Linguistic relativism is a protean notion. A useful systematization of it has
been offered by the social psychologist Joshua Fishman, who identified four
possible levels of interpretation of the Sapir-Whorf hypothesis, namely in
terms of:
(i) differences in lexical structure between languages that correlate with cul-
tural/psychological differences;
(ii) differences in lexical structure that correlate with non-linguistic behav-
ioral differences;
(iii) differences in syntactic structure between languages that correlate with
cultural/psychological differences;
(iv) differences in syntactic structure that correlate with non-linguistic behav-
ioral differences.
The first level is the least novel and, therefore, the least interesting. It is a
truism that different languages have different semantic structures. It is well
known that languages may differ in the way they organize semantic fields. In
English, for instance, long thin bodies of water are distinguished by width:
creek, stream, river, etc., while in French, the more central distinction is how
the long thin body of water flows: a rivière flows into a fleuve, and a fleuve
flows in the ocean. The only thing that results from distinctions such as these,
Fishman tells us, is that “it is much easier to refer to certain phenomena or to
certain nuances of meaning in certain languages than in others” (1960: 326).
That is, we have here only a relationship between some aspect of language
and some aspect of language behavior that may or may not have any deep psy-
chological or cultural meaning. The English in/on versus Korean kkita debate
turns on a level-one interpretation of Sapir-Whorf, since it is clear from the
Hespos and Spelke experiments that “mature English speakers have little dif-
ficulty distinguishing tight-fit from loose-fit categories once they are pointed
out, and many English speakers discover the categories on their own” (Hespos
and Spelke 2004: 455). Indeed, a distinction between tight-fitting and loose-
fitting clothes or shoes is one that is often and easily made in English.
The second level points to a relationship between some aspect of language and
some aspect of non-linguistic behavior. Here it can be said that there is evidence,
for instance, that the color terminology used in a particular language has an effect
on certain non-linguistic processing and memory tasks associated with color: that
is, colors that can be named with a single word in language X exhibit a shorter
response latency when they need to be named than do colors that do not have a
particular word in language Y; in addition, named colors will be more readily
recognized or remembered when they must be selected from among many colors
after a period of delay after their original presentation. Still, here, we are not in
any particularly deep waters, and a version of this level is more or less engaged in
a series of experiments on cultural conceptions of time, described below. Indeed,
those studies point to a conclusion of a weak determinism.
The third and fourth levels engage with comparing language structures, i.e.,
grammatical categories, parallel to the way that the first two levels engage
with comparing lexical items and inventories. That is to say that the third level
has (some of) the problems of the first level: there is not much to show for how
the grammatical differences from one language to the next affect anything
beyond the linguistic differences. It is difficult to relate particular grammati-
cal features to particular cultural features, and the efforts sometimes get lost
in clichés of the type that “the heaviness of German philosophy is due to the
heaviness of the German language.” According to Fishman, the third level of
analysis has not normally been able to supply independent confirmation of
the existence of the cultural something that the grammatical data is taken to
indicate. A notable exception is Whorf’s work where he relates large-scale
cultural differences that correlate with linguistic differences in the conceptu-
alization of time, described below.
This brings us to the fourth level that Fishman identifies – the Big One,
so to speak, where linguistic structure as such can be said to have some kind
of major cognitive effect. This is to say that the fourth level moves through
the forms of a particular language in order to discover whether those forms
and their relationships affect particular aspects of non-linguistic cognition and

behavior. This is something more than saying that the presence or absence of
a lexical item or a grammatical structure will make it more likely that an indi-
vidual will behave in a certain way in a certain context. This is also something
more than saying that it is possible to translate the general idea of some stretch
of one language into another language, thereby enabling some amount of con-
sensual coordination between or among languagers from different languaging
backgrounds. Furthermore, this fourth level puts at issue the dual assertion
that: (i) language serves to integrate cognitive functions that would not be inte-
grated without language and makes certain mental processes possible; and (ii)
particular linguistic forms and structures that are intertwined both socio-histor-
ically and ontogenically with conceptual structures and activities may integrate
mental processes differently in different groups of languagers. To exemplify
what is meant here, Pierre Pica et al. note at the very end of their article on the
Mundurukú, a group that lacks words for numbers beyond five, that:
the availability of number names, in itself, may not suffice to promote a mental repre-
sentation of exact number. More crucial, perhaps, is that the Mundurukú do not have a
counting routine. Although some have a rudimentary ability to count on their fingers,
it is rarely used. By requiring an exact one-to-one pairing of objects with the sequence
of numerals, counting may promote a conceptual integration of approximate number
representations, discrete object representation, and the verbal code. (2004: 503)
Thus, it is not the presence of words for numbers, say, from one to one hundred
that is important, it is the activity of counting – what one does with those num-
bers, what one asks children to do with those numbers – that has the cognitive
effect. Western children around the age of three seem to catch on to the idea
that each count-word refers to a precise quantity, while “the ‘crystallization’
of discrete numbers out of an initially approximate continuum of numerical
magnitudes,” Pica et al. write, “does not seem to occur in the Mundurukú”
(2004: 503). We have moved far beyond philologism here, far beyond the
kinds of information that one finds in grammar books that travel the borders
between languages. As Fishman has said, this fourth level is the most demand-
ing, because it requires detailed technical training at both the predictor and the
criterion ends of the relationship to be investigated.
Given that this chapter opened with the assertion of the unbroken coinci-
dence of our being, our knowing and our doing, Fishman’s categories may look
entirely arbitrary, since there is no way to fully distinguish one’s behavior from
some kind of neural organization that affects and is affected by it. And, indeed,
the example of the lack of a behavioral counting routine for the Mundurukú
was chosen precisely to exemplify a relationship where the behavioral and the
neurological cannot be distinguished one from the other. It is also to assert
that the Mundurukú cognize about numbers differently than languagers who
cognize about precise quantity. Nevertheless, Fishman’s categories are useful

because they highlight what has traditionally been considered linguistic gold
when it comes to the Sapir-Whorf hypothesis, that is, finding some non-trivial,
deep-down cognitive differences that cannot be dismissed as “mere” behavio-
ral differences and that can be attributed to the particular language that a group
of people language in.
The discussion of emotion at the end of Chapter 2 was intended, among
other things, to set up the discussions of number, color, time and space in
this chapter. Recall that Griffiths argued that a cross-cultural understanding of
higher cognitive motivational complexes such as envy, guilt, jealousy, amae
and others requires a different descriptive framework than either the terms
of the affect program allow (human emotions found across the world are all
instances of the same thing) or what the social constructionists have offered
(human emotions found across the world are highly variable). It is precisely
the developmental systems framework that calls for descriptions in terms of
heterogeneous construction that will account for both the similarities and the
differences to be found in how love, embarrassment, guilt and other emotions
are instantiated and experienced in the world’s languaging groups.
Number
Although it is becoming rarer and rarer these days to stumble upon a previ-
ously undiscovered language, newly emerging data from obscure languages,
which may have features that lie outside expectations, can shake things up
theoretically, since such so-called exotic features might provide new wedges
into prying open the relationship between language and thought/cognition.
Two relatively recent reports on two relatively small groups in the Amazon,
the Pirahã (c.200 people) and the Mundurukú (c.7,000 people), bring the
language–thought relationship into high relief around the issue of the pres-
ence – or, more precisely, the absence – of a counting system and its effect on
numerical cognition.
The Pirahã use a “one-two-many” system of counting that consists of
the words hói (‘one’ or even ‘roughly one’), hoí (‘two’) and baagi or aibai
(‘many’). Hói can also mean “small” and can contrast with ogii (‘big’). The
researcher, Peter Gordon, had his participants perform tasks that involved non-
verbal numerical reasoning and that required some combination of cognitive
skills, such as the need for memory, speed and mental-spatial transformations.
The opening task required that they match in one-to-one correspondence a
linear array of AA batteries with another set of AA batteries. The tasks then
progressed to, e.g., matching nuts to the battery line and to matching a battery
line that was unevenly spaced. In these experiments, “participants responded
with relatively good accuracy with up to two or three items, but performance
Number 207
deteriorated considerably beyond that, up to eight to ten items.” Other tasks

involved the researcher putting, say, up to eight nuts in a can and then with-
drawing them one by one. The participants were asked, each time a nut was
withdrawn, whether any nuts were left in the can. Again, performance accu-
racy fell off when numbers over three were involved. Gordon notes that the
nonverbal numerical abilities of the Pirahã are greatly affected by their limited
counting system and states that “the present study represents a rare and perhaps
unique case for strong linguistic determinism” (2004: 498).
The first task that Pierre Pica et al. asked of their Mundurukú participants
was to state how many dots were present in displays of one to fifteen dots. In
this way they were able to determine that the Mundurukú have set terms only
for numbers one to five, that these expressions are long and that they often have
as many syllables as the corresponding quantity. Predictably, Mundurukú do
not perform well on tasks involving exact arithmetic with numbers larger than
four or five. It seems that they do not use their numbers for counting sequences
but rather for approximations. Thus, the word for five can be translated as “one
hand” or a “handful” and was used to refer to six, seven, eight and even nine
dots. Pica et al. then were interested to study whether the lack of a Mundurukú
counting system above five affected their ability to apprehend large numbers.
Here the answer seems to be No. That is, they succeeded in consistently iden-
tifying the more numerous set when asked to compare two sets of, say, twenty
to eighty dots. Pica et al. conclude, further, that the Mundurukú can mentally
represent numbers beyond their naming range and that they do not confuse
number with other variables such as size and density. They are also able to add,
subtract and compare their approximate representations. Pica et al. state that
“approximation is a basic competence, independent of language, and available
even to preverbal infants and many animal species. We conclude that sophisti-
cated numerical competence can be present in the absence of a well-developed
lexicon of number words” (2004: 503). The idea is that their study provides a
qualification of Gordon’s, in that the lack of a number lexicon does not com-
pletely limit the ability to work with abstract number concepts.
Yet Gordon acknowledges the general consensus that there seems to be two
kinds of numerical ability: (i) an ability to enumerate accurately small quantities
up to three; and (ii) an ability to estimate rather accurately large quantities with-
out overt counting (2004: 498). In their fascinating book Where Mathematics
Comes From: How the Embodied Mind Brings Mathematics into Being, George
Lakoff and Rafael Núñez state that this ability is called subitizing, from the
Latin word for “sudden” (2000: 19), and they acknowledge that newborn babies,
along with a range of species from primates to pigeons, are able to subitize, to
estimate numbers, and to perform the simplest addition and subtraction. Pica
et al.’s point of contention with Gordon seems to be one of emphasis, of how
strongly we wish to interpret the influence of language on cognition.
Our purpose here is not to deny the Pirahã and the Mundurukú the numeri-
cal abilities they have. Neither is it to collapse obvious differences into some
instance of “the same” numerical abilities to be found in all languaging living
beings in the world. Now, the potential for more elaborated forms of arithme-
tic and mathematics may be present in all humans. Indeed, it is reported that
Pirahã children easily learn number words in Portuguese, but, as one might
expect, Pirahã adults, who are farther along in their ontogenic drifts, are not
similarly adept. It is reported that they “lose interest” in such number lessons.
However, this losing of interest is not only related to an individual’s age; it
also involves other aspects of the environment that either would or would not
support an interest, e.g., the existence of a counting routine or the custom of
counting on fingers, which the Mundurukú do not have. As is evident from the
case of the Pirahã and the Mundurukú, not all human groups have found it use-
ful, desirable or necessary to develop more elaborated forms of arithmetic and
mathematics. Another way of saying this is that the more elaborated forms of
arithmetic and mathematics that are commonplace in many parts of the world
today have long and complicated histories, the particular effects of which can
be felt in the places where they have been important. The kind of mathematics
that is used in today’s computer technology, for instance, is not the invention of
a single human but is rather the result of a complex tradition that has taken its
course in a variety of social settings and as a result of many cultural contacts.
It is a commonplace to note, for instance, that it took Western mathematicians
many centuries to fully grasp the benefits of the Arabic numeral zero and that
the zero radically transformed Western arithmetic.4
In The Cultural Origins of Human Cognition, Tomasello engages, in several
passages, with a range of historical and cultural factors affecting mathematical
practices in an effort to show, precisely, how social are those practices, despite
the fact that “nothing seems less social than mathematics” (1999: 185). Lakoff
and Núñez, for their part, have a similarly multilayered view of mathematics
as a “product of the neural capacities of our brains, the nature of our bodies,
our evolution in our environment, and our long social and cultural history”
(2000: 9). Their particular emphasis is on mathematics as a systematic exten-
sion of the mechanisms of everyday cognition, conceptualizations that involve,
e.g., collections of objects in a bounded region of space (class), repeated
actions (recursion), rotation (complex arithmetic), and motion and approach-
ing boundaries (calculus) (2000: 28–29). As we have been saying throughout,
the mechanisms of our everyday cognition have been significantly shaped by
4
The debate between Nevins et al. (2009) and Everett (2009) in a recent issue of Language does
not concern itself with numeracy but rather with features of Pirahã syntax and whether or not
they (a) conform to Everett’s proposed Immediacy of Experience Principle and (b) confirm or
disprove a Chomskyan notion of Universal Grammar (however that is interpreted). The position
taken in Linguistics and Evolution is that of Levinson (2003) and Evans and Levinson (2009).
Color 209
our languaging together, and it follows that thus so should our mathematics. It
also follows that an individual’s mathematical cognizing will be affected by the
environment in which that individual’s being, doing and knowing is realized.
Color
Compared to the discussion about number cognition, which has arisen only
recently with the reports on the Pirahã and Mundurukú, the discussion about
color perception and the world’s languages has a long history. It has been
known for quite a few centuries that different languages provide different terms
for color and that those terms can cover quite different boundaries in the color
spectrum. This led to the view that color terms, like so many other semantic
categories found in the world’s languages, were arbitrary and could be said to
“vary without assignable limit.” As H. A. Gleason put it in his Introduction to
Descriptive Linguistics: “There is a continuous gradation of color from one
end of the spectrum to the other. Yet an American describing it will list the hues
as red, orange, yellow, green, blue, purple, or something of the kind. There is
nothing inherent either in the spectrum or the human perception of it which
would compel its division in this way” (1961: 4, quoted in Berlin and Kay
1969: 159). Such a view also prompted a desire to discover the extent (if any)
to which the presence or absence of color terminology in a language correlates
with any perceptual differences in its languagers.
Brent Berlin and Paul Kay’s Basic Color Terms: Their Universality and
Evolution (1969) is a landmark study that altered the theoretical landscape on
color research by showing: (i) that the foci of color terms for different languages
are highly convergent, even if their boundaries are highly divergent, meaning
that humans the world over perceive color similarly, and (ii) that there is a
hierarchic sequence in the order in which color terms are included or excluded
in language that unfolds historically in seven regular stages, suggesting that
the way English (or any other language) has cut up the spectrum has quite a lot
to do with something inherent in the spectrum and the way humans perceive
it. In other words, the perception of color and the lexification of color percep-
tion do not vary without constraint. The Berlin and Kay model is a version
of Levinson’s assertion that, given the fact of gravity, linguistic solutions on
the vertical plane do not vary and thus our experience of space on the vertical
plane is not affected by language (although Levinson will say that our experi-
ence of space on the horizontal plane is dramatically affected by the languages
we speak: see below). That is, the human experience of color is universal and
reduces not only to physics and neurophysiology but also precisely to the color
experience of Western science, which is based on spectral hues.
Berlin and Kay (1969) has not escaped criticism, particularly for its bias
toward hue over saturation and for the assumption that neurophysiological
perception is unmediated by experience. As a case in point, the philosopher

Angus Gellatly reports on an interesting color experiment performed by the
Russian psychologist Alexander Luria on a group of local women in Uzbekistan
in the 1930s. The women were given a pile of skeins to sort by color, and they
did so by groupings based on brightness and saturation and not by hue. Upon
request, many were able to modify their sorting according to hue, but some of
the women refused, arguing that the skeins sorted by hue did not go together.
Even the women that did sort by hue were not happy about it. The point is that
“contrary to the usual interpretation of Berlin and Kay, their colour experience
was not organized around focal colours” (Gellatly 1995: 211). Interestingly
enough, Berlin and Kay themselves report on such a case, quoting a turn-of-
the-twentieth-century anthropologist who studied aboriginal Australian tribes
and who wrote:
there was a natural tendency to put together all the wools to which the same name
was given, thus in Murray Island, Holmgren’s green wool was often called kakekakek
(white) or pipi (grey), and there was a tendency to place with it other unsaturated wools
of any colour to which the same name would be applied. One would often hear a native
saying ‘kakekakek’ to himself as he picked up a colourless wool to place with the green.
(1969: 147)
Berlin and Kay use this example not to question their assumption about the
importance of focal colors, but rather to question the anthropologist’s pro-
nouncement that the Murray Island group had a definite term, in this case,
for green. Berlin and Kay think rather that the Murray Island group counts
as a Stage I language, in that it has two color terms, one for light colors and
one for dark colors. Now, the point for Berlin and Kay in their commentary
about this passage is to argue against the earlier anthropological idea that the
Murray Islanders (or any other so-called primitive peoples) lacked the abil-
ity to perceive colors as a European does and to argue for the idea that their
naming practices were one thing, but their perceptual abilities another. As Kay
later wrote: “Languages differ semantically but not without constraint, and …
linguistic differences may induce nonlinguistic cognitive differences but not so
absolutely that universal cognitive processes cannot be recovered under appro-
priate contextual conditions” (Kay and Kempton 1984: 77). From my point of
view, the passage underscores rather the importance of saturation over hue in
certain color naming practices.
More recently, Kay has made explicit the cultural context for the evolu-
tion of color terms that was implicit in Berlin and Kay all along. The relevant
dimension is technological complexity. In a technologically simple society,
color is more predictable, since it is rare that naturally occurring objects are
distinguishable only by color. In a technologically complex society, however,
“artifacts are frequently to be told apart only by color. The limiting case is
Color 211
perhaps color coding, as used in signal lights, electric wires … almost every
kind of material thing we encounter in daily life – clothing, books, cars,
houses – presents us with the possibility that two tokens of the same type will
be distinguishable only, or most easily, by their colors” (Kay and Maffi 2000:
756). This explains why color terminology has evolved in some societies and
not in others, but it still does not address the differing experiences of color
that members of those societies accumulate as they realize themselves through
their languaging together and organize themselves around their skein-sorting
or clothing-book-and-car-identifying activities.
Deacon, for one, acknowledges a neurological bias acting as a relentless
selective force in social evolution and states that these biases “underlie the uni-
versality of color naming.” At the same time, he also acknowledges that we are
not biased to see colors as discrete, nor are we biased:
only to name certain points of the spectrum (the advertising and paint industries have
proven that thousands of color distinctions can be given individual names), but rather
that this will bias the co-evolution of color names. The patterns of errors in use and
transmission of color terms will be biased like a loaded die, so that over time the lin-
guistic references will converge to match the neuropsychological foci of perceptual
experience. (1997: 119)
Levinson follows in Deacon’s path when he writes that treating color terminol-
ogies as being in a push-pull relationship between neurophysiological biases
on the one hand and cultural pressures on the other has three advantages: (i) it
allows for exceptional cases, like Russian light versus dark blue and Hungarian
with two basic terms for red; (ii) color terms and technological complexity
come into alignment; and (iii) “the long, slow emergence of colour terminolo-
gies, sometimes along exceptional trajectories, can be explained as the fixation
of perceptual biases in cultural tradition over time” (2003: 321). Once again,
in a developmental systems account, one does not have to sacrifice all sense
of same in order to account for the particular, nor to subtract particularities in
order to ground a universal theory of human perception. What colors people
think do and do not go together are a result of that push-pull relationship in
ontogeny, and that push-pull relationship will result in different experiences for
differing languaging groups.
It could be that the more researchers work with sophisticated non-invasive
techniques for measuring perceptual and conceptual differences that correlate
with language differences, the better we will come to understand the subtle-
ties involved in linguistic relativity. The first demonstration of a relationship
between native language and unconscious, preattentive color discrimination
was published by Guillaume Thierry et al. in 2009, and this study distinguishes
itself from other studies of color categorization that have relied on conscious
and overt behavior. In Thierry et al.’s experiment, they compared speakers of
English and Greek on the early stages of visual perception using the vMMN,
an electrophysiological index of perceptual deviancy detection. Greek has two
color terms – ghalazio and ble – distinguishing light and dark blue. The vMMN
findings show a greater distinction between different shades of blue than differ-
ent shades of green in Greek participants, whereas English speakers show no
such distinction. The findings reveal not only an effect of the native language
on implicit color discrimination as indexed by preattentive change but even,
as the authors point out, “electrophysiological differences occurring as early
as 100 ms after stimulus presentation, a time range associated with activity in
the primary and secondary visual cortices” (2009 p. 3 of 4). Russian similarly
distinguishes light blue, goluboy, from dark blue, siniy. Russian speakers are
faster than English speakers to discriminate two colors when they fall into dif-
ferent linguistic categories in Russian (one siniy and the other goluboy) than
when they were from the same linguistic category (both siniy or both goluboy)
(Winawer et al. 2007). A lifetime of distinguishing among things one way or
the other does, indeed, seem to create perceptual and conceptual differences.
Time
In “The Relation of Habitual Thought and Behavior to Language,” first pub-
lished in 1941, Whorf was interested to answer the question: “Are our own
concepts of ‘time,’ ‘space,’ and ‘matter’ given in substantially the same form
by experience to all men, or are they in part conditioned by the structure of par-
ticular languages?” (1956: 138). He takes up the question because the category
time cannot be experienced or apprehended directly, that is, in and through our
perceptions and/or practical dealings with the world, the way that the category
space is (but see below and the next section on space). Since the experience of
time is possible only in and through language, it made an excellent category
for Whorf to investigate possible cultural/behavioral differences that might co-
vary with linguistic differences.
In “An American Indian Model of the Universe,” first published in 1940,
Whorf maps out the contrast between the Standard Average European (SAE)
conception of time and the Hopi. In Hopi metaphysics, Whorf identifies two
grand cosmic forms that he calls manifested and manifesting (or unmanifest).
The objective or manifested “comprises all that is or has been accessible to the
senses, the historical physical universe, in fact, with no attempt to distinguish
between present and past, but excluding everything that we call future.” The
subjective or manifesting, on the other hand:
comprises all that we call future, but not merely this ; it includes equally and indis-
tinguishably all that we call mental – everything that appears or exists in the mind, or,
as the Hopi would prefer to say, in the heart , not only the heart of man, but the heart
of animals, plants, and things, and behind and within all the forms and appearances of
Time 213
nature in the heart of nature … and, so charged is the idea with religious and magical
awesomeness, in the very heart of the Cosmos, itself. (1956: 60–61)
The manifesting (or unmanifest) thus includes not only the future but all of
mentality, intellection and emotion: the essence and typical form of which is
the striving of purposeful desire, intelligent in character, toward manifesta-
tion – as Whorf tells us – a manifestation that is much resisted and delayed but
is nevertheless inevitable.
Two verbal categories reflect this worldview, the inceptive, referring to
the edge of emergent manifestation belonging to the manifested side, and
the expective, referring to the edge of emergent manifestation belonging to
the unmanifested side, that which is beginning to be done but is not yet in
full operation. The inceptive, interestingly, implies the ending of the work of
causation at the same time that it states the beginning of manifestation. For
instance, when a phrase that has a suffix which corresponds to our passive, e.g.,
“the food is being eaten,” receives an additional inceptive suffix, we have the
beginning of manifestation and the end of causation. The translation would be
something like, “The food stops getting eaten.” Without knowing the underly-
ing Hopian metaphysics, there would be no way of distinguishing the inceptive
from the expective in translations, and it would be impossible to understand
how the inceptive could denote both starting and stopping.
The Hopi term tunátya is often translated as “hope” but also contains, accord-
ing to Whorf, our ideas of “thought,” “desire,” and “cause.” It is, in effect,
the Hopi word for the manifesting realm: the subjective, unmanifest, vital and
causal aspect of the cosmos, the mental-causal activity that is forever pressing
upon and into the manifested realm. If one adds the inceptive suffix to the term
to produce tunátyava, we get “comes true, being hoped for.” Tunátyava is thus
the term for the manifested realm.
The Hopi language has no word that would be the equivalent of the SAE
concept time. If there were to be a way to describe how things proceed or come
into being, my reading of Whorf’s description suggests an activity something
like the following: Imagine a field of newly planted corn. Go to the heart of
each corn seed/budding stalk. Fix your imagining around a vertical axis from
the center of the earth to the skies. Put your own heart and mind and everyone
else’s heart and mind and hope and striving into the coming-into-being of the
corn. Run the whole through a very fast time-elapsed camera that implies a
kind of horizontal axis. Now, take away the horizontal axis. Then take away the
vertical axis. You are in the realm of the expective and tunátya.
When Whorf compared the Hopi system with SAE, he noted that Hopi lacked
anything resembling an SAE tense system. In addition, they classified events
by duration, where “events of necessarily brief duration (lightning, wave,
flame, meteor, puff of smoke, pulsation) cannot be anything but verbs.” Whorf
furthermore noted that Hopi had evidentials, that is, morphological indications
of the utterer’s attitude toward the utterance, namely whether it is a statement

of current fact, a statement of fact from memory, a statement of expectation,
a statement of generalization, and so on. Whorf correlated these grammatical
features with an outlook on life that he deemed to be timeless and ahistorical
in the sense that past, present and future are seen as a continuity of duration,
experience being cumulative and unchanging for countless generations. For
Whorf, then, “As a result of the ‘timelessness’ of Hopi life, it is of greater
importance for Hopi speakers to distinguish between the duration of events and
their certainty than to indicate when they occurred” (Fishman 1960: 332).
Whorf contrasts Hopi metaphysics with that of the SAE notion of time
as a smooth-flowing continuum that can be cut up into a past, a present
and a future. Hopi “time” does not represent itself as kinematic motion but
rather as an accumulation of what he called a vitality or spirit (heart/mind/
thought) that proceeds from a so-called unmanifested realm into the so-
called manifested realm. Whorf relates the Hopi propensity for preparing
with an emphasis on persistence and constant insistent repetition – e.g., in
planting, harvesting or hunting ceremonies – to their linguistic background
that disposes them to understand that what manifests itself “in the present”
is a result of all that has come before, including both the overt and covert
support (i.e., intentions) that the group has put into bringing forth the mani-
fested “present.” Whorf notes that the Hopi would understand our proverb
“Well begun is half done,” but not our “Tomorrow is another day” (1956:
148). There is no “turning over a new leaf” for the Hopi, since whatever
comes into manifestation at any moment always and continually carries for-
ward all that occurred in the “past.”
The SAE notion, in contrast, disposes its languagers (my term, not Whorf’s)
to treat time as an object that is available to be partitioned and put in con-
tainers. He relates our linguistic background to the SAE propensity for: (i)
records, bookkeeping, diaries, accounting, mathematics; (ii) interest in exact
sequence, dating, calendars, chronology, clocks, time wages, time graphs, time
as used in physics; (iii) annals, histories, the historical attitude, interest in the
past, archeology, attitudes of introjection toward past periods, e.g., classicism,
romanticism (1956: 153). In extending this Whorfian analysis, we could say
that SAEers think nothing of manipulating time in the sense that “time out”
exists as a feature of many sports. A Hopi might ask, “How it is possible to
stop such a thing?” The answer would be: “Well, since we decided that the
game was going to be an hour in length, we wanted to make sure that all sixty
minutes were devoted to actual play. And that makes it necessary for many
reasons that arise within the game to, in effect, stop time, i.e., turn the clock
off.” In many universities, a tenure clock may be suspended, especially when a
woman’s biological clock is ticking, because the latter is not as easily manipu-
lated as the former.
Time 215
THOUGHT
GENES LANGUAGE
FLOW
BLOOD MONEY
WATER CURRENCY
TIME
HISTORY
Figure 7.1 The flow paradigm for Standard Average European
Going further, we could say that for SAEers, the idea of time as a smooth-
flowing continuum – one that arises out of the past, moves through the present
and ribbons on into the future – resides within a paradigm of things/events/
constructs that are similarly perceived as flowing. This paradigm organizes a
significant swath of SAE culture. In considering things that flow in SAE (here
English), we find entities depicted in Figure 7.1.
Just as water (and any liquid) can be said (and seen) to flow, so time and
history flow, and we speak, consequently, of historical currents and, e.g., new-
wave cinema. Money has a flowing property as well, and we all have experi-
enced (often problems with) cash flow. Money in English is called currency,
and in French is argent liquid. (In Romanian, it is bani-gheaţa, “money-ice,”
which has the sense of hard cash; here we would say that this formulation
reflects Romanians’ recent historical lack of experience with flowing cash and
their continuing experience of lei as soft, that is, as a currency that does not
survive a border crossing.)5 A liquidation sale is one where the proprietor of
an establishment has too much solid stock on hand and perceives the need to
sell it so he can get back into a situation of positive cash flow. Accounts can, of
course, be frozen, thereby preventing any kind of flow whatsoever. In addition,
we can do the same things with money that we do with time: we can spend it,
save it, invest it, waste it; they share a similar attribute in that both are valu-
able. The similarities are so close that we have a phrase Time is money, and we
do not perceive it as metaphoric. It is rather the reality of everyday life. There
exists, for instance, a government-decreed hourly wage. Our entire banking
and economic systems, not to mention how we fill out our tax forms, depend on
time–money complexes such as interest rates, appreciation and depreciation.
Now, just as money flows, so we say that a language one knows well is one that
5
Of course, English has the phrase cold, hard cash to contrast with a promissory note, IOU, or
any other token or barter exchange, or even a check that either will take time to clear or could
bounce.
is spoken fluently. Money and language have a long history of being compared
in the Western tradition as exchange systems, and we can and often do coin a
phrase. Thought comes into the picture in the phrase stream of consciousness
(coined, incidentally, by William James). And finally, we return to the begin-
ning, namely to physical bodies, between and among which genes are said to
flow the way that visible blood does in an individual body. Gene flow creates
gene pools.
Whether or not the above flow model can withstand close scrutiny as a mas-
ter modeler for Western cultural life, the overall point here is that Whorf was:
(i) concerned with the ways that large-scale linguistic patterns correlate with
habitual thought patterns and with behavioral and cultural norms; and (ii) inter-
ested in habitual behavior. Clearly, making good on the first feature is difficult,
since the researcher must master more than one language and culture in order
to observe and describe systematic differences. The second feature is easier
to deal with, and Whorf’s emphasis on habitual behavior is quite consistent
with the ontogenic script sketched in this and the previous chapter, because
our approach predicts that languagers living in different languaging communi-
ties will think (however deeply or superficially we wish to construe the term)
and behave differently, that is, in accord with the norms of their communities,
unless asked to do otherwise.
The question of habitual thinking and conceptions of time is at issue in “Does
Language Shape Thought? Mandarin and English Speakers’ Conceptions of
Time” (2001) by Lera Boroditsky. Her article investigates the relationship
between a particular language and its effects on how one tends to think about
time. For instance, in English, horizontal metaphors are used for time as well
as for space (“good times ahead, hardships behind” // “full speed ahead, don’t
lag behind”), while in Mandarin there are two possibilities: in addition to hori-
zontal metaphors for time as well as for space, there is also the systematic use
of vertical metaphors to talk about time. The spatial morphemes shàng (“up”)
and xià (“down”) are frequently used to talk about the order of events, weeks,
months, semesters and more.
Boroditsky devised three experiments that compared reaction times between
English speakers and Mandarin speakers in processing horizontal versus verti-
cal statements about time. She showed that, although all the experiments were
conducted in English, the Mandarin speakers were still “thinking in Mandarin”
while processing English instructions. For instance, Mandarin speakers were
faster to confirm that March comes earlier than April if they had just seen
a vertical array of objects than if they had just seen a horizontal array, and
the reverse was true for English speakers. Boroditsky notes that: “Overall,
Mandarin speakers who learned English later in life were more likely to think
about time vertically. The propensity to think about time vertically was related
to the length of pure Mandarin experience (before any English was learned),
Time 217
but not to the length of English experience” (2001: 16). She readily acknowl-
edges that certain cultural factors could account for the differences, e.g., writ-
ing direction, where English is written horizontally from left to right, while
Mandarin is traditionally written in vertical columns that run from right to left.
However, she was able to establish that writing direction does not account for
all the data of Mandarin speakers. She was also able to establish that the effect
of language is not absolute: English speakers who were told that they would
learn “a new way to talk about time” and who were trained with vertical meta-
phors produced results that looked more like those of Mandarin speakers than
those of untrained English speakers. In this respect, Boroditsky’s conclusion is
not different from that of Hespos and Spelke, and it is evidently true that lan-
guaging living beings at almost any stage of development can and do recognize
conceptual distinctions made by other groups of languaging living beings.
However, it does not follow that all distinctions are innate, since the cat-
egory innate is not conceptually coherent. Nor does it follow that individuals,
having once recognized or had pointed out a distinction not made in their cus-
tomary language, are able to immediately redirect their attention and/or verbal
behavior with respect to that distinction in accord with the norms of the other
group – a fact to which every foreign language teacher in the world can attest.
Boroditsky’s research confirms this last statement, when her results showed
that Mandarin speakers were still said to be “thinking in Mandarin” when pro-
cessing English instructions.
Overall, Boroditsky understands her work as supporting a weak version of
the Sapir-Whorf hypothesis or linguistic determinism (see also Davies and
Corbett 1997). This weak version might correspond to a level-two relativism,
because Boroditsky is not attempting a large-scale cultural/behavioral correl-
ation (except for the differences in writing direction), but she is able to tease
out certain processing-time differences between two different languaging
communities.
In answering the question “how do we become the particular languaging liv-
ing beings that we are?”, I want to open a space for differences in the ways that
individuals may live their lives in the so-called same community. Traditional
linguistics certainly makes a space for the concept of idiolect, often defined
as an individual’s way of speaking. For an ontogenic script in a developmen-
tal systems framework, there must also be room for an individual cognitive
fingerprint.
For a number of years now, I have asked students to draw their vision for the
sequence of the months of the year with the question: what does your visuali-
zation of one month after the next look like? I also asked them to sketch their
vision of the sequence of years with the question: Where are 1980, 1850, 1776,
0, 100 BC, etc. in relationship to one another? For as long as I can remember,
my own image of the twelve months of the year is superimposed on a clock
face, such that 12:00 is where midnight December 31 becomes January 1. The
sequence of months is clockwise, with February 1 corresponding to 1:00. I did
not even realize that this was not a cultural norm until a few years ago when,
somehow, the topic came up in a seminar I was teaching. (I always assumed
that my vision of the sequence of years was completely idiosyncratic, because
it is a wavy stream with particular bends at particular years.) There were six
of us sitting around the table, and all six of us had different images of the
sequence of months. Upon learning that my image was not a cultural norm, I
asked my sister to draw her picture of the way the months follow the months
in her head. She drew a clock, as did my brother. Obviously, the three of us got
the image from the same source, but we have no idea what it is. (I suspect it
was my father’s image.) Since that seminar, I always ask the students to draw
a picture of their mental images.
Despite the many representations of the sequence of time that abound –
e.g., grade school timelines and calendar pages flying off the wall as if blown
away by the wind, as depicted in old Hollywood movies – the way the pas-
sage of time is pictured in any one of our heads ranges from slightly to highly
idiosyncratic. One person I polled drew a clock face as I had done, but she
put September at 12:00, and the movement of the months for her is coun-
terclockwise. Another has a complicated 3-D elevator image that moves in a
zigzag fashion inside a large “time” structure. Another has the summer months
off on a bubble to the side. Another has two columns, with January–June on
the left and July–December on the right, just as she had seen it posted on a
grade school wall. Yet another has himself situated inside a solar system which
revolves around him. And so forth.
Some patterns do, however, emerge: (i) a relatively common picture for those
with European language backgrounds is of a horizontal straight line, extend-
ing indefinitely in both directions, with the Now marked out at some midpoint.
There also tends to be a left-to-right orientation, with the past lying to the left of
Now and the future to the right; and (ii) Chinese-American students I have had
in my classes, especially ones whose first language is some variety of Chinese
(usually Mandarin), often have a vertical view of the twelve months of the year,
with the sequence of years moving from right to left. There do not seem to be any
differences that line up according to gender. One young woman who is native
speaker of Mandarin and Arabic with near-native proficiency in English drew a
combination of the horizontal and vertical: January, February, March and April
ran horizontally left to right, then May and June went vertically below April,
while July and August ran horizontally right to left off of June, then September,
October, November and December fell vertically again below August. The
whole was not symmetrical by any means. The picture was interestingly lop-
sided toward the spring months. The young woman laughed when she realized
that she had hedged her bets between the vertical and the horizontal.
Space 219
So, despite the fact that we live in a culture that is very invested in record-
ing and representing time – analog clocks, digital clocks, agendas, day tim-
ers, planners and calendars are seemingly everywhere – that is to say that the
abstract notion time is scaffolded by a quite a lot of visual representation in
our daily lives, and despite the fact that languaging living beings in our culture
(and in a good bit of the world these days) tend to devote quite a lot of languag-
ing to the topic of time, individual mental representations of this conceptual
terrain can be extremely varied.
It, therefore, does not come as a great surprise that at least one group of peo-
ple, namely the Aymara in the Andes highlands, has a concept of the future that
seems to be metaphorically behind, whereas the past appears to be in front. This
seems to be an extremely rare case, since most languages construe the future
in front and the past behind. (I do not know where the Hopi fit in this.) The
Aymara system may also be an areal feature (Núñez and Sweetser 2006: 404).
The point here is that, given the range of individual idiosyncrasy that exists
alongside some basic general cognitive schemas that are widespread in a cul-
ture and may also be found worldwide, the Aymara linguistic conventions seem
to have organized themselves around an (some individual’s?) idiosyncrasy.
Space
Whorf believed that the category of time is directly affected by the languag-
ing environment through which one lives. However, about the category of
space, Whorf writes: “There is no such striking difference between Hopi and
SAE about space as about time, and probably the apprehension of space is
given in substantially the same form by experience irrespective of language.”
However, he adds, “But the concept of space will vary somewhat with lan-
guage, because, as an intellectual tool, it is so closely linked with the concomi-
tant employment of other intellectual tools, of the order of ‘time’ and ‘matter,’
which are linguistically conditioned”(1956 [1941]: 159). Whorf’s distinction
between apprehension of space versus concept of space may well be a play on
nature/nurture with apprehensions referring to preexisting cognitive endow-
ments and concepts being given through language.
There is no need to spend a lot of energy on the contemporary approach
to the issue. In Language and Space, edited by Paul Bloom et al., the edi-
tors acknowledge in their final co-authored chapter that a culture may influ-
ence how one refers to spatial attributes and even which spatial attributes are
referred to, but it will not alter in any fundamental way spatial understand-
ing, i.e., the ways “primitive CRs,” or pre-given “conceptual representations,”
process spatial information. They state in the form of an assumption that
“experience does not fundamentally alter perceptual and cognitive processes”
(1996: 571–72). When they state that “It is important not to overemphasize
the differences between speakers of different languages” (1996: 571), the

question must be asked: why? (Well, of course, we wouldn’t want to overem-
phasize, since this already implies inappropriate emphasizing. But, how about
we “simply” emphasize the differences, appropriately?) Emphasizing the cog-
nitive spatial differences appropriately is exactly what Steven Levinson has
done, most remarkably in Space in Language and Cognition: Explorations in
Cognitive Diversity (2003).
Whorf did not pursue possible linguistic/cultural divergences in the category
of space in the same way he did with time, and Levinson’s work adds a signifi-
cant corrective to the notion that space is, somehow, organized in human heads
independently of the languages those heads move in and through. He focuses
on distinctions on the horizontal plane – and with good reason. Although per-
ceptual cues for the vertical may not always coincide, Levinson tells us, “they
overwhelmingly converge, giving us a good universal solution to one axis. But
the two horizontal coordinates are up for grabs: there is simply no correspond-
ing force like gravity on the horizontal” (2003: 35). Thus, as might be expected,
languages diverge in their solutions for how to specify angles or directions on
the horizontal. Levinson shows how the various solutions that have been found
across the world’s languages affect their languagers in more profound ways
than has been previously suspected.
The point of Levinson’s thorough investigation of a wide range of languages
is to question what seems to be the logic, given the importance of spatial cogni-
tion to many aspects of survival, that it is encapsulated in a module. He shows,
in rather convincing fashion, that the predominant frame of reference in a lan-
guage – intrinsic, relative or absolute6 – has surprisingly far-reaching cognitive
effects: on mental imagery, memory, inference, navigation and gesture. Absolute
ways of thinking, for instance, make for good dead-reckoning abilities, keep a
mental compass constantly running and tend to produce individuals who decou-
ple spatial memory from view-centered systems of vision and motoric action. In
contrast, relative ways of thinking make for good piloting (but bad dead-reckon-
ing) and an ability to conceive of “strip maps” (but not of overviews). Consistent
with everything that has been laid out this far, Levinson’s work provides valu-
able evidence that: (i) language may actually induce conceptual development
in a number of domains, including spatial cognition; (ii) particular languages
in and through which one languages may induce different kinds of cognitive
engagements in the world; and (iii) cognitive and linguistic environments do not
6
A relative frame is viewer-based, e.g., “the book is to my left.” An intrinsic frame is object-
centered, e.g., “the book is in front of the television.” An absolute frame is organized around
cardinal directions, e.g., “the book is to the north of the television.” All three frames have: (i) a
system of labeled angles; (ii) coordinates that are polar, with one primary coordinate system;
(iii) points of reference, ground, viewpoint, anchor point, etc.; and (iv) an anchoring system that
locks the labeled angles either to an anchor point or to a slope.
Space 221
exist in a vacuum but are always situated in a social/cultural context that will
be intertwined with some of the cognitive and linguistic particularities of those
languaging living beings who make their lives in it.
With regard to the first point, it was already made in the previous section on
number cognition, given that the conceptualization of precise number occurs
relatively early on in Western developmental systems but not in that of, say,
the Mundurukú. The second point takes within its purview the length of the
developmental system – here Waddington’s chreod from Chapter 2 might have
some use – and includes the idea that the varying conceptual developments
take place over a rather long period of time, many years in many cases. That
is, there is quite a difference between a three-year-old’s grasp of the concept
precise number and the cognitive activity that person will be engaged in some
eleven or twelve years later when taking the SAT (no matter what score is
received). Levinson makes clear that the acquisition of the frame(s) of refer-
ence in use in a particular languaging environment does not come all at once.
In European languages, the intrinsic frame of reference is clearly learned first,
but “such notions as intrinsic ‘in front of’ are not mastered in production before
the child is nearly four years … and intrinsic ‘left/right’ not fully mastered till
eight, while relative ‘front’/‘back’ comes into the child’s production just before
five, and relative ‘left/right’ must wait till as late as age eleven” (2003: 308).
Incidentally, the intrinsic frame of reference is “close to linguistic bedrock, in
that it is near universal” (2003: 81). What Levinson admirably demonstrates
is the way that the cognitive abilities that correlate with the various frames of
reference found in the world’s languages are constructed results of both socio-
historical and ontogenic development. In addition, Levinson deftly undoes
centuries of speculation about human spatial cognition, e.g., Kant’s contention
that the relative (viewer-based) system was the most fundamental.
The second point blends into the third, namely that the whole way of organ-
izing the cultural environment in egocentric coordinates, which correlates
with relative frame systems, makes possible many organizational founda-
tions of complex societies, from traffic rules to writing systems. The organi-
zation of cultural space in groups who language with an absolute system
may support and reflect that system by the way they organize their towns on
the landscape, orient their houses and place their front doors. The feedback
loops between spatial frame(s) of reference and cultural correlates extend to
all aspects of daily life, big and small, from unreflective gesture to the games
children play. Levinson mentions an aboriginal Australian language that uses
the absolute frame. In this community, there exists a kind of “blind man’s
bluff” that serves to amuse and educate children into the absolute system.
Children under eight are not expected to be able to play the game perfectly,
and it is with the help, in part, of the game that they learn to master the car-
dinal direction system (2003: 313).
In the Western tradition, many types of knowledge are not expected to be in

the heads of individuals. Instead individuals are trained in ways of finding that
knowledge, e.g., they are taught to read, to understand graphs, to use indexes,
etc. Take navigation. Levinson points out that while there has been much study
devoted to navigation in non-human species, there is not such a large literature
on human navigation. Of course, there is a vast literature about marine naviga-
tion, but “in our own culture this has been largely formalized since at least the
fifteenth century, and is the province of experts. Such expert knowledge is not
necessarily in the navigator’s own head, for it is an accumulated lore and sci-
ence, built into maps, instruments and procedures for their uses” (2003: 217).
The same may be said of mathematical knowledge. Not all of us want to or are
able to use the floating-point arithmetic now used in all computers that Lakoff
and Núñez describe in Where Mathematics Comes From (2000). Experts are
trained to use it over a lifetime – or a good portion thereof – and it may well
not be the case that everything that is known about floating-point arithmetic is
in any one person’s head.
The last page of Space in Language and Cognition relates to the heart of
Linguistics and Evolution when Levinson writes: “A language ‘canalizes’ the
mental landscape, offering complex concepts that would otherwise be mostly
out of the range of independent invention – complex concepts, like specific
instantiations of frames of references, that then come to dominate the internal
coding of states of affairs and events” (2003: 325). Although Levinson does not
invoke Waddington in this passage, it is worthwhile to remember Waddington’s
images of genetic canalization already discussed in Chapter 2.
As a reminder: canalization is a property of the developmental system that
has the capacity to reduce the effect of perturbing influences on the pheno-
type. The canalization of the phenotype explains how, in a given species, e.g.,
Drosophila, all individuals will look remarkably alike when, in fact, their geno-
types are remarkably different. Given all that Levinson lays out in his book, he
evidently does not mean to say on the last page that language as such canalizes
human mental capacities so that, despite linguistic diversity, human mental
landscapes end up looking remarkably the same the world over. However, in
the dimension of assessment comparing human cognition versus non-human
cognition and/or comparing the cognition of an adult human languager versus
an adult human non-languager, e.g., the well-known and unfortunate case of
Genie, we would say that all human adult languagers are similar in that they
have “languaging mental landscapes” while non-humans and human non-lan-
guagers do not. Now, Levinson apparently does mean that while all languages
have spread themselves out over and through human cognitive capacities, dif-
ferent languages offer different conceptual contours. It may well be the case
that some of those differences are small enough such that a mental pathway
that is pushed up a hillside for a while may find its way back down, farther
Summing up 223
along in development, to the valley of some human norm. However, it may also
well be the case that some of those differences create watersheds such that over
time, a mental pathway is pushed over a hill and finds itself running down to
the bottom of another valley. Sapir called these ruts or prepared grooves.
Because much of the renewed interest in linguistic relativism was prompted
by studies of bilingualism (Hunt and Agnoli 1991), it is interesting to consider
the ways that bilingualism itself induces or enhances certain neurological proc-
esses. For instance, it seems that older bilinguals are able to suppress distract-
ing information better than monolinguals because they have spent a lifetime
suppressing one language or another. Then, too, bilinguals seem to be more
easily able to reverse ambiguous figures, for instance the Necker cube, than
monolinguals (Bialystok and Shapero 2005). As it turns out, the language con-
trol processes engaged in contexts during which both languages need to remain
active involves the left caudate and the anterior cingulated cortex (Abutalebi
et al. 2008), a brain region that was earlier noted as playing a fundamental
role in the evolution of communication in primates. The neurological possibil-
ities become even more interesting when we consider bilingual languagers of
a spoken and a signed language. Given the signers’ need to adjust to whether
their interlocutor is left-handed or right-handed and given the strong prefer-
ence of the right hemisphere for spatial perception, signers’ right hemispheres
are sure to show enhanced spatial abilities that those of us who language only
with spoken languages lack. This is the point of a recent article by Evans and
Levinson to the effect that the diversity that the world’s language exhibits is
its most important property, and they caution that a great deal of work in the
cognitive sciences risks “being vitiated, at least if it purports to be investi-
gating a fixed human language processing capacity” (2009: 1). It is my hope
that the present study, with its understanding of languaging as an orienting
activity, offers a coherent and thorough alternative to a linguistics anchored in
Universal Grammar.
Summing up
We have covered an immense amount of territory in order to account for how
languaging living beings are constructed results of a long series of phylogenic
and ontogenic interactions. As a species, we can say that we languaging living
beings share more traits in common with each other than we do with non-
languaging living beings, but once we take a species-internal perspective, we
acknowledge that all of us languaging living beings are different one from the
others. When it comes to looking at particular subgroups, say, that of English-
speaking, relative-frame-of-referencing North Americans, it will again be the
case that we are all different from one another, but if we compare ourselves as a
group to those who language through an absolute frame of reference – roughly
about a third of all human languages from Mesoamerica, to New Guinea,

to Australia, to Nepal (Levinson 2003: 48) – we will find some consistent
between-group differences on a cognitive and behavioral level. The same can
be said for what colors we think go together; for some of the particular body
states we experience in the context of certain events, i.e., how we emotion with
respect to them; for how we conceptualize time; for certain ways of interact-
ing that we find polite or rude; for some of our eating preferences, which may
begin in the womb; for the ways that English speakers, who tend to conflate
motion and manner in the main verb, will also do so in gestures that accom-
pany those verbs, while speakers of languages that conflate motion with path
will not reveal manner in their gestures; and so on. To answer our questions we
have not needed to offer yet another putative object of study carved out from
among the welter of linguistic data. In fact, the answers to our questions have
depended on the idea that it is no longer useful to treat the construct language
as any kind of object whatsoever, even a theoretical one.
But what exactly is meant here by the term thinking, however deeply or
superficially we wish to construe the term? A verb that tends to turn up in these
discussions is the verb to attend. The linguistic anthropologist Dan Slobin
offers what amounts to a definition of attending in his article “From ‘Thought
and Language’ to ‘Thinking for Speaking’”:
In my own formulation: the expression of experience in linguistic terms constitutes
thinking for speaking – a special form of thought that is mobilized for communication.
Whatever effects grammar may or may not have outside of the act of speaking, the sort
of mental activity that goes on while formulating utterances is not trivial or obvious, and
deserves our attention. We encounter the contents of the mind in a special way when
they are being accessed for use. That is, the activity of thinking takes on a particular
quality when it is employed in the activity of speaking. In the evanescent time frame
of constructing utterances in discourse one fits one’s thoughts into available linguistic
frames. “Thinking for speaking” involves picking those characteristics of objects and
events that (a) fit some conceptualization of the event, and (b) are readily encodable in
the language. I propose that, in acquiring a native language, the child learns particular
ways of thinking for speaking. (1996: 76)
Attending, then, can be thought of as thinking for speaking. In order to formu-

late any utterance in any language, one’s attention is drawn in this direction,
rather than that direction, one is under this pressure or constraint to notice or to
ignore certain features of the situation rather than that feature of the situation.
One attends selectively, with the selections based on the categories available to
the languager in the language through which s/he is languaging.
Languaging in a language thus directs/induces/requires languagers to attend
to certain features of their environment rather than others, to take certain
perspectives rather than others. Koreans will regularly attend to tight–loose
distinctions and Chinese will tend to think of time vertically, while English
Summing up 225
languagers will regularly attend to containment-versus-support relationships

and tend to think of time horizontally. Yes, we can all switch perspectives
if the need arises – if we are subjects of a psycholinguistic experiment, if
we move to another country and want to learn the language – but we do not
switch perspectives as a rule because speaking and thinking of something,
e.g., time, in whatever fashion we are used to, e.g., in vertical terms, can be
said to work for us.
When we take a developmental perspective and place our brains in our bod-
ies, and our bodies in our worlds, we see that particular ways of attending
to aspects of our environment are often supported by other features of our
environment and are thus reinforced on an ongoing basis. We may be sur-
rounded, for instance, by other languagers who also language about time in
vertical terms. We may typically write vertically, and we may be used to using
calendars where March is always seen above April. We have hardly any reason
to deal with time other than vertically. The word conspires easily enters the
picture when we begin to see that particular features of a language and particu-
lar features of the languaging environment seem to be in cahoots. Those who
experience the consistent congruence of their linguistic and physical environ-
ments experience it as reality, and we say that those who deal effectively with
it display common sense. It is often only through a comparison with a con-
trasting congruence of linguistic and physical environments that one comes to
see, as did Sapir, that “the worlds in which different societies live are distinct
worlds, not merely the same world with different labels attached.”
The question arises: do the Pirahã attend to numerical distinctions above
the number two and the Mundurukú above the number five? It seems that they
do not. They have no reason to, and nothing in their environment supports
such distinctions. If asked, could they attend to such distinctions? The answer
depends, in part, on how far along they are in their ontogenic drift. The farther
along they are, the less they are able and the less relevant it is for them to make
distinctions such as “seven fish” or “twelve trees.” The answer also depends,
in part, on whether any other aspects of their daily lives will change to support
the numerical distinctions.
Although evolutionary biologists and neurobiologists are still exploring the
ways that humans are realizing their sapien (and homo-hominid-primate-mam-
malian-etc.) selves, enough is known about human neurology and biology with
respect to our languaging behaviors to make basic statements about it possible.
From there it is possible to describe the ways that a human baby comes into
languaging by relying on the most current usage-based accounts that detail
how and when children begin to respond to and produce the many patterns
that linguists have so carefully described over the decades. And it continues, as
we have just seen, into areas traditionally covered under the rubric of linguis-
tic relativity, where the universal versus relativity debates are more usefully
recast in a developmental systems framework, which requires researchers to

have enriched knowledge of the cultural context in which particular languaging
behaviors occur. The developmental systems linguistics outlined here does not
make an effort to be theoretically ecumenical. Instead, it advocates that lin-
guists pursue cognitive accounts of languaging in the brain in both an improved
psychological framework – namely, one that actually takes psychology into
account – and a cultural framework that exceeds by far information found in
traditional grammars. The developmental systems linguistics suggested here
has tried to offer both a historiographic psychoanalysis of the discipline and
a constructivist account of our subject matter so that a linguistic account of
languaging as an orienting behavior may be more fully integrated into work
in the psychological, neurological, biological and evolutionary sciences and,
perhaps, inspire researchers in those fields to new perspectives on what they
think they are investigating.
Part III
What to do next
8 Revisit: Skinner, Chomsky and
construction grammar
One of the most fascinating episodes in the relatively recent history of linguis-
tics involves the publication of Verbal Behavior by the American behaviorist
psychologist B. F. Skinner in 1957 and Chomsky’s review of it that appeared
in the Linguistic Society of America’s main journal Language in 1959. A less
circumspect way of describing this episode is to say that Chomsky’s review of
Verbal Behavior caused a theoretical and methodological trauma in linguistics
from which the discipline is still recovering. In short, Chomsky (1959) had the
immediate effect of exiling the consideration of the circumstances that contrib-
ute to the production of any given utterance from so-called hard-core linguistic
research, thereby keeping so-called sociolinguistics and discourse analysis on
the soft sidelines for many years and the tradition of languaging as an orienting
behavior completely out of bounds. One purpose of this chapter is to reattach
the piece that Chomsky broke off, so that we can move on from a new whole.
At the time Chomsky wrote his review, he had published his first monograph
Syntactic Structures only two years earlier and was a relative newcomer to the
field, while the well-published Skinner was undoubtedly the most well-known
American psychologist of the era, behaviorist or otherwise. With his review,
Chomsky made an almost instant name for himself in linguistic circles as the
champion of the new rationalism and innatism. He succeeded not only in cre-
ating the impression that he had completely dismantled Skinner’s account but
also in demonizing the man and his lifework in the linguistic community. The
disparagement and misrepresentation of Skinner’s work lives on. Take, for
instance, Foundations of Language: Brain, Meaning, Grammar, Evolution by
Ray Jackendoff, who is interested to acknowledge the contributions of numerous
non-transformational approaches to generative grammar that developed during
the 1980s and 1990s and to integrate them into a reformulated account of the
field for the twenty-first century. However, he is not yet ready to acknowledge any
contributions from the field of behaviorism, as the following passage shows:
Various approaches to meaning seek to treat it as something other than a mental phe-
nomenon, as part of the “environment.” Perhaps the most extreme is behaviorism, now
mostly extinct, which claimed (e.g., Watson 1913) that thinking is nothing but subvocal
speaking, and that the idea of a “concept” behind the language is nonsense. However,
229
230 Revisit: Skinner, Chomsky and construction grammar
behaviorism never attempted to explain more than the most trivial of linguistic facts,
and those in dubious fashion (Chomsky 1959). (Jackendoff 2002: 280)
Chomsky (1959) is, of course, Chomsky’s review of Verbal Behavior, and
Skinner’s name goes unmentioned. With Skinner’s name “under erasure” –
or sous râture in the Derridian phrase – there is no room to note that, pre-
cisely on the topic of meaning, which is the subject of Jackendoff’s passage,
Skinner was highly critical of the position taken by the early twentieth-century
behaviorist John B. Watson – not to mention that Skinner was highly critical
of many aspects of Watson’s behaviorism in general. In addition, of the many
things one can say about Verbal Behavior, one must say that it is considerably
more than an attempt to explain “the most trivial of linguistic facts.” And as
far as Chomsky’s review goes, Chomsky may have asserted the dubiousness of
Skinner’s account, but he in no way demonstrated or proved it.
An additional purpose of this chapter is to identify and honor the contribu-
tions of both Skinner and Chomsky in order to have a view of a whole that we
want to move on from as well as a view to where we would like to go in the
arena of syntax. Since Skinner is presumably the tougher sell to linguistics
audiences, it is necessary, before attempting to suggest how his work on ver-
bal behavior might speak to linguists today, to address the question: how did
Skinner’s Verbal Behavior get buried in such a deep grave in the first place?
Skinner’s exile
From the beginning, and as mentioned earlier, Chomskyan linguistics has had
a kind of Bauhaus aesthetic appeal. Like modernist Bauhaus design, genera-
tive analyses are stripped of ornamentation and operate in the hushed world of
the idealized speaker–listener in a streamlined, dust-free speech community.
Syntactic derivations look crisp and scientific on a page, their neat sentential
triangles as perfectly Euclidean as Bauhaus architecture. The early transforma-
tions and then principles and parameters were once valued for their elegance
and economy, with the Minimalist Program arising in the 1990s when it came
to be seen that “economy considerations could have a larger role than previ-
ously assumed” (Chomsky 2002: 95). Within the framework of formalist lin-
guistics and now minimalism, the study of the clean, cool, finished monologic
utterance continues to be an art. By comparison, Verbal Behavior, with its rush
of details, its humor and its eccentricities, must have seemed cluttered and
inelegant to the young Chomsky.
Skinner was willing to follow his subject matter wherever it took him, and
it took him into discussions of advertising, autism, guessing, logic, lying,
poetry, rhetorical strategies, translation, verbal games and puzzles, verbal ritu-
als including prayer, in addition to grammar and syntax, and not forgetting to
mention discussions of problem solving, remembering and thinking, which are
Skinner’s exile 231
for Skinner behaviors like any others. Now, of all the criticism leveled against
Skinner from linguistic quarters, no one (as far as I know) ever accused him
of bad style – and he does have interesting things to say on the subject of
style, as well. However, Verbal Behavior is a frankly noisy book, resounding
with exclamations of Fire! and Water! and illustrated with a welter of lively
examples. The point here is that the aesthetic feel of Verbal Behavior has lit-
tle in common with a Bauhaus building and much in common with, say, the
Paris Opera with its sculptures and gargoyles, its cornices and scrollwork. This
Beaux-Arts masterpiece has no scale because it has every scale, just as Verbal
Behavior has every scale: from a major rethinking of the conception of “lan-
guage” to, indeed, the most “trivial” observations, e.g., Skinner’s comment on
the “delight” of “a good palindrome – for example, A man, a plan, a canal –
Panama” (1957: 292).
Style is never superficial; it cannot be modified, and thereby dismissed, by
the adjective mere. I am often entertained when I open Verbal Behavior at ran-
dom and skim a passage or two. I find Skinner’s style to be very playful and his
approach to be welcome relief from Chomsky’s unrelenting seriousness. Let
us take the phenomenon of ambiguity as a case in point. One of the early argu-
ments in favor of the analysis proposed by transformative generative grammar
was its ability to demonstrate simply and elegantly the differences between
pairs of sentences out of context that exhibit structural ambiguity, e.g., The
chickens are ready to eat and Flying planes can be dangerous. Skinner has an
original way to think about ambiguities, which he discusses in a chapter enti-
tled “Multiple Causation,” one example of which demonstrates how the same
words may have different effects on different audiences:
A distinguished scholar used to acknowledge complimentary copies of books by writ-
ing immediately to the author: I shall lose no time in reading the book you have so
kindly sent me. With respect to the audience of which the author was a member, this was
synonymous with I am anxious to read your book or I am going to read your book as
soon as possible. With respect to an audience, of which the scholar was himself a mem-
ber, it was synonymous with I shan’t waste my time on such stuff. (1957: 232)
Skinner notes that playing to two audiences is also known as irony. I am not
saying that this observation is laugh-out-loud hilarious, but it did provoke in
me a slight smile. The point is that Skinner and Chomsky have very different
personalities, and there is no doubt that Chomsky did not like Skinner’s.
Skinner’s and Chomsky’s styles are also worth talking about since they
express not only the deep contrast in cognitive taste that exists between these
two major thinkers but also the contrast in disciplinary orientation between
twentieth-century structural linguistics and behaviorist psychology. When
Chomsky declared linguistic theory to be concerned with “an ideal speaker–
listener, in a completely homogeneous speech-community” whose know-
ledge is “unaffected by such grammatically irrelevant conditions as memory
limitations, distractions, shifts of attention and interest, and errors (random or

characteristic)” (1965: 3), he was, perhaps unconsciously, echoing Saussure’s
pronouncement that “An absolute state is defined by the absence of changes,
and since language changes somewhat in spite of everything, studying a lan-
guage-state means in practice disregarding changes of little importance, just as
mathematicians disregard infinitesimal quantities in certain calculations, such
as logarithms” (1959 [1916]: 101–02). By way of sharp contrast, for Skinner
there were no changes of little importance or “irrelevant conditions,” since the
study of the changes, however big or small, in the context of their conditions
was the very stuff of his research.
From a vantage point that was unavailable when Saussure was writing and
not yet commonplace when Chomsky began, we are now aware that mathema-
ticians no longer ignore changes of little importance and that physicists and
meteorologists, among others, have come to understand that so-called irrele-
vant conditions are anything but when it comes to explaining certain natural
phenomena; and these scientists have revised their theoretical orientations
accordingly. It is only the philologist who has nothing to do with a language
except study it who can judge, post hoc, which conditions are (or have become)
irrelevant for a particular group of speakers at a particular time. The unweav-
ing of the Skinner/Chomsky episode is perhaps most useful in providing an
opening for reweaving a linguistics in which individual variation in the social
setting is a built-in ingredient of the discipline, rather than an add-on – a point
which should make variationists happy.
Verbal Behavior is one of the most elaborated accounts of our subject mat-
ter to be taken from the perspective of languaging as an orienting behavior.
Although Skinner himself saw his work on verbal behavior as standing com-
pletely alone, it shares family resemblances with others in a post-Darwinian
tradition. To begin is Skinner’s understanding of the intersubjective nature of
our subject matter where activities such as thinking and constructs such as
mind and consciousness are not bounded by an individual’s body but, rather,
are viewed as processes and products of social interactions that are worked
out, in no small measure, through language. Weaving Skinner’s contribution
into the mainstream of linguistic thinking requires understanding the four plot
points that have kept him out until now.
The inside/outside story

The usual charge against behaviorist accounts is that they are said to stop at
the skin, and therefore they cannot help us understand the inner workings of
organisms. This is certainly true for some forms of behaviorism, Watson’s for
example, but it is not true of Skinner’s. As mentioned above, for Skinner, activ-
ities such as thinking, problem solving and feeling do not differ in principle
from public human activities such as driving a car, talking and greeting another
person. In Skinner’s form of behaviorism known as radical behaviorism, psy-
chological activities occurring inside the privacy of one’s skin are behaviors
and can be described in the same terms as any and all other behaviors. The
difference between mental phenomena and physical phenomena is thus not
metaphysical, that is, mental phenomena are not immaterial predecessors of
material actions, since they are in and of themselves actions. Rather, the diffe-
rence between mental phenomena and physical phenomena is epistemological:
private mental events are simply less accessible than public events. What is
remarkable about Skinner’s brand of behaviorism is its non-dualism – its
refusal to use separate vocabularies when speaking of observable behavior as
opposed to non-observable behavior. This has made it possible for many crit-
ics of radical behaviorism to call the interior world of a Skinnerian organism
empty or blank. It certainly would look that way to a linguist or a psychologist
or any other social scientists looking for descriptions of mental events that fit
traditional descriptions of mental events.
At the same time, Verbal Behavior is not a book that devotes attention to how
children become languagers of the language of the environment they are born
into. However, if that is a defect, then it is one that many treatises in linguistics
share with it, including everything that Chomsky has ever written. However,
Chomsky’s work has inspired an entire research program for child language/
ing acquisition/learning, whereas Skinner’s did not.1 The reason for this dis-
crepancy in traditional language acquisition research programs has everything
to do with the metalinguistic plot line.
The metalinguistic story

Skinner did not write Verbal Behavior for the linguistic community, but rather
as a result of an offhand remark by the philosopher Alfred North Whitehead,
Skinner’s colleague at Harvard. Whitehead told Skinner that he was willing to
accept behaviorism’s ability to account for human behavior provided one made
an exception of verbal behavior. The remark was made in 1934, and Skinner
spent the next twenty years working on his account of human verbal behavior.
There is no evidence that Skinner was ever terribly concerned with the disci-
pline of linguistics as such. Although it is clear that he is familiar with linguistic
issues involving phonemes and morphemes and sentence structure, he mentions
only three linguists – Jespersen, Sapir and Alan H. Gardiner – and only in pass-
ing in the rather long book that he eventually produced. Nor is there evidence
1
Roger Brown’s A First Language is sometimes cited as a Skinnerian account of language acqui-
sition. While it is true that Brown is rightly critical of Chomsky (1973: x) and complimentary of
Skinner (1973: 388), his work nevertheless strikes me as proto-constructionist.
that he was profoundly provoked by Chomsky’s review of his book. This only
makes sense because he did not write it for the linguistic community. He would
not, therefore, have considered himself under the control of that audience, for as
he himself said: “Practically all verbal behavior is … controlled by an audience”
(1957: 52). This was not a statement Chomsky would have liked, for, as we will
see shortly, Chomsky has long held to the position that one of the distinctive
properties of the ordinary use of language in everyday life is “freedom from
control by external stimuli and inner states” (1988: 138).
One of the more salient features of Verbal Behavior is Skinner’s innovative
terminology, and one of the more salient features of Chomsky’s review was
the constant effort to collapse that terminology back into the traditional defini-
tions. At various instances, Chomsky comments on Skinner’s use of terms: “It
appears that the word ‘control’ here is merely a misleading paraphrase for the
traditional ‘denote’ or ‘refer’” (1959: 33). Or: “We are no doubt to interpret the
terms ‘strength’ and ‘probability’ in this context as paraphrases of more famil-
iar locutions such as ‘justified belief’ or ‘warranted assertability,’ or something
of the sort” (1959: 35). And: “The phrase ‘X is reinforced by Y (stimulus,
state of affairs, event, etc.)’ is being used as a cover term of ‘X wants Y’, ‘X
likes Y’, ‘X wishes that Y were the case’, etc.” (1959: 38). Chomsky’s dis-
like of Skinner’s terminology is thoroughgoing. He does not like mand or tact
or echoic operant, and he certainly does not like anything resembling textual
behavior or intraverbal operant. Writes Chomsky:
A verbal response to a written stimulus (reading) is called “textual behavior.” Other
verbal responses to verbal stimuli are called “intraverbal operants.” Paradigm instances
are the response four to the stimulus two plus two or the response Paris to the stimulus
capital of France. Simple conditioning may be sufficient to account for the response
four to two plus two, but the notion of intraverbal response loses all meaning when we
find it extended to cover most of the facts of history and many of the facts of science; all
word associations and “flight of ideas”; all translations and paraphrase; reports of things
seen, heard, or remembered; and, in general, large segments of scientific, mathematical,
and literary discourse. (1959: 51–52)
With the benefit of hindsight, we can see just how similar are Skinner’s notion
of the intraverbal and the understanding of discourses – those grand realms of
words and texts and traditions we are disposed to calling scientific and mathe-
matical and literary – as critiqued by the French philosopher Michel Foucault.
It is not, as Chomsky would have it, that the notion of intraverbal response
“loses all meaning” when extended to cover these discourses; it is rather
that these discourses can no longer make certain kinds of truth claims under
Skinner’s description. With such a thorough discrediting of Skinner’s termi-
nology by Chomsky, it is no wonder that Skinner wrote some years later that
Chomsky’s review was “not really a review of my book but of what Chomsky
took, erroneously, to be my position” (1972: 346).
Hindsight also highlights how Skinner’s understanding of the autoclitic

function of verbal behavior – which brings the speaker’s responses to the
speaker’s own verbal behavior into the picture – anticipates much of what has
had to be included piecemeal into introductory linguistics textbooks in the past
few decades. What Austin calls speech acts are what Skinner calls descriptive
autoclitics, which is the verbal behavior that a speaker acquires to describe his
own behavior (1957: 313), e.g., I declare (a state of war), I observe (that he
is absent today), I call it (a shame), etc. (1957: 315); what Schiffrin calls dis-
course markers are what Skinner calls an autoclitic that acts as a mand upon the
listener, namely one that starts, stops or deflects the listener’s reactions, e.g.,
Now, then; So!; Oh, well, etc. (1957: 321); and what traditional grammarians
call agreement and case markers are what Skinner calls relational autoclitics,
e.g., the -s plural and the ’s, the latter being a minimal tact indicating singu-
larity but also serving the autoclitic function of denoted “possession” (1957:
333). Skinner identifies a range of types of autoclitics, my favorite being spe-
cial manipulative autoclitics that connect remote responses, signal temporary
digressions, pick up dangling threads, and so on, exemplified by: Incidentally,
by the by, meanwhile, we shall return to this in a moment, but first, paren-
thetically, to go back for a moment (1957: 353). Skinner’s discussion of the
autoclitic might be the most interesting to linguists today in that a distinction
between grammatical phenomena and pragmatic phenomena does not exist in
his account; all such phenomena are viewed as part of a coherent system that
has conceived of the relationship among the phenomena as manifestations of
the same function.
The effect of Chomsky’s review was to bar the possibility of Skinner or
any other behaviorally oriented researcher from participating in the intraver-
bal behavior called linguistics. Chomsky opened his review with terms like
language and language behavior and closed it with repeated instances of the
phrases present-day linguistics, grammar, theory of language, construction of
grammar and mastery of language, thus anchoring his review in its own textual
tradition. At the very least, the Chomsky/Skinner exchange provides a cau-
tionary tale for any linguist wishing to propose new metalanguage and a new
epistemological framework to the linguistic community.
The ethical story

American linguistics immediately prior to Chomsky was behaviorist-oriented;
or perhaps it is more accurate to say that Leonard Bloomfield, the grand old
man in the discipline just before the time that Chomsky came along, was behav-
iorist-oriented. If Chomsky was interested in launching a new research para-
digm for linguistics in the late 1950s, it would hardly have been politic of him
to have launched into a similar critique of the disciplinary insider Bloomfield.
Far better to aim his anti-behaviorist lance at an outsider who transgressed

disciplinary boundary lines.
Despite my claim that I find behaviorism fun, some readers may be feeling
the same discomfort Whitehead felt about a behaviorist account of language. It
is true that behaviorism’s metalanguage is often severe. For example, Skinner
uses the term punishment to describe a turn in a verbal episode when someone
chooses not to pass the bread when asked. In that episode, described below,
Skinner has assumed a listener already predisposed to reinforcing the speaker
with bread. If the listener were not so predisposed, the listener could be said
to punish the speaker with silence, a shrug of the shoulder, a move to put the
bread further out of the reach of the speaker or, say, the words, “No way will
I pass you the bread.” For some, the term behaviorism itself may call to mind
the specters of scary words such as manipulation and control, beginning with
Skinner’s declaration that, “The ultimate aim of the study is the prediction
and control of verbal behavior” (1957: 12). However, let us keep in mind that
what he is trying to predict and control and then account for are classes of
verbal behavior that occur under the kinds of conditions that prevail in human
communities.
For example, Skinner wants to account for the conditions under which we
may get someone to say “violin.” He offers three possibilities: we may resort to
a verbal stimulus plus a mand for echoic behavior: Say “Violin”; we may use a
textual stimulus with an appropriate mand, Read this: VIOLIN; or we may ask
Tell me what this is while designating a violin (1957: 90). In a later passage,
Skinner imagines conditions under which we could get someone to say “pen-
cil.” This may include creating conditions in which the person is to sketch the
picture of a cat and has been given paper but no pencil (1957: 253–54). These
cases are hardly the stuff of my worst nightmares. They are rather the descrip-
tion of my everyday interactions – which is, precisely, what Verbal Behavior
undertakes to describe: the dynamics of our everyday interactions.
Chomsky’s review played up and played on the worst fears engendered
by behaviorist approaches to human activity. Chomsky’s review is thirty-one
pages long. On thirteen of those pages, as already mentioned in Chapter 4,
Chomsky refers to rats or Skinner’s bar-pressing experiments, often more than
once per page – although nowhere in Verbal Behavior is there mention of rats.
Furthermore, whenever I dip into the book, as I have said, I always find it to
be fresh, often amusing and sometimes idiosyncratic – in short, very human.
Nevertheless, Chomsky framed his entire review of Verbal Behavior with a
five-page discussion of Skinner’s earlier Behavior of Organisms (1938) and
recurrently associated Skinner’s innovative terminology for human language
use with rats pressing bars. The most egregious example was Chomsky’s dis-
missal of the term verbal behavior some nineteen pages into the review (along
with a misrepresentation of the term):
Consider first the term “verbal behavior” itself. This is defined as “behavior reinforced
through the mediation of other persons.” The definition is clearly too broad. It would
include as “verbal behavior,” for example, a rat pressing the bar in a Skinner box, a
child brushing his teeth, a boxer retreating before an opponent, and a mechanic repair-
ing an automobile. (1959: 44–45, emphasis mine).
Chomsky’s rhetorical move, that is, to interlard his review of Skinner with
repeated references to “rats pressing bars,” was extremely effective. Among
other things, it caused a theoretical retreat away from the always dynamic,
always variable, always fascinating verbal interactions that occur between and
among walking, talking bodies.
The Chomskyan retreat is complemented by the Chomskyan charge into the
mysteries of the genome, where the linguist is supposed to find – somewhere,
somehow – the principles of the language faculty and the elements of so-called
intrinsic human nature. What is there to fear in non-essentialized accounts of
human behavior, verbal and otherwise? Chomsky tells us:
Quite typically, intellectuals have been ideological and social managers, serving power
or seeking to assume power themselves by taking control of popular movements of
which they declare themselves to be the leaders. For people committed to control and
manipulation it is quite useful to believe that human beings have no intrinsic moral and
intellectual nature, that they are simply objects to be shaped by state and private man-
agers and ideologues – who, of course, perceive what is good and right. Concern for
intrinsic human nature poses moral barriers in the way of manipulation and control. […]
In accordance with these conceptions, human rights are rooted in human nature, and we
violate fundamental human rights when people are forced to be slaves, wage slaves, serv-
ants of external power, subjected to systems of authority and domination, manipulated
and controlled “for their own good”. (1988: 165–66; see also Chomsky 1971)
I appreciate the moral high road Chomsky takes here. However, it is notable
that his argument against the non-essentialist behaviorist account of human
behavior is obliged to shift from an epistemological grounding to a moral one.
The positing of an intrinsic human nature serves as the moral barrier against
any – and therefore all – kinds of manipulation and control, including presum-
ably routine interactions of the type: “Could you do me a favor, and please
close the door? I’m feeling a draft, and I’m blocked from getting to the door
at the moment.” It seems that, in the most simplistic terms, an anti-behaviorist
stance operates under the anxiety that knowledge of a behaviorist account of
human beings would be disastrous for humanity. I believe it is this ethical
story that explains most pertinently why Skinner’s work on verbal behavior has
been stigmatized in linguistic circles, e.g., by Jackendoff, as treating only “the
most trivial of linguistic facts.” If the means by which one person may induce
another person to say “violin” or “pencil” are trivial, then they are also true. It
so happens that it is … happily? sadly? wonderfully? miserably? … the case
that such trivial truths make our verbal world go round.
Chomsky’s linguistics has its roots in an eighteenth-century epistemology

and is inspired by Descartes. His ethics seems to derive more from Marx.
That is to say that for both his linguistics and his ethics, he wishes to retain a
theological rationalist humanism whose enemy is naturalism. But what of this
idea of intrinsic moral and intellectual human nature that is akin to the notion
of inalienable rights on which the American government and its legal institu-
tions are founded? What are we to do with it, if we take a behavioral stance on
this issue and deny an essentialized human nature? The first thing to say is that
while I wish to preserve the operating values of the legal institutions we have,
I do not feel it necessary to preserve the eighteenth-century epistemological
framework in which those legal institutions were engendered in order to pur-
sue twenty-first-century linguistics, biology, evolutionary theory and/or any
other science or social science. Conversely, I do not believe that pursuing work
in the twenty-first-century sciences and social sciences threatens the operat-
ing values of our legal institutions. In other words, I am comfortable with
the position taken by pragmatists like Richard Rorty who argue that it is not
necessary to believe in God in order to believe in the civic value of upholding
the speech-act formula used in courtrooms I swear to tell the truth, the whole
truth and nothing but the truth, so help me God that legally binds witnesses
to their words. The second thing to say is that, although radical behaviorism
does not engage with the concept and complexities of agency, an empirically
adequate account of moral and/or legal agency does not require freedom from
causes. Rather, it seems to require embeddedness in a causal world where
one is engaged by the exchanges that constitute human life. “Only there,” as
Oyama says, “can one be moved to encourage some outcomes and prevent
others; and only there can one be positioned among others who regard one as
responsible” (2000: 183).
Chomsky’s antagonism toward Skinner’s work extended beyond Verbal
Behavior to include Skinner’s Beyond Freedom and Dignity (1971), which
Chomsky reviewed in the New York Review of Books (1971). It exceeds the
scope of this chapter to engage in a critique of this second review, but we might
note Skinner’s reaction to it:
I have never actually read [Chomsky’s] long review of Beyond Freedom and Dignity
though I have read three answers to it none of which the New York Review would pub-
lish. I have never been able to understand why Chomsky becomes almost pathologically
angry when writing about me but I do not see why I should submit myself to such verbal
treatment. If I thought I could learn something which might lead to useful revisions of
my position I would of course be willing to take the punishment, but Chomsky sim-
ply does not understand what I am talking about and I see no reason to listen to him.
(Skinner letter cited in Murray2)
2
1977, an exchange of letters between Murray and Skinner, 1977, personal communication.
Although it would have been interesting and, perhaps, useful for Skinner to have
engaged in a verbal exchange with Chomsky, it is understandable why he saw
no percentage in doing so. Skinner’s own verbal behavior was both consistently
behaviorist (he speaks freely of punishment) and gentle. Verbal Behavior ends
with the words: “But whatever the reader may think of the success of this ven-
ture, I hope he will agree that the analysis has shown respect for human achieve-
ment and that it is compatible with a sense of dignity” (1957: 460). Chomsky’s
own verbal behavior has been, by way of contrast, consistently severe.
Perhaps now is a good time to quote a comment Charles Darwin made in the
sixth edition of On the Origin of Species: “Great is the power of steady misrep-
resentation; but the history of science shows that fortunately this power does
not long endure” (1962 [1872]: 421). Which leads to the final plot line.
The evolutionary story

As pointed out in Chapter 4, Verbal Behavior opens and closes with evolution-
ary considerations. The memorable first paragraph proposes a double continu-
ity of verbal behavior with all other human behaviors, and of verbal behavior
with the behaviors of other creatures:
Men act upon the world, and change it, and are changed in turn by the consequences of
their action. Certain processes, which the human organism shares with other species,
alter behavior so that it achieves a safer and more useful interchange with a particular
environment. When appropriate behavior has been established, its consequences work
through similar processes to keep it in force. If by chance the environment changes, old
forms of behavior disappear, while new consequences build new forms. (1957: 1)
Skinner closes with an appendix entitled “The Verbal Community,” which is

devoted to a discussion of “the old question of the origin of language” (1957:
461). The beginning and the ending are thus tied together by one of Skinner’s
enduring insights, that of selection by consequences, where behaviors, both for
the species and for the individual, are selected and maintained by effects in an
evolutionary framework. Skinner notes the parallel between natural selection
and operant conditioning: “The selection of an instinctive response by its effect
in promoting the survival of a species resembles, except for enormous differ-
ences in time scales, the selection of a response through reinforcement” (1957:
462–63). Skinner’s account of verbal behavior is, thus, from beginning to end,
continuous with evolutionary theory.
In light of the growing interest in evolutionary scripts for language/ing in
the linguistics community, three useful features of Skinner’s approach come
into high relief:
(i) in contrast to Skinner’s account of verbal behavior, evolutionary consider-
ations were not part of Chomsky’s original thinking, and this has caused
consternation in the linguistics community. The disciplinary problem was

clearly expressed by Steven Pinker and Paul Bloom, who write: “Because
we are impressed both by the synthetic theory of evolution and the theory
of generative grammar, we hope that we will not have to choose between
the two” (1990: 708). With Verbal Behavior, one does not have the dilemma
to begin with. As was noted in Chapter 5, Chomsky has been more likely
to address evolutionary issues in the past few years, and he is now claim-
ing that he has always done so. For instance, he has recently remarked that
linguists should take the ethological perspective seriously and claims that
he has been arguing for it for forty years (2002: 77–78). Following the foot-
note to this sentence, curious to learn about Chomsky’s long engagement
with an ethological perspective, I was surprised to find only a reference
to Chomsky (1959). In his review of Verbal Behavior, there is only one
passage, namely one on what Chomsky calls “unrewarded learning,” that
suggests that “insights arising from animal behavior studies … may have
the kind of relevance to such complex activities as verbal behavior that
reinforcement theory has, so far, failed to exhibit” (1959: 41). Otherwise,
the review is an extended argument against the relevance of animal behav-
ior studies to understanding verbal behavior; and there is only one passing
mention of the theory of evolution in Chomsky (1959), but in relation nei-
ther to Skinner’s perspective on the subject nor to the theory itself but rather
as an example of the ways scientific theories are confirmed (1959: 35);
(ii) the features of small-scale context and of temporality that are built into
Skinner’s understanding of verbal episodes transfer easily to an under-
standing of the large-scale context and temporality demanded of evolu-
tionary explanations; and
(iii) an account, such as Skinner’s, which is predicated on variation, that is,
differences among organisms of the same species, will have less diffi-
culty playing on an evolutionary axis than one predicated on idealized
sameness, since natural selection does not work on idealizations but rather
selects among variations that exist at any given time.
Skinner and Austin

Because J. L. Austin’s pragmatic account of speech acts in How To Do Things
With Words (1962) has had such good uptake in the linguistic community, it is
worth comparing his account with Skinner’s, now with fifty years’ hindsight.
The first verbal episode Skinner analyzes in Verbal Behavior is a bread-pass-
ing episode. The model shows us the minimum requirements for any analysis,
namely a four-term contingency relation: (1) initial stimulating events (the pres-
ence of bread, an audience), (2) the verbalizing action (gesturing or speaking
or writing, “Please pass the bread”), (3) the consequating action that mediates
Skinner and Austin 241
the verbalization (passing the bread), and (4) the consequence (obtaining the
bread). When compared with Saussure’s talking heads, the bread-passing epi-
sode puts a whole speaker (not just a head) in the presence of a whole listener
(not just a head). In addition, the speaker and listener are contextualized with
respect to bread; the episode depicts several verbal exchanges; specific verbal
operants are represented; the episode unfolds in real time; the speaker and lis-
tener behave differently with respect to each other, and they affect one another;
they actually move. The overall effect upon the speaker is the reinforcements
of the mand (“request”) which occurs with the presentation of the bread, and
this completes the account as far as the speaker is concerned.
By studying verbal behavior in terms of its effectivity, Skinner has no place
for the sameness and symmetry and stasis that figure in both Saussurean and
Chomskyan linguistics. He also has no reason to posit the (pre)existence of a
stable and/or static code to secure mutual intelligibility and no need to propose
a theoretical object abstracted away from, or existing prior to, any given ver-
bal episode. That is, he never pauses to freeze-frame his analysis, since verbal
behavioral episodes are always ongoing. (Current effects are effects of prior
effects of prior effects of prior effects… not unlike long, Austinian speech-act
braids.) Skinner’s analysis is thereby unequivocally non-representationalist.
Any statements he might wish to make about the (internal) state of a partici-
pant in an episode will be a result of the verbal dynamics rather than any kind
of (mental) representation of the verbalizing action. Mention of the possibil-
ity of a reference to an internal state may surprise someone unfamiliar with
Skinnerian behaviorism. Skinner does include internal events such as remem-
bering and thinking and problem solving within his purview, as mentioned
above. However, he sees no reason to describe them in terms different from
those he uses to describe any and all other behaviors.
Furthermore, Skinner does not assume the conversation to be the paradigm
case of a verbal episode. More importantly, he distinguishes himself by ques-
tioning the assumption deeply embedded in, say, certain schools of linguistics
and philosophy, e.g., those prevalent in most current schools of linguistics,
that the speaker is a causal agent responsible for the structure and character
of verbal behavior. Skinner, by way of contrast, presents verbal behavior as a
repertoire of responses, some of them minimal in size (Bread, please!), others
complex (Hamlet’s soliloquy in the repertoire of many actors), existing in vari-
ous states of strengths under the control of variables in the environment and his-
tory of the speaker. He thereby necessarily reduces (the need for an explanation
in terms of) the (supposed) contribution of any inner agent. Skinner identifies
the speaker as “the organism which engages in or executes verbal behavior,”
but he no longer views that organism as a source of the activity but rather as
a locus of the activity, quite literally, “a place in which a number of variables
come together in a unique confluence to yield an equally unique achievement”
(1957: 313). The two instances of the word unique in the preceding quote mark
a stark contrast between Skinner’s analysis of verbal behavior, which is always
differential, and both Saussure’s concept of langue and Chomsky’s concept of
I-language (or knowledge or competence) which are theorized to be essentially
the same among all speakers/human beings.
In reducing the space of the so-called the inner causal agent in his explana-
tory framework, Skinner does not reduce human behavior to simple machine-
like responses to stimuli – that is, there is no Newtonian so-called billiard ball
determinism here. Rather, human behavior is dynamic, ever changing and
always responsive to a dynamic and ever-changing environment, both medi-
ated and unmediated by other humans, by other creatures and by other objects.
In this respect, human behavior is like the behavior of all other organisms.
The difference is that human behaviors are more responsive to a wider range
of stimuli, e.g., the vast and complex range of verbal stimuli, than most other
organisms.
The parallel here with Austin is clear: both Skinner and Austin emphasize
total speech episodes, with the participants in them behaving in asymmetric
ways. In addition, Skinner and Austin do not methodologically separate the
spoken from the gestural. Skinner’s definition of “verbal behavior as behav-
ior reinforced through the mediation of other persons [which does] not, and
cannot, specify any one form, mode or medium” (1957: 14) finds a parallel
in Austin’s understanding that both utterances and actions can be performa-
tive. “Suppose I bow deeply before you,” Austin writes, “it might not be clear
whether I am doing obeisance to you or, say, stooping to observe the flora or to
ease my indigestion.” In order to make clear both that a conventional ceremo-
nial act is at issue and which act is at issue, Austin goes on to say that “the act
(for example of doing obeisance) will as a rule include some special further
feature, for example raising my hat, tapping my head on the ground, sweeping
my other hand to my heart, or even very likely uttering some noise or word,
for example ‘Salaam’.” Austin explains this case as one of “actions which are
non-linguistic but similar to performative utterances in that they are the per-
formance of a conventional action (here ritual or ceremonial)” (Austin 1962:
69–70). The difference between Skinner and Austin on this score is termino-
logical: Skinner would identify both acts – ceremonial bowing and the uttering
of “Salaam” – as verbal behavior and would distinguish between them by call-
ing them non-vocal verbal behavior and vocal verbal behavior, respectively,
while Austin identifies both acts as performatives and distinguishes between
them by calling them non-linguistic performatives and linguistic performa-
tives, respectively.
Austin, like Skinner, does not methodologically separate the spoken from
the written. When it comes to the performative act of warning, for example,
Austin proposes a series of expansions (1962: 62):
Skinner and Austin 243
‘You are warned that the bull is dangerous’ is equivalent to ‘I, John Jones, warn you that
the bull is dangerous’ or
This bull is dangerous. (Signed) John Jones.
In other words, for Austin, the written message “The bull is dangerous.
(Signed) John Jones” is equivalent in its illocutionary force to its spoken vari-
ants. Skinner would certainly agree that all three instances of this segment of
verbal behavior could have similar effects, i.e., causing the recipient to move
out of the way of the bull. He would even approve that Austin had not made
one form more primary than the other, because, although Skinner is aware that
some written forms of verbal behavior arise only after vocal behavior has been
established, he also says that this is not necessarily so. I concur. In interpreting
my own writing life, I would say that the novels I have written are non-vocal
in origin, while the present book, while non-vocal in form, is not so explicitly
non-vocal in origin.
For most of the twentieth century, it was considered major disciplinary her-
esy not to separate the study of the spoken from the written and not to privilege
the spoken over the written. Saussure referred to the “tyranny of writing” and
called for a radical separation between the study of langue and writing, stigma-
tizing spelling pronunciations, e.g., pronouncing the historical and now-silent
‘t’ in often, to be “monstrosities.” He was willing to allow that they “belong to
langue,” but he said that they “do not stem from its natural functioning. They
are due to an external influence. Linguistics should put them into a special
compartment for observation: they are teratological cases” (1959 [1916]: 32).
Chomskyans, and other linguists certainly, have continued to maintain the spo-
ken/written separation, usually casting it in a “nature versus nurture” argument
structure. That is, they judge speaking (or signing) to be a product of nature,
while reading to be a product of nurture. In these arguments it is usually said
that learning to read is a largely effortful undertaking requiring much explicit
instruction, while learning to speak (or to sign) takes place rapidly and sponta-
neously, that is, without explicit instruction.
Although the supposed ease or difficulty of these activities is debatable
(never mind the elision of listening), the real problem with their separation is
that it truncates an understanding of languaging living beings’ experiences with
environments that include an emphasis on literacy. Such an understanding will
go a long way to explaining some very common features of that languaging
living being’s behavior: a responsiveness to visual doublets such as light and
lite, laughs and laffs, which also suggest that semantic change can arise through
writing, and a tendency to speak punctuation: “She thinks of herself as a quote-
unquote journalist.” “He is a body builder slash personal trainer.” “I want to
hear no more on this subject. Period.” People now use text abbreviations in
speech, e.g., the student I overheard saying “JK” (just kidding) to the person he
had been teasing. With writing everywhere, including all over our clothing and
mobile phones, it is simply to state the obvious that the spoken and the written
interpenetrate the lives of a great many languaging living beings on this planet.
We do not honor environments that traditionally do not emphasize literacy by
ignoring that dimension of those environments that do.
In the opening pages of his review of Skinner, Chomsky points out that “one’s
estimate of the relative importance of external factors and internal structure in
the determination of behavior will have an important effect on the direction of
research on linguistic (or any other) behavior” (1959: 27). Here we have the
idea that he is going to offer a moderated corrective to the lack of attention to
internal explanations characteristic of behaviorist approaches at the time. What
happened, as we know, was a complete replacement of that framework with a
rationalist insistence on knowledge that is fixed and unresponsive. Fifty years
later, the corrective that is in order is the rehabilitation of Skinner, who highlights
the dynamic and reactive character of behavior. But why not just take Austin and
leave Skinner behind? The answer is that Austin offers the philosophic argument
for the primacy of the performative in the context of the total speech episode,
while Skinner offers the stimulus – literally – that puts it into play.
Consider, for example, the statement by Pinker and Bloom in their “Natural
Language and Natural Selection” to the effect that: “presumably there is a large
selective advantage conferred by being able to learn in a way that is essentially
stimulus-free … Children learn from a parent that a food is poisonous or a partic-
ular animal is dangerous; they do not have to observe or experience this by them-
selves” (Pinker and Bloom 1992: 481–82, emphasis mine). Their point is not
difficult to interpret. Pinker and Bloom are commenting on what is sometimes
called cultural learning. Skinner made a similar point when he noted that once
vocal verbal behavior began to be shaped and maintained by its reinforcing con-
sequences, people could prime the behavior of others, not only by showing them
what to do, but also by telling them. Skinner puts the matter thusly: “Advice can
be useful on more than one occasion, and it is then often given or taught in such
a way that it is passed on from person to person or from generation to generation.
Maxims (‘great saying’) and proverbs (‘saying put forth’) are examples” (1990:
1206). Pinker and Bloom’s idea is also known as the tart maxim: “Experience is
the wisdom of fools.” The only point of interest is that they configure such learn-
ing to be stimulus-free. Evidently, their own verbal behavior has been shaped in a
way that makes them avoid using the term stimulus. At the very least, rehabilitat-
ing Skinner will relax the taboo on the use of the word that has tied one hand of
the linguist behind his back for far too many years.
The impoverishment of the stimulus

In his review of Skinner, Chomsky was interested not only to dismantle the ter-
minology Skinner created to describe verbal behavior but also to render all of
The impoverishment of the stimulus 245
behaviorist terminology such as Skinner used it – stimulus, response, response

strength, reinforcement, conditioning, etc. – empty and pointless.3 It would
take another twenty years before Chomsky actually coined the phrase poverty
of the stimulus, which first appeared in Rules and Representations (1980: 34;
see Thomas 2002). The phrase is rich with implications.
The first concerns the language learner and points to the idea that language
inputs are too impoverished both in quantity and in quality to explain the rap-
idity and accuracy with which children learn their native languages. Chomsky
has called this “Plato’s problem,” to wit, “How comes it that human beings,
whose contacts with the world are brief and personal and limited, are able to
know as much as they do?” (1988: 3–4). If individuals cannot glean what are
considered to be the important features of the grammar of their language from
the surface structures they encounter as language learners – e.g., how to rela-
tivize from a prepositional phrase, since relativized prepositional phrases have
a low frequency of occurrence – then the problem of how they learn their lan-
guage at all is solved by “loading them for language at the factory.” The task
for language acquisition specialists, then, is one of discovering how the child
converges on the rules and other components of grammar through a combin-
ation of innate knowledge and deductive (non-analogic) reasoning.
The second implication of the phrase poverty of the stimulus concerns the
(more or less adult) language user and points toward the idea that no mat-
ter what situation the user is in, the amount and/or force of the stimuli are
too impoverished to determine what the user says and why the user says it.
Chomsky called this “Descartes’s problem” or “the creative aspect of language
use.” It is tied to the view of the individual that is the backbone of Cartesian
epistemology, namely that the individual is a unique and bounded entity, an
actor rather than a locality, endowed by the Creator – or, in Chomsky’s case,
by the genetic code – with free will. Recall the passages from Language and
Problems of Knowledge quoted in Chapter 3, where Chomsky fudged the terms
cause and evoke. This fudge promoted and preserved a continuous feed-forward
system, with no feedback. Already in his review of Skinner, Chomsky says that
the ability, say, of an adult to read a newspaper and understand “countless new
sentences which are not at all similar, in a simple, physical sense” and that he
will also be able to detect slight distortions or misprints indicates “that there
must be a fundamental processes at work quite independently of ‘feedback’
from the environment” (1957: 42).
A third implication of the phrase poverty of the stimulus involves the idea
that certain surface features of language underspecify the properties of the
grammar that produces them, thus making it difficult for learners to learn
3
Unpacking the terminology, along with Skinner’s use of it and Chomsky’s criticisms, goes
beyond the scope of this chapter.
those properties. An example of such a feature would be the notion of struc-

ture dependence as a hypothetical language universal, introduced by Chomsky,
who asserted that children create only structure-dependent grammars although
no feature of the input eliminates structure-independence. “Chomsky’s key
example – that English-learning children do not form questions like *Is the
man who ___ tall is in the room? – eventually became … the ‘parade case’ in
discussion of the poverty of the stimulus” (Thomas 2002: 56). That is, there
are supposedly some grammatical mistakes that children might logically make
on the basis of so-called simple induction but do not (see Tomasello 1995: 144,
for a critique). Thus, if the information cannot be read off the surface form and
there is no positive evidence to prevent children from making such mistakes,
then their knowledge of this non-occurring syntactic structure must be given in
Universal Grammar.
The fourth implication of the phrase poverty of the stimulus concerns the
phrase itself and points toward the idea that stimuli as such are somehow
impoverished in the sense of degenerate, i.e., speakers produce sentences all
the time that are incomplete or inadvertently ill-formed for some reason or
another or are ambiguous or unclear or contain performance errors, like slips
of the tongue. At this point, let us simply turn the phrase on itself and say that
the only thing impoverished about the stimulus in the Chomskyan framework
is the Chomskyan notion of it, since stimuli supposedly play little or no role in
either language acquisition or adult language production. And it is not a ques-
tion that the Chomskyans have misspoken, in a sense, when they talk about
the poverty of the stimulus, when what they have really meant all the while is
that the spare inputs children receive are enough to construct their grammars,
i.e., that there is no impoverishment at all and instead a sufficiency of stimuli,
although slight. Rather, there is no mystery why stimuli would be so impov-
erished in this style of analysis: once the philologist-linguist has surgically
removed the sentence from a context, then drained it of all meaning so that
only the empty forms remain, there are no stimuli left to consider, and there is
no possibility that a stimulus of any kind could have any effect anyway on what
remains, namely a syntactic form. However, when we fill linguistic forms back
up with content and then place them back into the contexts in which they are
produced, we discover plenty of stimuli to go around.
From Bloomfield to Chomsky

Chomsky’s Syntactic Structures of 1957 is sometimes said to have revolution-
ized the discipline of linguistics, even to have set it on its ear. Historiographically
speaking, it did not set linguistics on its ear but rather set it back several hun-
dred years by (re)asserting the science of language as a kind of theoretical
psychology – a special science of human minds – in a rationalist framework
From Bloomfield to Chomsky 247
that is the legacy of the seventeenth and eighteenth centuries. Linguists of the
prior era, e.g., Leonard Bloomfield, would have said that their work was to
describe the structure of human languages. Chomsky reoriented the field by
claiming that to describe language (here understood already as an abstract
property of the mind) is to describe a central feature of the human mind. The
research agenda in linguistics thus shifted from the Bloomfieldian approach:
“how do we (linguists) describe our subject matter?” to the Chomskyan ques-
tion: “what is the best way for us (linguists) to describe how speakers have
language organized in their heads?”
Up through the mid 1950s, linguists had been working to identify the sound
(phonological) patterns and the word formation (morphological) patterns of the
various languages of the world. What seemed to be just beyond the linguist’s
grasp was the ability to capture regularities that obtained over larger stretches
of linguistic forms, e.g., sentences. In the chapter entitled “Form-Classes and
Lexicon” in Language, Bloomfield understood that while it was possible to
list all the words in a language, it was not reasonable to make a list of all the
sentences in a language, “since the possibilities of combination are practic-
ally infinite” (1933: 275). Thus, while it makes sense to compile a dictionary,
it makes no sense to (try to) compile a “sentence-ary.” Bloomfield is willing
to identify sentence types, e.g., “the actor-action form.” In an earlier chapter
of Language entitled “Sentence-Types,” Bloomfield names “the actor-action
construction” to be one of two “favorite sentence-constructions” in English,
the other one being “the command” (1933: 199). In that earlier chapter, he dis-
cusses different “favorite sentence-forms” in various languages, but he never
strays far from describing what we might call the surface appearance of any
particular language’s structures.
In “Form-Classes and Lexicon,” Bloomfield also suggests that the relation-
ship between the observer, i.e., the linguist interested in describing the structure
of a particular language, and the speaker, the one producing utterances, is one
of precedence and of perspective. As already noted in Chapter 3, the speaker–
linguist relationship is one of precedence in that a linguist can only observe,
record and/or describe after the fact what a speaker produces and of perspec-
tive in that a linguist is not limited to observing, recording and/or describing
the productions of any one speaker but can take a more global perspective.
However, Bloomfield acknowledges that this supposedly global perspective
itself will always be limited, since any one observer (or group of observers)
cannot necessarily have access to all utterances that all speakers of a language
have uttered. In the case of the sentence A red-headed plumber bought five
oranges or any other particular combination an observer may chance to hear,
an observer cannot say for sure whether or not it has ever been uttered.
One key term in Bloomfield’s discussion of the large number of pos-
sible actor-action forms is analogy: “A grammatical pattern (sentence-type,
construction, or substitution) is often called an analogy. A regular analogy per-

mits a speaker to utter speech-forms that he has not heard; we say that he utters
them on the analogy of similar forms which he has heard” (1933: 275). The
term analogy is doing double duty as a descriptive feature of a language: “a
grammatical pattern … is often called an analogy”; and as a mental operation:
“[a speaker] utters [speech-forms that he has not heard] on the analogy of
similar forms that he has heard.” In both cases, analogy is a phenomenon that
hovers at or just below the so-called surface of an utterance. As a descriptive
feature, that which is observed to be analogical in a language is a statistical
property based on frequency and distribution of elements that actually occur.
As a mental process, an analogical formation is one that is responsive to the
frequency and distribution of elements in a language that the individual produ-
cing the analogical formation has actually encountered. Bloomfield identifies
two principles according to which analogy operates: (i) the larger class attracts
the smaller class; and (ii) the more frequent patterns resist change, which is
why irregular forms appear chiefly among the commonest words and phrases
of a language.
The forms on which the analogy is supposed to be based must occur or must
have been encountered; otherwise, there is nothing for an analogical forma-
tion to be analogical to. In other words, analogy does not create new gram-
matical patterns; it simply extends the range of a pattern that already exists.
Of course, for the pattern to exist in the first place, an individual or group of
individuals will have had to have extended it by the mental operation of anal-
ogy. Bloomfield does not elaborate on analogy as a mental operation, and he
makes no grand or deep claims about it in relationship to the mind or how the
mind works. However, he is aware that the extension of a grammatical pattern
is as much a social process as a mental one. He acknowledges that not all – and
perhaps very few – instances of what the linguist perceives as the extension
of the grammatical pattern result from the same kind of cognitive process that
produced the first occurrence of the extension. Bloomfield is perfectly willing
to distinguish between these two processes – the mental extension of a form
and the social uptake of that extension – and he wants to limit the term to the
process of describing the mental operation of the individual, who, for example,
having habitually produced kine as the plural of cow, produced the novel form
cows without ever having before heard the word cows (1933: 405).
Another key term in his discussion is form. On the one hand, a linguistic
form refers to something with physical presence. In the same passage from
“Form-Classes and the Lexicon,” Bloomfield writes: “[we] may say that any
form which a speaker can utter without having heard it, is regular in its imme-
diate constitution and embodies regular functions of its constituents, and any
form which a speaker can utter only after he has heard it from other speakers,
is irregular” (1933: 274). Here Bloomfield is using the term form to refer to
something physical, something utterable, namely a phonetic sequence. This is

very common usage, and linguists regularly speak and/or write about linguis-
tic forms as what words sound like when spoken, e.g., the form cat [kæt]. On
the other hand, another sense of the term form lurks in Bloomfield’s passage
in phrases like the actor-action form or favorite sentence forms, and this sense
suggests precisely the absence of any physical/perceptible linguistic material.
What happened next? Chomsky took up the latter sense of form with a ven-
geance and produced a theory of syntax replete with grammatical categories
absolutely empty of content. Those categories were abstract outlines or tem-
plates, only waiting to be filled with content. The double sense of form – the
(messy) presence of linguistic forms (plural, particular) versus the (cleaned-
up) absence that reveals The Form (singular, universal) – has been in linguis-
tics since the turn toward structuralist descriptions, and probably quite a long
time before that.
Bloomfield, a representative twentieth-century American linguist, had, like
his European counterparts, turned his attention from nineteenth-century his-
torical and comparative work in order to concentrate on descriptions of the
so-called language state in the absence of change. However, American lin-
guists never signed on to the Saussurean agenda of describing the abstract con-
struct langue, so in the time period immediately prior to Chomsky, American
linguists were less interested in x-raying the langue that existed in potentia
beneath the welter of parole and were more interested in analyzing structures
that appeared in a representative sample of language called a corpus. Linguists
use corpora to make statements about frequency of usage. The ranges and sizes
of linguistic corpora are variable; they may attempt to cover the language as
a whole, taking extracts from many kinds of texts, or they may be selective,
providing a collection of material that deals only with a particular linguistic
feature. Corpora can serve legal uses as well, for instance, as evidence to sup-
port a witness’s contention that s/he did not understand the meaning of a word,
e.g., carcinogenic, based on the number of times it appears in millions of run-
ning words. In the style of American linguistic analysis that prevailed until the
mid 1950s, the relationship between an actual, unwieldy and never complete
corpus of data and the description of linguistic structures observable in that
corpus was very tight.
Chomsky altered the theoretical landscape by arguing that “the possibilities
of sentential combinations,” while, indeed “practically infinite,” as Bloomfield
had said, are not, in fact, infinitely variable. They, too, exhibit patterns.
However, these patterns are not to be read from surface structures, that is, they
are not to be found directly in the surface strings of the practically infinite set
of sentences themselves; and they are not a function of anything so cognitively
low-tech as analogy. Rather, they are more abstract. In the early days of trans-
formational generative grammar, sentence production was discussed in terms
of phrase structure rules, e.g., S → NP VP, to show the order and hierarchy of
sentential elements, and transformational rules such as Equi NP Deletion and
For Deletion, to show how elements could move. Then came a middle period
where there was talk of constructs such as principles and parameters, which
latter refers to sets of alternatives for a particular syntactic phenomenon made
available by Universal Grammar, e.g., whether a language can drop pronouns
or not. Today, in the Minimalist grammar, two basic operations, merge and
move, are proposed to account for the link between the sensory-motor sys-
tem (SM) and the conceptual-intentional system (CI) (Chomsky et al. 2005,
Appendix p. 2). The task is to map the interfaces between these two levels
in the most efficient way possible, where the best possible linkage solution
is language. Of note here is the extremely minimal elaboration of a technical
definition of language: a best possible solution to the problem of linking (a
preexisting) SM and (a preexisting) CI.
Because these patterns cannot be read from surface structure and because
the focus of analysis is on the relationship between pairs of contextless sen-
tences, there is no longer any point to working with corpora. If one is inter-
ested to understand the relationship between pairs of sentences such as “John
is expected to be intelligent” and “John seems to be intelligent,” it does not
make sense for linguists to search databases to find examples; it is more effi-
cient for them to rely on their own intuitions to produce the relevant examples.
Furthermore, the relevant limiting cases of grammaticality are precisely the
kinds of structures that do not occur. Thus, searching for them in a database
would be, by definition, futile. When linguists began to rely on their own intui-
tions, the distinction between the observer and speaker characteristic of the
Bloomfieldian program collapsed. Speakers’ utterances no longer had prec-
edence over observers’ analyses, since the observer and the speaker were one
and the same. And linguists’ perspectives on syntactic phenomena were nec-
essarily narrowed to the sentences they generated out of their own so-called
native-speaker competences and either judged to be grammatical, i.e., well-
formed (Whom does Mary see?), or judged to be ungrammatical, i.e., not
occurring (*Does see Mary whom?).
Under the Bloomfieldians, it was already bad enough that corpora were seen
as little more than monumentalized stretches of utterances that had been severed
from the contexts of their production and use. However, the situation worsened
when the need to work with corpora disappeared, because as the corpora went,
so went even those bits of surrounding verbiage that might have supported the
analysis of any particular (syntactic) phenomenon under review. In Chomskyan
linguistics, the interest in analyzing any particular sentence no longer resides
in understanding its relationship to the stretches of discourse that precede and
follow its possible occurrence, much less the context in which it might con-
ceivably be uttered. In other words, the reliance on so-called native-speaker
intuitions in the Chomskyan era automatically effaced all discursive triggers

that could conceivably be taken into consideration to account for languaging
behaviors, just as the reliance on corpora in the previous era had automatically
effaced any and all environmental and/or perceptual triggers. Worse still is the
fact, which variationists have known for decades, that introspective judgments
about language, especially syntax, are often highly unstable and at variance
with attested behavior (Rickford et al. 1995: 127).
In stepping away from corpora and into abstraction, the Chomskyan meth-
odology identifies structures that can be viewed retrospectively and from the
point of view of the linguist-speaker. Thus is formal syntax brought into exist-
ence, for only in retrospect and in isolation from all surrounding phenomena –
context – can the structure of any particular utterance be seen to preexist its
appearance in discourse. The construct competence – i.e., the mental system
that underlies a person’s ability to speak and understand a given language as
well as the so-called proper object of linguistics – serves as the harbor for
sentence pattern templates as they await instantiation. In their preexistence,
then, the abstract rules are theorized to represent the speaker’s ability to prod-
uce sentences. Just as the observer/linguist and the speaker have merged, so
the linguistic description of non-surface regularities, seen from the perspective
of the outside observer now inside the speaker’s head, doubles as the mental
explanation of that behavior.
What a difference fifty years make. Recently, certain psychologists and
linguists have taken renewed interest in the role that stimuli – aka inputs or
statistical properties of a language based on frequency and distribution of ele-
ments – play both in learning by infants and young children and in comprehen-
sion and production by adults. This has led to a revival of the use of corpora
both in formal theoretical linguistics and in language acquisition studies, where
written corpora are providing evidence that children are, indeed, exposed to
examples of the kinds of complex grammatical structures that Chomsky said
children never hear or virtually never hear, an absence that required him to
posit the existence of inborn knowledge of very abstract features of language
(Tomasello 2003: 288; Pullum 1996; Pullum and Scholz 2002).
When the dynamic life of languaging in verbal interaction is restored and
the richness, rather than the poverty, of so-called inputs is recognized, a new
relationship of linguist to subject matter becomes possible and new descriptive
terrains open up. As suggested in Chapter 3, Bloomfield, like any good scien-
tist, desired completeness, but he recognized a dual limitation, that of descrip-
tive and observational inadequacy: no linguist could hope to list all the items a
grammar could contain “since the possibilities of combination are practically
infinite,” just as no linguist could hope to have access to all utterances whose
characteristics should be accounted for in a grammar. Chomsky collapsed
the observer into the speaker and solved the problem of limitation by putting
the conditions for generating a complete grammar inside every human being,
along with a complete knowledge of the (linguistic) environment that is evoked
at the appropriate times. In DST linguistics, by way of contrast, the drive for
completeness transforms itself into a taste for thick descriptions of individual
development through multilayered temporal axes and historical trajectories, as
I hope the present study shows.
The cognitive turn and (cognitive) construction grammar

At the turn of the new century, the discipline of linguistics is, in a way, in the
place where Chomsky was fifty years ago. For some decades now, discourse
analysts have been offering their fellow linguists descriptions of patterns that
occur in larger stretches of our languaging lives. There is no reason to suppose
that the work that is now coming out of discourse analysis has fewer cognitive
implications than the work that is coming out of, say, formal syntax. In fact,
the understanding of the cognitive dimension of the discourse marker oh is the
very point of Schiffrin’s work when she writes:
although oh is a marker of information management tasks which are essentially cog-
nitive, the fact that it verbalizes speakers’ handling of those tasks has interactional
consequences. Thus, use of oh may very well be cognitively motivated. But once an
expression makes cognitive work accessible to another during the course of a conversa-
tion, it is open for pragmatic interpretation and effect – and such interpretations may
become conventionally associated with the markers of that work. (1987: 100)
The discourse analytic approach also fits nicely with the cognitive perspec-
tive promoted by languaging as an orienting behavior in that languaging has
to occur for a languaging living being to be oriented in his or her cognitive
domain. Languaging living beings do not language with abstractions. It is per-
tinent to see that only now that linguists have taken the description of structure
up a level from the sentence and situated utterances in their contexts is the dis-
cipline of linguistics finally ready to leave philology – the study of monumen-
talized language – behind. For it is in situ that languaging lives, that the two
poles of an electric current meet in order to create an energy that is perceived
as light, and we are able to see that “actually, every utterance is some kind of
speech act.” And, then, necessarily, every speech act has some kind of cogni-
tive consequence. The study of those consequences in phylogeny and ontogeny
is central to the project of a developmental linguistics.
A developmental linguistics, therefore, is interested to incorporate the more
psychologically satisfying and pragmatically informed line of thinking about
grammar that extends back to Ronald Langacker’s Foundations of Cognitive
Grammar (1987) and extends forward to Adele Goldberg’s Constructions at
Work: The Nature of Generalization in Language (2006), the latter forming
the basis of the exposition in this section. Goldberg identifies her approach as
Cognitive construction grammar 253
cognitive construction grammar (CCG) and places it in a lineage that includes

George Lakoff’s Women, Fire, and Dangerous Things: What Categories Reveal
About the Mind (1987), Charles Fillmore, Paul Kay and Mary Catherine
O’Connor’s “Regularity and Idiomaticity in Grammatical Constructions: The
Case of LET ALONE” (1988), as well as Fillmore and Kay’s “Grammatical
Constructions and Linguistic Generalizations: The What’s X doing Y?
Construction” (1999). CCG has not made an appearance in any current intro-
ductory textbook, although occasionally Fillmore’s, Langacker’s or Lakoff’s
name may be mentioned in passing.
Pertinently, the treatment of the let alone construction in Fillmore, Kay
and O’Connor (1988) is parallel to Austin’s treatment of the performative. At
the beginning of their argument, Fillmore et al. acknowledge that a construc-
tion such as let alone is usually seen as a minor sentence type or special
construction. However, by the end, they claim it as central, just as the so-called
minor case of performatives eventually swallows the seemingly larger case of
constatives. They write that “a large part of a language user’s competence is
to be described as a repertory of clusters of information including, simultan-
eously, morphosyntactic patterns, semantic interpretation principles to which
these are dedicated, and, in many cases, specific pragmatic functions in whose
service they exist” (1988: 534). This approach thus departs significantly from
the traditional generative approaches that do not countenance direct pragmatic
interpretation of syntactic structures unmediated by the proposition expressed.
CCG has long been recognized as a viable alternative to transformational gen-
erative grammar by many linguists and language theorists who are not them-
selves cognitive construction grammarians. Tomasello, for one, in his review
of Pinker’s The Language Instinct, cites the work of Langacker and others as
just such an alternative that is established not on Chomsky-style formal lan-
guages but rather “on the basis of psychologically meaningful constructs such
as symbols, categories, schemas, frames, images, discourse perspectives, and
the like” (1995: 149). It is of interest to note that Pinker, along with Jackendoff,
has come around to CCG, which they call a “substantial rearrangement of the
grammatical universe” (Jackendoff and Pinker 2005: 220).
Just as there are many constructivisms and constructionisms in social and bio-
logical theorizing these days, so there are any number of approaches to gram-
mar in linguistics that may called be constructionist. In general, (grammatical)
constructionists understand constructions to be form–meaning pairings, e.g.,
the ditransitive construction Subj V Obj Obj2, which has a counterpart expres-
sion employing ‘to’ or ‘for.’ The two possibilities are exemplified by: Mary sent
John a book and Mary sent a book to John. For constructionists, the interest in
this pair of utterances does not lie in the fact that they are paraphrases. Rather
it lies in the many, often subtle ways they differ from one another either seman-
tically or pragmatically (or both). One such difference involves the pragmatic
conditions that determine whether the recipient argument (here: John) and/
or the theme argument (here: a book) is likely to be pronominal or not. It is
remarkable to discover that, in both corpus and experimental studies, the theme
argument of the ditransitive strongly tends to be new information, that is, it is
not already given in the discourse context, and this tendency stands in con-
trast to the recipient argument which is typically pronominal. In other words,
“the recipient argument of the ditransitive construction rarely introduces a new
argument into the discourse” (Goldberg 2006: 139). This observation accounts
for why She gave him a book is fully acceptable and why She gave a man them
sounds so odd. The alternative construction with to is not so constrained. She
gave a book to him and She gave it to a man are equally good.
In addition, for constructionists, it is the construction itself that is the primi-
tive unit of syntactic representation and neither so-called rules nor grammat-
ical categories such as verb, noun, adjective, subject, object and the like. Now,
aspirant language users, i.e., young children, (in English) tend to exhibit verb-
centered conservatism, meaning that they will readily substitute new nouns into
particular and preferred general-purpose verbal frames. There is also evidence
that adults retain much verb-specific knowledge. However, one of the many rea-
sons to like CCG is that it can be shown that constructions are sometimes better
predictors of overall meaning than many verbs. Goldberg notes that “when get
appears in the VOL [verb-object-location] pattern, it conveys caused motion, but
when it appears in the VOO [verb-object-object] pattern, it conveys transfer:
(1) a. Pat got the ball over the fence.
get + VOL pattern →‘caused motion’
b. Pat got Bob a cake.
get + VOO pattern →‘transfer’” (Goldberg 2006: 106)
It is pertinent to note that the strings VOL, VOO, and Subj V Obj Obj2 exist for
ease of notation only, as nothing more than a way to represent a semi-abstract
cognitive schema. The point here is that our judgments about She gave a man
them and how we interpret one kind of get-phrase from another do not come
from consulting an internal grammar that operates on sentences out of context;
rather we are responding to the match or mismatch of a construction in com-
parison to the large experience we have had of similar constructions heard in
contexts that have built up semi-abstract schemata to which cling chunks of
semantic and discursive triggers.
Goldberg’s most recent work in CCG is devoted to understanding how
people generalize over individual items and even exemplars, but she neverthe-
less recognizes that there is more item-specific knowledge in an adult grammar
than has been acknowledged in generativist approaches (cf. Tomasello 1999:
141–42). CCG further differs from generativist approaches in that the latter
have long separated the semi-regular and unusual (unpredictable) patterns in

the so-called periphery from the predictable patterns in the so-called core of
language, whereas CCG contends that “it’s constructions all the way down”
(Goldberg 2006: 18), meaning that patterns are stored as constructions even if
they are fully predictable as long as they occur with sufficient frequency.4 That
is to say that they are learned and not prefigured in Universal Grammar.
Goldberg notes that “certain mainstream generative grammar frameworks”
seem to share the basic idea that a construction is a conventionalized form–
meaning pairing and, to the extent that syntax plays a role in contentful meaning,
these frameworks could be and are often referred to as constructionist. However,
she distinguishes her CCG and related proposals from those of the generativists,
aka Syntactic Argument Structure (SAS), by way of this helpful list:
1. SAS accounts do not adopt a non-derivational (monostratal) approach to
syntax.
2. They do not emphasize speaker construals of situations; the emphasis is
rather on rough paraphrases.
3. On SAS accounts, “constructions” are pairings of underlying form and
coarse meaning instead of surface form and detailed function.
4. On SAS accounts, only certain syntactic patterns are viewed as instances of
constructions; words or morphemes are assumed to be stored in a separate
component, and most syntactic generalizations are assumed to make no ref-
erence to semantics or function.
5. According to SAS accounts, constructions are assumed to be universal and
determined by Universal Grammar.
6. SAS accounts have not addressed language-internal generalizations across
distinct but related constructions.
7. On SAS accounts, constructions are assumed to be compatible with
Minimalist architecture and assumptions instead of providing an alternative
way to view our knowledge of grammar. (2006: 205–206)
Based on these differences, I will take the term cognitive construction gram-
mar to exclude SAS accounts. I will add that while CCGers do appeal to their
intuitions to make grammaticality judgments, they rely on corpora to deter-
mine how frequent certain patterns are and which constructions occur with the
widest variety of types because these factors will affect how likely construc-
tions are to be extended.
The relevance of CCG to Linguistics and Evolution is not surprising in that
CCG continually: (i) relates grammatical strategies to non-linguistic cognitive
4
On this point Goldberg distinguishes herself from, e.g., Jackendoff, who wants to preserve a role
for traditional syntax and argument linking that is nonconstructional (Goldberg and Jackendoff
2004: 548).
strategies; (ii) situates utterances in their discursive and pragmatic frameworks,

i.e., is usage-based; (iii) takes a developmental perspective in that both the
historical axis and the ontogenic axis are of interest in understanding how
constructions change over time and how children learn them. Goldberg puts
the last point thusly: “The aim of [constructionist] frameworks is to represent
grammatical knowledge in such a way that it can interface transparently with
theories of processing, acquisition, and historical change” (2006: 215); and (iv)
employs a vocabulary of grammatical construals that suggests an important
slice of the microdynamics involved in the kinds of cognitive orientings that
languaging induces. All this is to say that constructionists thankfully put the
traditional speaker back into the grammatical picture and into the grammar-
external world and then language about constructions in a way that is continu-
ous with the theoretical perspective of the present book.
In relating grammatical strategies to non-linguistic cognitive strategies, CCG
reintroduces the terms analogy and analogical extension to linguistic theorizing
in a way that harks back to Bloomfield. There is renewed appreciation for the
dynamics of analogy in making both linguistic and non-linguistic judgments –
in drawing inferences, making predictions and extending patterns – which goes
hand in hand with the recognition that language users are influenced by the
relative frequencies of particular constructions to which they are exposed and
the kinds of properties that those constructions exhibit. However, if CCG is a
neo-Bloomfieldian structuralism, then it is a sophisticated form of it, since it
benefits from: (i) the intervening decades of research in cognitive psychology;
(ii) the state of the post-Chomskyan world where the details of many syntactic
phenomena left unnoticed or, at least, undiscussed in Bloomfield’s time have
come into full view, e.g., the ditransitive and subject-auxiliary inversions; and
(iii) the existence of corpora and language acquisition databases unimaginable
to Bloomfield and unavailable to early transformational grammarians that are
now often only a google away.
Results produced by cognitive psychologists relating to non-linguistic
exemplar-based models of, e.g., priming and counterfactual reasoning, can be
shown to relate to actual corpora regarding linguistic priming and counter-
factual reasoning. An interest in the form of learning called priming has long
been alive and well in neurolinguistics. For instance, it has been shown that
amnesiac patients are not limited to learning motor skills; they are also able
to improve performance on perceptual tasks through priming, that is, where
recognition of words or objects is facilitated by prior exposure to the words
and visual cues. In short, subjects recall cued items better than other items
for which no cues have been provided (Kandel et al. 1995: 654). Turning to
recent work on syntax and non-amnesiac language users, Goldberg notes that
“the existence of structural priming may be an important factor underlying the
fact that there are generalizations in language. The same or similar patterns are
easier to learn and produce. At the same time, priming of course is not particu-
lar to language – repetition of the same motor programs also leads to priming
effects” (2006: 121). Effects of priming can be seen not only in linguistic and
non-linguistic behavior but also in their intersection.
The point here, however, is not that CCG is a kind of recycled pre-Chomskyan
behaviorism, because, as has just been said, in the past fifty years linguists and
language theorists have not only altered what they considered significant to
discuss and explain, they have also come to a more layered understanding of
the different kinds of cognitive effects that produce and respond to the number
and kinds of utterances language users respond to and produce. A case in point
is Goldberg’s explanation of how generalizations are constrained, and such an
explanation must be undertaken in order to counter the generativist claim that
evidence for non-occurring forms does not exist, i.e., there is no positive evi-
dence against them, and therefore the constraints against producing such non-
occurring forms must already be there in the Universal Grammar. As suggested
above, the ditransitive construction is distinct semantically and pragmatically
from the construction using the prepositional phrase. Thus, She explained me
the story, which does not (often) occur and which most of us would judge to
be ill-formed, is not in any simple way preempted by the occurrence of She
explained the story to me, for which there is much positive evidence. Goldberg’s
argument for this case centers on counterfactual reasoning, and she argues for
a statistical form of preemption: “In a situation in which construction A might
have been expected to be uttered [She explained me the story], the learner can
infer that construction A is not after all appropriate if, consistently, construc-
tion B [She explained the story to me] is heard instead” (2006: 96).
Pertinently, Goldberg backs up the general ability for counterfactual reason-
ing by citing a study that young children are capable of this kind of logic in
a non-linguistic domain, that of anticipating whether an adult would turn on
a light switch using her head instead of her hands and mimicking the adult’s
action. When the adult’s hands were free, 69 percent of babies mimicked the
adult’s head action to turn on the light, while only 21 percent did so when the
adult’s hands were occupied. In addition to underscoring how readily humans
imitate, this experiment suggests infant reasoning to be that if the experimenter
had her hands full, she had to use her head, but if her hands were free and she
used her head, she did so for a reason. Goldberg writes: “Similarly, the reason-
ing required for pre-emptive processes to work is ‘if the person meant to use
the other formulation she would have; therefore, since she didn’t, perhaps she
used the alternate formulation for a reason’” (2006: 96). Statistical preemption,
as a method by which indirect negative evidence can be gathered by learners,
is related, of course, to frequency patterns, such that low frequency or seman-
tically or pragmatically specialized constructions resist preemption, given the
fact that more specific items are preferentially produced over items that are
licensed but are represented more abstractly. CCG remains very close to the
surface of things.
Goldberg’s most interesting chapter concerns the Subject-Auxiliary
Inversion (SAI), because she is able to convincingly counter the generativist
claim that SAI is the premier example of a formal generalization with no func-
tional underpinning. Her analysis, when allied to Pullum and Scholz’s strong
undermining of generativist claims that SAI is crucially underrepresented in
utterances that a child is exposed to and is therefore unlearnable (2002: 36–45),
surely shows how far current work in grammatical analysis has gone beyond
the generative framework. At the same time, it also shows a possible legacy for
the generative framework in that it has left a rich residue of kinds of syntac-
tic phenomena that are interesting to analyze – even if the framework itself is
shown to be insufficient – and that yield perhaps surprisingly rich pragmatic
analyses.
Instead of seeing the SAI construction as some kind of formal idiosyncrasy,
Goldberg understands that a coherent category of functions is associated with
it and that there are no unified syntactic properties (other than the surface form
of subject-auxiliary inversion itself) that account for it. As a sentence pattern,
it is decidedly non-prototypical in English, and its non-prototypical form cor-
relates with two other features: (i) SAIs differ from prototypical sentences
(declarative, positive assertions with predicate-focus information structure) by
being non-positive declarations of one sort of another; and (ii) many SAIs,
unlike prototypical assertions, which do not depend on another clause, cannot
stand alone. These two features account for a wide range of extensions, listed
in the table below.
Goldberg works through the details of these constructions in a way that that
does, indeed, make good on her effort to represent grammatical knowledge
in such a way that it can “interface transparently with theories of processing,
acquisition, and historical change.” Just as certain constructions are in cur-
rent extension,5 so others are receding, e.g., the archaic comparative in the list
below. Goldberg represents these extensions in terms of what she calls a “halo
of constructions” that radiate out from a prototypical sentence type to include
the nine possibilities listed below (2006: 179).
No attempt to do full justice to Goldberg’s CCG has been made here, much
less to convey all the richness that exists in the field of cognitive construc-
tion grammar broadly conceived. Clearly CCG is building out at right angles
from the sensibility to fine syntactic detail that the generativists brought to
5
The got-X? construction is now being worked out in American English. Since its appearance in
a milk commercial some years ago, the Got Milk? construction has extended from everything to
a Got Mulch? banner seen at a nursery to Got Major? balloon seen at the Fall 2006 majors fair at
Duke University.
Yes-No question: Did he go?

Counterfactual conditionals: Had he found a solution, he would take time off.
Initial negative adverbs: Not until yesterday did he take a break.
Optatives: May you live a good life!
Exclamatives: Boy, is it ever hot!
Rhetorical questions: Boy, are you tired, or what?
Comparatives (optional): He has read more articles than have his classmates.
Comparatives (archaic): He has read more articles than have they.
Likewise constructions: (context: His girlfriend was worried.) So was I.
the study of language – which comment serves as a context for giving credit
where credit is due. In his determined turn toward the inside, Chomsky nearly
single-handedly prepared the way for linguists to become interested in the rela-
tionship between their discipline and the cognitive and brain sciences while,
at the same time, sadly, occluding the cognitive investigation. In addition, and
although not all linguists would go as far as Chomsky does when he says that
“more has been learned about language in the last twenty years than in the
preceding 2,000 years” (2002: 95), the kinds of grammatical phenomena that
the Chomskyan style of analysis has made visible will provide a rich legacy
for many years to come. It has been the Chomskyan’s sharp attention to syn-
tactic microdynamics that paved the way for the more psychologically realistic
construction grammars which, like Labovian sociolinguistics and the study of
phonetic change in progress, make it furthermore possible to study syntac-
tic change in progress. We keep the important generativist observations and
change the explanatory framework.
9 Revise: introductory linguistics textbooks
At the beginning of Chapter 5, I noted that the difficulty of accounting for

languaging on the evolutionary axis does not derive solely from the complex-
ity of the behavior or the variety of disciplines involved in tackling it – though
these aspects of the undertaking surely do pose distinct challenges. Rather,
the difficulty of the problem lies in the fact that the question of the evolution
of language – or, rather, the question of the instantiation, development and
maintenance of the human languaging network – has some of the most histori-
cally embedded conceptual baggage to unpack before one can move forward.
In fact, the very subject matter of the discipline of linguistics – usually called
language – is similarly bedeviled.
Throughout Linguistics and Evolution, I have used historiography as a kind
of talk therapy for the discipline and, over the course of the chapters, I have
effected some chiropractic adjustments. First, I enlarged the scope of linguistic
history by weaving a minor tradition of thought about our subject matter into
our history, that minor tradition being concerned with “languaging as an ori-
enting behavior.” Second, I argued that a shift toward understanding of our sub-
ject matter not as language but as languaging eases the incorporation into the
discipline of the latest findings in the social and biological sciences. Third, I
outlined the abandonment of a rationalist-realist framework and offered devel-
opmental systems theory in its stead. Because developmental systems theory
factors in time, both ontogenic and phylogenic, into its very framework, the
anxiety of how to account for languaging on the evolutionary axis does not
become easier but it does become less problematic, and ontogenic explanations
open onto the whole of a languaging living being’s lifetime. If what has been
said throughout these chapters has a use, then I would like to think it could be
in the service of reshaping introductory linguistics textbooks.
The problem: theoretical disintegration

The concern in this last chapter is to imagine the ways that the field may take
shape in introductory linguistics textbooks in the future. But why bother with
them, other than out of concern for the education of the young minds that
260
Problem: theoretical disintegration 261
come into linguistics classrooms? The answer is because introductory text-

books serve a useful and, perhaps, unintended purpose: they can be reliable
guides to problems in the current state of a discipline’s theory and practice.
John Searle has observed, “Often … we can find out more about what is going
on in a culture by looking at undergraduate textbooks than by looking at the
work of more prestigious thinkers. The textbooks are less clever at conceal-
ment” (1998: 20). Searle most likely means that these texts, in presenting
material whose complexities must necessarily be sifted out for beginning
students, more readily expose to experienced observers the bare bones of a
discipline’s theoretical problems. The problem is that introductory linguistics
textbooks attempt to do too much by trying to survey the increasing number
of legitimate topics, e.g., Language and Brain, Language and Evolution,
Language and Society; and at the same time, they do too little in that they do
not pause to reimagine the whole. What they end up presenting is a fractured
understanding of our subject matter that the students may or not be aware of
by the end of the introductory course.
The current problems in introductory textbooks can be traced to two major
developments in the field of linguistics in the last fifty years whose implications
have been undertheorized in standard introductions to date. The first pertains
to the newer understandings of linguistic form and function that issue from
discourse analysis, which is allied both to pragmatics (the study of contexts)
and to speech act theory (an account of the conditions of intelligibility: what
it means to mean in a community). It can be said that up through the 1950s,
linguists had been working to identify the sound patterns and the word forma-
tion patterns of the various languages of the world. What seemed to be just
beyond the linguist’s grasp was the ability to capture regularities that obtained
over larger stretches of linguistic forms, namely sentences. In specifying the
patterns that could account for sentence formation, Chomsky – arguably the
most famous linguist in the world of the past half-century – intended to take
linguistic theory, as it is often said, “to the next level.” As it turned out, adding
a Chomskyan approach to syntax could be done without disturbing the order of
presentation of the material in introductory textbooks, the basic organization of
which extends to Leonard Bloomfield’s masterful Language (1933). That is to
say that the addition of a Chomskyan-style syntax required only updating the
chapters that Bloomfield entitled “Sentence Types” and “Syntax.”
Now, as we move farther into the twenty-first century, the field of linguistics
is in a situation best described in the words of Barbara Johnstone, who notes
that “linguists have long been interested in the structure of words (morphology)
and sentences (syntax). Discourse analysis has moved the description of struc-
ture up a level, looking at actual stretches of connected text or transcripts and
providing descriptions of the structure of paragraphs, stories, and conversa-
tions” (2002: 5, emphasis mine). This time around, however, the incorporation
of newer approaches to usage-based construction grammar into an introductory

exposition of the field cannot be responsibly done by adding yet another chap-
ter entitled “Pragmatics” to follow the chapter entitled “Syntax.” The point is
that most linguists have learned the lesson that syntax – and all other linguistic
phenomena, for that matter – cannot fully be understood apart from context;
and this same lesson renders the long-conventional organization of introduc-
tory textbooks to be little more than a convenient pretense.1
The second development of the last half-century concerns the fact that
linguists have increasingly recognized the importance of producing a coher-
ent evolutionary account of the emergence, development and maintenance of
language in the human species. Here again, we encounter Chomsky’s influ-
ence, and the discussion of human language/ing abilities and evolution in
introductory textbooks usually notes that the topic is said to be controversial,
after which a kind of nature versus nurture story gets told with emphasis put
on whichever so-called side of the so-called divide the textbook authors are
most comfortable with. However, at this point, it should be said that the topic
remains controversial only if one is willing to ignore constructivism and to
defer incorporation of all the information about evolutionary biology and cog-
nition that now exists. This does not mean that Linguistics and Evolution is
perfect (it is surely incomplete), that all the information presented herein is
accurately understood by the present author and/or is uncontroversial, or that
its arguments are unassailably watertight. It is only to say the workable alter-
native to understanding human languaging presented here does suggest the
lines along which a future, theoretically (re)integrated, introductory linguis-
tics textbook might be written.
The solution: stepping up to the Austinian challenge

Historiographers do not believe that reading the historical record of linguis-
tics will yield truths about our subject matter that have been forgotten. The
activity is rather more of a method, a way of understanding our subject matter
by reviewing the variety of ways it has been approached and theorized over
time. Nevertheless, there are nuggets of wisdom to be gleaned from the old
texts, one such gem being Ferdinand de Saussure’s remark in the Cours de
linguistique générale that “it is often easier to discover a truth than to assign
to it its proper place” (1959 [1916]: 68). I was put in mind of the observation,
some years ago, when I encountered a good third of the way through such a
standard text as Fromkin and Rodman’s An Introduction to Language, the
1
The term context has a wide variety of meanings. A valuable discussion of it can be found in
David McNeill’s Gesture and Thought and can be summarized as “the background from which
a psychological predicate is differentiated” (2005: 107).
Solution: stepping up to the Austinian challenge 263
unremarkable yet startling statement: “Actually, every utterance is some kind

of speech act” (2003: 215).2 The question immediately arose: what would a
linguistics textbook look like if the central insight of speech act theory were
found on page one?
Introductory textbooks have certainly incorporated the innovative speech act
terminology found in J. L. Austin’s How To Do Things With Words (1962).
Students are duly introduced to the useful distinction between such terms as
constative and performative; to the designation of performative acts in terms of
being felicitous or infelicitous, and to the subtleties involved in distinguishing
the locutionary from the illocutionary from the perlocutionary. Nevertheless,
the textbooks are only halfway there, because they have not yet met the full
challenge that Austin’s work presents to the discipline. This challenge can be
summarized in terms of the following two questions:
1. What is the basic unit of analysis of speech act theory?
2. What is the purpose of issuing utterances?
Two more questions can be asked here, although nothing in How To Do Things
With Words makes one think that Austin necessarily cared about them. They are
brought up here because I care about them and because speech act theory offers
interesting perspectives on them. They are:
3. What phenomena are properly included in a linguistics derived from speech
act theory?
4. What is the most central and/or most primary feature of languaging derived
from a speech act framework?
I address each in turn.
What is the basic unit of analysis?

First and foremost is the fact that the basic unit of analysis for Austin is not the
sentence in isolation but rather the speech act episode, as mentioned already in
passing in Chapter 8. At the end of Lecture IV, at a time in his argument when
he is still trying to maintain a distinction between constatives and performa-
tives, he writes: “We must consider the total situation in which the utterance is
issued – the total speech-act – if we are to see the parallel between statements
and performative utterances” (1962: 52). By Lecture XI, not only has the dis-
tinction between them collapsed but the parallels have, too, and we find the ori-
ginally and supposedly larger class of constatives swallowed by the originally
and supposedly smaller class of performatives. Austin says: “Once we realize
2
This sentence does not appear in the 9th edition of this text in the section on speech acts (2011:
215–16).
that what we have to study is not the sentence but the issuing of an utterance
in a speech situation, there can hardly be any longer a possibility of not seeing
that stating is performing an act” (1962: 139). Unable to put the genie back in
the bottle, Austin concludes that statements such as “The cat sat on the mat”
are every bit as performative as utterances such as “I bet.”
As for the reason why statements have gotten a bye, so to speak, from the
grammarians and philosophers all these years, Austin says, in perhaps the most
insightful passage of the entire argument:
The most that might be argued, and with some plausibility, is that there is no perlocu-
tionary object specifically associated with stating, as there is with informing, arguing,
&c; and this comparative purity may be one reason why we give “statements” a certain
special position. But this certainly would not justify giving, say, “descriptions,” if prop-
erly used, a similar priority, and it is in any case true of many illocutionary acts. (1962:
139–40)
Thus, while in stating something we may be performing perlocutionary acts

of all kinds, those acts are not as conventional or as controllable, perhaps, as
are the perlocutionary acts typically secured by other types of speech acts.
Nevertheless, statements are as bound to contexts for their appropriate uptake
as are all other types of speech acts. Here Austin introduces the notion of the
dimension of assessment – which was a useful tool in Chapter 4 in unpack-
ing the continuity/discontinuity conceptual complex in evolutionary explana-
tions – and argues that the statement as it is typically understood is responsible
to the world as it is given within a particular dimension of assessment. The
statement “France is hexagonal,” for instance, cannot be assigned a true/false
value independent of a particular dimension of assessment. It is not a true-
or-false description, as Austin says, it is a rough one – one that may be good
enough, say, for a top-ranking general but not for a geographer. Since no state-
ments ever exist outside of any dimension of assessment whatsoever, there is
simply no way to interpret an utterance without also knowing the dimension of
assessment in which it was issued. For instance, without knowing the dimen-
sion of assessment in which the words “The rock of Gibraltar” are issued,
one cannot know whether they: refer to a limestone and shale peninsula at the
southern tip of Spain that Spain ceded to Britain in 1713; function as a tribute
to the strength of someone’s character; are a coded answer to a question about
what time a shipment of cocaine is coming in; and so on.
The thrust of speech act theory is, furthermore, to reverse the relationship
between meaning and context that was held by grammarians and philosophers
of Austin’s day and continues to be held by many even today. The traditional
way of thinking about the relationship is to believe that meanings stabilize
contexts, whereby things are said or assumed to work out felicitously or not
in linguistic interactions because either individuals know and understand
the meanings of words and thus respond appropriately or else they do not
know and understand them and thus respond inappropriately. The idea that
meanings are properties of words reflects the perspective of the philologist,
defined in Chapter 1 as an observer of language who stands at the borderlands
between languages. From this vantage point, the philologist is able to report to
those on various sides that the forms biblios (βιβγιος), Buch, carte, hon ( 本 ),
kniga (книга), livre, shū (書,书), etc. “mean” “book.” However, if we whole-
heartedly pursue the consequences of speech act theory, we are led to the idea
that, from the perspective of the language user, contexts stabilize meanings.
We find that so-called meanings are shaken-out effects of contexts that have
tended to work out felicitously enough over time in such a way as to stabilize
those meanings. For the language user, “I bet” comes to mean “I bet” because
uttering such a sequence of sounds in the right circumstances tends to yield
felicitous results.
Now, the problem of meaning might just as well be called the muddle of
meaning. Meaning is sometimes said to be: (i) properties of words or group-
ings of words; (ii) properties of the linguistic system, e.g., langue, or of syntac-
tic structures; (iii) properties of perception or of categories of perception; (iv)
derived from contexts; (v) effects of interactions; or (vi) a term that we cannot
do without and can do little with. Another way of understanding approaches
(i)–(vi) is to see them as answers to the (usually unasked) question: where is it
we must look to find meanings? The answers then become: (i) inside words or
groupings of words; (ii) inside the linguistic system; (iii) distributed inside our
gray matter as a whole; (iv) in the speech event itself; (v) in the results of inter-
actions; and (vi) in a long history of traditional conceptions of language, both
in commonsense ideas (so-called folk linguistics) and in the formalizations of
those ideas in the academic discourses of language (hermeneutics, linguistics,
logic, rhetoric, philology, the philosophy of language). Many theories combine
approaches. Discourse analysts might combine (iii) and (iv), while a range of
semioticians combine (ii)–(v). Traditional linguists have tended not to look
beyond (ii), and some do not stray very far from (i).
The major dividing line exists between those linguists who look for (and
find) meanings within the linguistic system and those who look outside the sys-
tem for explanations, the latter often called functionalists. In their classic work
Politeness, linguistic anthropologists Penelope Brown and Stephen Levinson
map out the functionalist approach:
The major justification for the bifurcation of the theory of meaning into semantics and
pragmatics was the basic Gricean observation that what is “said” is typically only part
of what is “meant,” the proposition expressed by the former providing a basis for the
calculation of the latter. In this perspective, indirection, together with related kinds of
mismatch between the said and the unsaid, is a central phenomenon, and has received
much technical attention. But why does the phenomenon exist at all? It was that moti-
vational question that our politeness theory was specifically designed to answer. (1987
[1978]: 49)
Their work seeks to account for some of the grammatical features of polite-
ness found in the world’s language in sources outside the purely linguistic
system, e.g., in the principles of cognitive processing or in the interaction of
language with other mental faculties. Distinguishing their work from what they
call “‘internal’ cognitive functionalist explanation,” Brown and Levinson are
interested to link linguistic structures to the organization of communication,
that is, to “‘external’ pragmatic theories” (1987 [1978]: 255–56).
Their approach opens up a new social cognitive space (indirection, implica-
ture) and a new site where meanings are negotiated (communicative interaction,
social pressures, usage). They hypothesize that some aspects of usage – namely
pragmatic inferences that are implicatures – are likely to acquire structural cor-
relates, and external functional pressures are always mediated through usage,
thereby creating a continuous feedback loop between form and meaning. For
instance, a phrase like Could you tell me the time? has come to be conven-
tionally understood as a request to be told the time rather than as a question
about a person’s ability to tell the time, and it arose historically as an output
of a politeness strategy. “Extended patterns of usage feed back into structural
properties,” Brown and Levinson write, “either through changes in meaning
or directly through formal reflexes of extended usages. Implicatures play a
crucial role in the mechanism (either through causing semantic shift or through
triggering form reflexes), and thus in language change” (1987 [1978]: 261).
Honorifics provide a case in point: the Japanese pronoun watakusi (‘I’ most
formal) originally meant slave or servant, although modern Japanese speak-
ers do not need to know or care that this is the case. The original meaning led
to the use of the term, while extended usage stabilized the form in its current
pragmatic function.
Approaches (v) and (vi) are the most radical. Approach (v) is Skinner’s. In
the present context it is pertinent to note that for Skinner “Meaning is not a
property of behavior as such but of the conditions under which behavior occurs.
Technically, meanings are to be found among the independent variables in a
functional account rather than as properties of the dependent variable” (1957:
13–14). (Note: here the term functional does not mean “outside the linguistic
system” but rather “to show how something works.”) In contrast to the well-
known classical conditioning model of Ivan Pavlov, which establishes con-
ditions under which an organism learns a relationship between two stimuli,
Skinner’s form of operant conditioning establishes conditions under which an
organism learns a relationship between a stimulus and the consequences that
the organism’s response to that stimulus has for the organism. Skinner was
interested in the strength of the relationship between antecedent stimuli and the
behavior that the stimuli come to occasion for the particular organism. In this
sense, then, Skinner, too, opened up a not purely linguistic space to explore sig-
nificance or meaning – a point with which he is not usually credited. Approach
(vi) characterizes the work of Barbara Herrnstein Smith, whose work is congen-
ial (but not equivalent) to that of the Soviet-Russian line Vološinov–Jakobson–
Lottman, along with the work of Austin, Skinner and Goffman. She is engaged
in elaborating accounts of interactive practice that place doing, rather than
meaning, at the center of theories of language (1997: 52).
Clearly, my druthers are for approaches (iii)–(v), and if I were to embrace
(vi) wholeheartedly I would say that introductory linguistics textbooks hardly
need to go much beyond foregrounding Jakobson’s metalingual function of
the verb to mean and to note that it occupies a large place in the verbal behav-
ior of preschool children (and curious adults) who want to know “What does
that mean?” The idea here is to get beyond the traditional view of meaning as
some kind of substance that can be accumulated, stored or lost; and that can
be psychologically described as “‘association,’ or ‘habit strength’ or ‘content’
at a mental address” (McNeill 2005: 107). No doubt there are many interest-
ing things to say about what might be called comparative semantics, i.e., the
different ways that various groups of languagers may organize their cognitive-
behavioral domains through their languaging together. But even here the term
comparative semantics seems to be pretty thin, when what is needed is an
understanding of the cognitive-behavioral domain as a whole of any particular
language user, and this is an understanding that cannot be easily gleaned from
grammar books, as Levinson points out (2003: 14). What I do mean to suggest
here is that Austin and his speech act theory cracked open the door on what had
become a rather claustrophobic domain of inquiry, and he let in some fresh air.
The task of the future is to open the door the whole way.
A significant aspect of Austin’s analysis is the way that a discontinuous tem-
poral range is built into the notion of the total speech act. If I promise at 2:00
to meet you at the bus stop at 5:00 or if I bet you ten dollars that it will rain
tomorrow, the total speech act encompasses in the first case at least three hours
and in the second case up to twenty-four hours. In those intervening hours it
is likely that I will engage in other, perhaps many other, speech act episodes
whose temporal ranges are likely to be discontinuous such that they will wrap
continuously around one another, turning my hours and days into long braids
of speech act episodes. This description neatly captures for me the embodied,
embedded experience of languaging.
Finally, a focus on the total speech episode brings into play the crucial issue
of uptake, a term which nicely nuances the social shaping of what may be called
a linguistic innovation and which has been used throughout the present book.
Austin notes that there are cases where someone initiates a procedure and that
sometimes he might get away with it. His example is from football and “the
man who first picked up the ball and ran. Getting away with things is essen-
tial,” Austin writes, “despite the suspicious terminology” (1962: 30). What is
required next is that the social group as a whole acts to accept the innovation or
not. All this is to say that, in Austin’s account, individuals can make up any new
maneuvers or social customs and, by extension, any new words or phrases they
please, but these maneuvers, customs and/or words and phrases do not become
part of any particular group until there is sufficient uptake in the community.
That is, an innovation is not an innovation until the community recognizes it as
an innovation. If there is no uptake in the language game, for instance, linguists
call such words and phrases nonce (= used only once) creations.3
What is the purpose of issuing utterances?

Although Austin does not come right out and say this, it is fair to state that
he does not think we say things because they are true; rather we say things
to have effects. Certainly we go around saying true things all the time. When
you say something like “It’s raining,” it is probably true that it is raining.
But you do not say it in order to make a true statement about the weather.
Rather, you say it because you want your roommate to know to take an
umbrella or you are sorry to be canceling the picnic or you are pleased that
the flowers you just planted are getting the water they need. Similarly, if
you say: “There are twenty chairs in the classroom,” you are not likely to
be making a truth claim about the number of chairs in the room. It is more
likely that you are involved in some episode where the number of chairs is at
issue, say, that twenty-one students have shown up for my class or when the
teacher next door needs to borrow one and I know there are fifteen students
in my class, so I can spare a few. And so on.
At the end of Lecture XI, Austin is willing to give an out, so to speak, to the
person who deep-down wants to preserve some notion of the non-performative
constative when he notes that the performative-constative extremes could be
exemplified by such utterances as “I apologize” or “‘The cat sat on the mat’,
said for no conceivable reason” (1962: 146). We can easily imagine the many
conditions under which the words “I apologize” are uttered and what their
intended effects must be, and thus they are prototypically or, it could be said,
purely performative. What makes “‘The cat sat on the mat,’ said for no con-
ceivable reason” the limit case of the constative is the phrase said for no con-
ceivable reason. The pure constative would be the utterance issued just for the
3
As many linguists have pointed out, the morphological process du jour in American English
noun formation is blending and is everywhere from names of Hollywood couples (Brangelina)
to dogs (labradoodle) to flavors (strawbermelon) to clothing (tankini) to cultural productions
(rockumentary, shockumentary, mockumentary and mockbuster). It is widespread in adjectives
as well (geektastic, adorkable, ginormous). While not all these will survive, I think that the blend
bridezilla has a good chance of remaining. A blend such as Boozachusetts, as uttered by a radio
DJ in the phrase “We’re gonna party in the Caribbean and turn Jamaica into Boozachusetts” has
the ring of a nonce form. The insult-as-endearment form, Masshole, to refer to an inhabitant of
Massachusetts may, however, have more staying power.
heck of it, the utterance that has no motive for being uttered and no uptake in
being uttered.
The question is: are there any utterances that are truly uttered for no reason
whatsoever? My answer would be: “No, not really. Now, not all utterances are
issued for deep and psychologically fascinating reasons, but there is always a
reason, albeit sometimes a completely trivial one. ‘Just for the heck of it’ is a
reason in and of itself.” And in the case of “the cat sat on the mat,” it is doubtful
that Austin pulled the example out of thin air, that is, invoked it for no conceiv-
able reason.4 Alternative answers to this question are available, as we have
already seen, where the question “what is the purpose of communication?”
came with the answer “to serve as an instrument for the free expression of
thought.” Here I am willing to interpret the phrase “free expression of thought”
as an equivalent of an utterance said for no conceivable reason.
What phenomena are properly included in a linguistics derived

from speech act theory?
Austin is interested in speech acts but not necessarily in speech as such, and so
he is perfectly willing to include in his analysis wordless actions. For Austin
(and for Skinner, as was noted in Chapter 8), words and actions are potentially
continuous. He thereby challenges traditional linguistic approaches, which,
while they certainly recognize the universal human propensity of gesturing
and are perfectly willing to explore cultural similarities and differences in ges-
turing systems, tend to view the analysis of words and the analysis of actions
as discontinuous, i.e., belonging to separate realms.
In recent decades, gesture studies have moved far beyond the examination of
what are called emblems, the “thumbs-up,” the “okay,” or the ceremonial bow
mentioned by Austin. In Hearing Gesture: How Our Hands Help Us Think
(2003), Susan Goldin-Meadow argues that the gestures we produce while
speaking are a form of thinking, that they are integral to communication, and
that the interface between gesture and spoken language is completely porous,
in that what gets visualized in gesture can end up later in speech and what
gets said in speech can end up later in gesture. She furthermore has evidence
4
Austin’s choice of the phrase “The cat sat on the mat,” which he claims one could say for no
conceivable reason was, perhaps, not chosen without a conceivable reason. For many years, the
phrase was the first sentence that English children used to copy out of copybooks in the bad old
days of nineteenth-century education. It was one of the first sentences that American children
were exposed to as well, since it was included in the McGuffey’s Primer at the turn of the twenti-
eth century. By the time Austin was putting together the lectures that made up How To Do Things
With Words, the sentence was already reverberating in philosophical discourse, usually in con-
nection with the so-called picture theory of meaning – what constatives are supposed to do, after
all – namely, mirror the world. A quick google of the phrase gives an idea of how widespread
it has become as an exemplar of a beginner’s sentence and as a canonical non-consequential
utterance.
to suggest that gesture improves memory, that gestures can prime speech, and
that speech–gesture mismatches in children are a sign that the child is ready to
learn something new. The idea here is that children are able to try out ideas in
bodily fashion that they are not yet able to articulate.
David McNeill’s Gesture and Thought (2005) takes a bolder conceptual
step by theorizing gesture and speech as two contradictory modes of think-
ing: gesture is holistic, imagistic, idiosyncratic, unconscious and (more or
less) motivated, while speech is analytic, linear, more regulated by conven-
tion, more conscious and (more or less) unmotivated (in the sense of the
arbitrariness of the linguistics sign). Like Goldin-Meadow, McNeill affirms
that gesture is co-expressive with speech and non-redundant. He makes use
of Vygotsky’s notion of the Growth Point to propose that there is a specific
starting point for unitary thought and that both speech and gesture are fun-
damental to unitary thought. He also describes the Growth Point as being
characterized by a benign instability, meaning that the two modes of think-
ing – gesture and speech – in being so unlike, must separate themselves in
order for the thought to come into expression. In short, the Growth Point is
both image and linguistic categorical content, that is, “an image with a foot
in the door of language” (2005: 115). In fact, as McNeill says in his open-
ing pages, the main theme of his book is that language is inseparable from
imagery, a statement he attributes to neuroscientist Antonio Damasio. Here,
Austin’s observation about the speech-act nature of certain gestures serves as
an easy springboard to incorporating into an introductory approach to linguis-
tic theory and practice the important work on the gesture–speech connection
coming mainly from psychology. It has also allowed in the multidimensional
space of conceptual imagery.
What is the most central and/or most primary feature of

languaging derived from a speech act framework?
To give the question an Austinian flavor, the extreme marginal answers to this
question can be said to be, on the one hand, that language users are able to pro-
duce an infinite number of (potentially infinitely) long novel sentences uttered
for no particular reason and, on the other hand, that languaging living beings
are able, in uttering the words, “Please pass the salt,” to engage the muscle
power of another languaging living being such that the latter will reach out his
hand, grasp the salt shaker and pass it to the requester, perhaps also uttering the
words “There you go. Here’s the salt.”
It is conceivable that Austin might allow that the so-called free (and poten-
tially uninterrupted) expression of someone’s thoughts is possible under the
right circumstances – e.g., when one party in Congress wants to delay or dis-
rupt a vote, when a Dane is facing a life-or-death dilemma, when a depressed
Ontogeny: applied Austin 271
patient shows up for an appointment – and with the right audience, e.g., mem-
bers of Congress, either no one else on stage or an entire theater listening to
Hamlet’s soliloquy, a psychotherapist. However, under normal circumstances,
we do not simply hold forth, and this for a variety of reasons: we run out of
breath, we don’t have anything more to say, we are interrupted, we feel that
it is someone else’s turn, we have to get going, we have something else to do.
Clearly, a speech act framework favors a vision of our languaging interactions
as, by definition, actions, that is to say interactions, that is to say consequential
involvements of language users with one another, going about our days getting
our business done with one another, where the consequences of that business
may be as trivial as passing the salt.
The word primary in the phrasing of this question was inserted in order to
trade on the sense of prior. Neither in phylogeny (as discussed in Chapter 5)
nor in ontogeny (as discussed in Chapter 6) does the so-called free expression
of thought precede the consequential interactions that the members of our spe-
cies have had with one another throughout time and that any particular member
born into one of our communities has as s/he becomes a languaging living
being. The ability to hold forth as well as the stamp of individual creativity that
may be put on any utterance of whatever length are results of interactions, not
the preconditions of them.
Austin’s work fits effortlessly into the languaging-as-an-orienting-behavior
line of thinking and talking about our subject matter. Just as the potential sting
of including Skinner in the disciplinary past of linguistics is blunted by com-
paring him favorably to someone with disciplinary approval, namely, Austin,
so Austin can do further service in our introductory linguistics textbooks. That
is because the glass is half full. He has already been let halfway into our text-
books. The next thing to do is to put speech act theory on page one and unspool
the ontogenic and phylogenic stories from there.
Ontogeny: applied Austin

Applying Austin to a long and ongoing debate in the discipline, the question is:
is learning one’s native language difficult or easy? One answer to the question
was provided by Jespersen over eighty years ago:
Many people will say it [learning a language] comes by “instinct,” as if “instinct” were
not one of those fine words which are chiefly used to cover over what is not under-
stood, because it says so precious little and seems to say so precious much. But when
other people, using a more everyday expression, say that it all “comes quite of itself,”
I must strongly demur; so far is it “coming of itself” that it demands extraordinary
labour on the child’s part. The countless grammatical mistakes made by a child in its
early years are a tell-tale proof of the difficulty which this side of language presents
to him. (1922: 128)
Jespersen’s entire account, beginning with the child’s first attempts at sounds,
which gets a subheading labeled “Mutilations,” is about the difficulties and
obstacles the child faces when learning his/her native language.
A more recent answer to the question is offered by Steven Pinker in The
Language Instinct (which he titled without, evidently, consulting Jespersen’s
opinion of the term instinct):
The three-year old … is a grammatical genius – master of most constructions, obeying
rules far more often than flouting them, respecting language universals, erring in sen-
sible, adultlike ways, and avoiding many kinds of errors altogether. How do they do it?
Children of this age are notably incompetent at most other activities. We won’t let them
drive, vote, or go to school, and they can be flummoxed by no-brainer tasks like sorting
beads in order of size. (1994: 276)
Pinker’s account, in contrast to Jespersen’s, stresses the ease and rapidity with
which children the world over learn their native languages(s) and how this
learning is different from other kinds of learning, e.g., sorting beads in order
of size. The idea that language comes by itself can be traced back, of course,
to Chomsky, who writes:
it seems reasonable to suppose that a child cannot help constructing a particular sort of
transformational grammar to account for the data presented to him, any more than he
can control his perception of solid objects or his attention to line and angle. Thus it may
well be that the general features of language structure reflect, not so much the course
of one’s experience, but rather the general character of one’s capacity to acquire knowl-
edge – in the traditional sense, one’s innate ideas and innate principles. (1965: 59).
Everyone agrees that most children eventually become successful representa-
tives of adult speech norms in their communities. The only question is whether
the path to that normative behavior is difficult and fraught with error or rapid
and relatively painless or whether a child cannot help constructing a particular
sort of transformational grammar, i.e., that it comes quite of itself.
Austin would say that the only way to answer the question is to understand
the dimension of assessment through which the questioner frames the answer.
For Jespersen, the dimension of assessment is the adult speech of the child’s
community. From this perspective, the child’s early articulations are anything
but smooth. For Chomsky and Pinker, the dimension of assessment is either
the second-language [human] learner after puberty or the efforts of primatolo-
gists to turn non-human primates into human language users compared to the
efforts required to turn human babies into human language users. From this
perspective, the three-year-old human does look to be a grammatical genius
compared either to an adult learner of a second language (sitting in a classroom,
perhaps) or to a three-year-old chimp (or to a computer). Whether a three-year-
old human errs in “sensible, adultlike ways” in the dimension of assessment
of an adult speaker in the child’s community is another matter entirely. Pinker
Ontogeny: applied Austin 273
evidently embraces a model of human development that assumes a child’s cog-

nition to be a miniature version of adult cognition, a position at odds with a
developmental linguistics.
The appropriate dimension of assessment through which a developmental
linguistics asks the question of the difficulty or ease of learning one’s native
language is the experience of the child (or chimp or computer) itself. Such
a view is taken, for instance, by Katherine Nelson in Young Minds in Social
Worlds (2007), where she is less interested in pronouncing on the issue of
language learning whether the child has it easy or not and is more interested
in investigating how the infant becomes the self-organized child in continuous
and dynamic adjustment to new conditions, whether internal or external.5 In
Linguistics and Evolution, it is axiomatic that the experience of the aspirant
language user (whether the child or the chimp) is neither difficult nor easy, nei-
ther slow nor rapid, neither error-ridden nor miraculously error-free. Clearly,
a child’s experience and a chimp’s experience will be very different because
of the different lineages to which they belong. However, all one can really say
is that their experience is what it is, that it may be difficult one moment, easy
another, and that it requires quite a bit of language use to be able to reflect on
the easy/difficult distinction.
Austin was not engaged with the topic of language learning, and so he
did not pronounce on whether it comes all by itself or not. Nevertheless, his
study of speech acts inspired Jerome Bruner’s work on the development of
language in children, mentioned in Chapter 6. Bruner’s work shows that learn-
ing a first language requires far more experience and help from adults than a
Cartesian-inspired account of language (and our textbooks) have been willing
to acknowledge and that learning cannot be divorced from situated contexts
involving cultural artifacts and the child’s own nonverbal activity in relation
to such artifacts. And there is no sense in which language can come by itself
because the child is never by herself. Human children become human adults
in and through their interactions with others, which always occur in particular
contexts.
It is time now to eliminate the pronouncements, oft-repeated in introduc-
tory textbooks, that first-language learning is rapid, effortless, uniform, accom-
plished by age four or five and with little exposure to inputs that are overly
narrowly conceived. Such pronouncements are tendentious, anti-developmen-
tal and essentially empty. They should no longer have a place in the way the
discipline is presented to incoming students. Our introductory textbooks are
5
Nelson makes only brief mention of the easy/difficult distinction (2007: 153 and 273, note 3).
Nelson’s study is devoted to a developmental systems theory account of how the mind undergoes
an expansion in complexity and in sources of meaning through experiential transactions (see
2007: 241).
no doubt informing the next generation of psychologists and neuroscientists.

In the future, it would be desirable for them not to be inclined to write, in a
high-profile journal such as Science and in an otherwise excellent article on the
new science of learning based on otherwise excellent neuroscientific research,
that “no computer has cracked the human speech code and achieved fluent
speech understanding across talkers, which children master by 3 years of age”
(Meltzoff et al. 2009: 286).
Now, it is well known that newborn infants are responsive to the speech of
the ambient language, but I would prefer not to call that responsiveness ‘rapid’
but rather evidence of the linguistic bath that human culture and cognition have
been steeped in for a very long time. And one can certainly make the observa-
tion that by the time most children are four years old, they are able to form gen-
eralizations over lists of attested instances of certain constructions and that this
ability suggests that these constructions have a neuronal instantiation at some
level of abstraction. However, the emergence of this ability is only one – albeit
interesting and important – moment in the entire life of the language user,
is by no means an end-point, and shows only the beginnings of the eventual
adult-like usages suggested by the term mastery. Nor is there any sense, once
again, in which language comes all by itself. The ability of a four-year-old to
form generalizations is a result of repeated and, indeed, massive exposure to
the relevant constructions.
Clearly, there is much work to be done to answer the question: “how does
the living being become the particular languaging living being that s/he does?”
It is an enormous topic that ranges over psychology and cultural anthropol-
ogy, and it feels as though I have only just opened a descriptive territory that I
have long experienced as frustratingly dispossessed and flat. Now the descrip-
tive territory feels almost overwhelmingly large because the spaces inside the
experiential languager include the layering into muscle, skin and gut of lan-
guage, perception, thinking, culture, identity, ideology and ethics, the spaces
where linguistically complex brain regions respond not only to events in the
world but also, proprioceptively, to cultural habits, skills, memory traces and
affects. This is the space of jokes and witticisms and of understanding what
lies behind the novelty cocktail napkin line “Where there’s a will … there
are attentive children” or how the cynicism of “Money is thicker than blood”
slyly rewrites the sentimentality of “Blood is thicker than water.” This is the
space of political persuasion, philosophical debate, sexual seduction, how and
even whom we greet in the street, and the I. The layers may be and often are
contradictory; an American from New York might find Southern speech ugly
but find the same twang as it turns up in Country Western music beautiful. In
all their multiplicity and contradictoriness, then, these layers inflect our loves,
our hates and every bodily disposition in between. In short, they are the stuff
of our everyday lives. They either help us or hinder us when we put one foot in
Phylogeny 275
front of the next – or into our mouth. Putting some of the richness of the neuro-
logical, cultural and ideological aspects of languaging living beings’ experi-
ence with whatever languages they language with into the opening chapters of
our introductory texts is a challenge, certainly, but one that calls out to be met.
Given the quality and quantity of work being done on far-ranging topics, from
children’s language, to gender, identity and ideology as well as in sociolinguis-
tics in general, the ontogenic script looks to have a bright and rich future.
Phylogeny
In standard introductory textbooks, the body is chopped up. A discussion of
respiration in relation to speech tends to occur in an early chapter and may be
separated by many, if not hundreds, of pages from a discussion of babbling.
The brain often gets its own chapter. The hands do not come into play, except
when the topic is ASL, while a discussion of footedness in relationship to hem-
ispheric control is simply out of the question. And what of basic genetics? Our
textbooks should reflect the practice of linguists who are increasingly recep-
tive to integrating research from related disciplines, e.g., evolutionary biology,
anthropology, psychology and the neurosciences, all of which have flourished
in the past several decades. It seems odd to have to say, one hundred and fifty
years after Darwin’s Origin, that introductory linguistics textbooks should take
the topic of evolution seriously, but there it is. The problem is that the evolu-
tionary story was written out of the discipline almost from its beginnings in
the nineteenth century, but now it is time to put it not only into our discipline’s
theoretical foundations but also into its introductory exposition. The evolution-
ary story requires that the whole body (interacting with other bodies) be taken
into account and presented as a whole. We are way past the time when we
should be trying to retrofit a given conception of language that was produced
without consideration of phylogeny or neurology to an evolutionary account or
to what is now known about the brain. It is far better to work with a conception
of language that is informed by evolution and neurology from the beginning.
The tools are in place to situate the story of the human languaging network in
the expanse of evolutionary time, and why not begin with genetics? – whether
looking at mitochondrial-DNA samplings and archeological evidence to trace
human migrations or at the relationship between ASPM-D (estimated to have
been derived with a 95% confidence interval between 0.5 and 14.1 thousand
years ago) and MCPH-D (estimated to have been derived with a 95% confi-
dence interval between 14 and 60 thousand years ago). The former allele –
which possibly alters the orientation of the mitotic spindle in certain areas
of the cortex – reaches high frequencies in Central and Western Asia, Europe
and North Africa and low frequencies in East Asia, sub-Saharan Africa, and
the Americas; the latter is very frequent in Asia, Europe and the Americas,
moderately frequent in North and East Africa, South-East Asia, and very rare
in sub-Saharan Africa. The presence or absence of these alleles, as noted in
an earlier chapter, has been correlated with the presence or absence of tone
languages. Those areas of the world where the new alleles are relatively rare
tend to be those areas where tone languages are common (Dediu and Ladd
2007). I bring up this particular example again and with this amount of detail
at the end of the book not because it is so new and promising (at the moment
it is only a hypothesis and may be subject to revision), but rather to underscore
what kind of information students of linguistics should be expected to know
in the immediate future: geography, language families, typological and areal
features, and, yes, population genetics. The detail in this paragraph should be
completely interpretable to students, rather than a jumble of unrelatable places
and statistics.
In addition, primatology has flourished in the past decades, and studies of
the facial, vocal and gestural communications of non-human primates stand
to help us humans in our understanding of our particular languaging mode
of communication. It is of interest that chest-beating displays in gorillas are
initiated with the right hand, whereas reaching for food tends to be with the
left. It is of further interest that gestures are evolutionarily younger, since they
show up in apes but not monkeys. They are also likely under greater cortical
control than facial/vocal signals (Pollick and de Waal 2007). Understanding
languaging to be a multimodal activity makes primatology more connectable
to linguistics – and potentially vice versa.
This is the place to acknowledge the great contribution of the philologi-
cal method and practice of the last several hundred years: the establishment
of eighteen or so language families for the world’s languages, although the
exact number is in dispute, and there remains a lack of consensus for the clas-
sification of a number of today’s languages, even some well-known ones with
large numbers of speakers, such as Korean and Japanese. However, most of
the world’s languages have been classified as belonging to one language fam-
ily or another, and there are surprisingly few language isolates, e.g., Basque,
which has no known structural or historical relationship to any other language,
although, for the case of Basque, at least, many have been proposed.
The current geographic distribution of the various language families reflects,
to a certain degree, the history of what might be called the first globalization:
the migratory patterns of the human groups who left Africa between 100,000
and 50,000 years ago and spread out over the world. Because we humans have
always been a restless lot, the living representatives of a language in any par-
ticular language family today may or may not be direct descendants of the rep-
resentatives of the protolanguage for that family. On the one hand, it is likely
that most living representatives of Nama (= Hottentot) have a direct cultural
and ethnic relationship to the Proto-Khoisans. On the other hand, it is doubtful
Phylogeny 277
whether most living representatives of English have any direct cultural and eth-
nic relationship to the North Germanic tribes who descended from the Proto-
Indo-Europeans. Despite the fact that lineages and language families have been
well established for the last several hundred years, this kind of information and
the maps to go with it are remarkable by their absence in introductory texts.
Maps are, to be sure, often (but not always) found in the chapters devoted
to historical linguistics.6 However, in general, there is little emphasis on this
material and a lack of perceived need to know it very thoroughly. Of course,
as the discipline developed throughout the twentieth century, what used to be
the staple of linguistics a hundred years ago, namely historical linguistics, got
crowded out of the introductory texts in favor of newer material. The idea here
is that it is time for a revival.
At the same time, I am not advocating a return to the tree model as the only
image to decorate textbook pages. In fact, I would like all but splitting fam-
ily trees to be removed. Rather, the image of the rhizome is much more apt
to capture both how languaging is laced throughout the cortex and the body
and how languages, e.g., the Romance languages, are related to one another.
The rhizome de-monolithicizes language and opens up an understanding
of origins, for example, the question of the monogenesis or polygenesis of
language(s).
The way to resolve the monogenesis/polygenesis problem is to improve the
terms of the discussion. According to the population geneticist John Maynard
Smith, around 60,000 years ago there existed about 5,000 human (breeding)
females. This means that there was probably something on the order of 35,000
humans living during that time. It is even likely, according to Smith, that the
human population before that time had been larger and that the 5,000 females of
60,000 years ago represented a reduction of the human population. Further evi-
dence suggests that the human population experienced several separate explo-
sions in size after that time, presumably by those populations that survived the
so-called bottleneck that occurred 60,000 years ago. Now, let us say that those
humans occupied several hundreds of square miles of territory in Africa, most
likely in the mid-north and east of Africa. From that area, the migrations out of
Africa would continue over the next tens of thousands of years.
It is highly likely that the human languaging network was up and running
60,000 years ago and had been for some time, even for a very long time. It is
not necessary that all humans spoke what could be identified as one language at
any one time or place 60,000 years ago. It is only necessary for some humans
and/or some groups to have been in contact with some others and those, in
6
In Linguistics: An Introduction to Language and Communication by Akmajian et al., the only
map to be found is of the state of Washington showing the geographical proximity of three dis-
tinct language families in the north-western United States (2001: 338).
turn, to have been in contact with yet others and/or the first group(s), thereby
producing long chains of communication whose interconnections do not form
any kind of tree trunk but rather a densely networked, highly scrambled rhi-
zome. It also does not matter whether groups migrated at widely different time
periods, which it seems they did. The point is not that the roots of the family
tree (or trees) of languages have been so lost to us in the far reaches of time that
we can no longer untangle them. The point is rather that the tangle itself is at
the beginning. There never was either monogenesis or polygenesis; there was
only genesis, an ongoing genesis that occurred in and through the individuals
(all 35,000 of them and everyone else before and after) who lived their lives
through a continuous drift of communicative interactions, some of which may
have stabilized enough to have perhaps, even probably, been considered lan-
guages by those who languaged with them.
A final point to make here is that our understanding of the languaging net-
work will not reduce to the evolutionary story that can be told about it. Rather,
the effort to understand our subject matter in evolutionarily consistent terms
requires making interdisciplinary cooperation an explicit methodological
premise of the discipline. Even if an individual linguist’s primary interest in
our subject matter is not the evolutionary story that can be told about it, this
methodological premise will nevertheless ensure a disciplinary platform that
is strengthened by its integration with, rather than its continuing and now out-
of-date premised methodological autonomy from, the latest findings in other
disciplines. It is an exciting prospect.
Austin gets the last word

The edges of linguistic theory and practice will always spill into neighbor-
ing disciplines. This exuberance is to be celebrated and encouraged rather
than feared and contained. For instance, my interest has long been piqued by
Austin’s observation – here making Austin the potential alpha and omega of
an introductory exposition – that when all is said and done, what is left of the
distinction of the performative and constative is this:
With the constative utterance, we abstract from the illocutionary (let alone the perlo-
cutionary) aspects of the speech act, and we concentrate on the locutionary: moreover,
we use an over-simplified notion of correspondence with the facts – over-simplified
because essentially it brings in the illocutionary aspect. This is the ideal of what would
be right to say in all circumstances, for any purpose, to any audience, &c. Perhaps it is
sometimes realized. (1962: 146)
In understanding the constative as the ideal of what would be right to say in
all circumstances for any purpose to any audience, etc., I see Austin’s move as
one in a long line of theorizing about what might be called right speech. In the
Austin gets the last word 279
eighteenth century, grammarians such as Robert Lowth7 introduced prescrip-

tive rules to show the way to right speech, namely what people should say,
not what they do say. Twentieth-century structuralists shifted from producing
prescriptive grammars to writing descriptive grammars and pronounced that
the utterances that the native speaker utters are, by definition, right. The gen-
erativists added their own layer to the discussion of right speech with contro-
versies and notations (*, ?, ??) surrounding their judgments of the grammatical
and ungrammatical and by proposing quasi-descriptive, quasi-causal rules to
account for those judgments.
Austin, in introducing an ecological dimension concerned with the right
thing to say to the right person at the right time, dares me to wonder whether
linguistics could extend itself into examining the conditions under which right
speech could be realized without losing its identity of linguistics as such. I am
flirting here with the relationship of the science of language to ethical behavior
and inquiry, clearly trading on multiple senses of the word right, as English
allows me to do. The idea is that an ethical inquiry in linguistics involves,
once again, the relationship between the observer, i.e., the linguist, and the lan-
guage experiencer, in this case also the linguist. Instead of acting as an outsider
observer, the linguist acts as an inside observer of his/her languaging behav-
iors. This activity is tantamount to entertaining yet a third question, in addition
to the first two (how does a living being become a languaging living being? and
how does the languaging living being become the particular one that it does?),
and that third question is: how am I to observe myself as an observer? I believe
that some elaboration of right speech would follow from that inquiry, and a
given instance of right speech may well depend on the language you speak.
An intriguing study showed that Arabic-Hebrew bilinguals showed more posi-
tive attitudes toward Jews when tested in Hebrew than when tested in Arabic
(Danziger and Ward 2010).
If, as I have argued, linguistics is to proceed without a theorized object of
study that is autonomous from other disciplines, it is still not the case that lin-
guistics does not have anything to offer to the social and biological sciences.
However, I would say that it is not the exportation of a particular model of
language that will now best serve interdisciplinary work; it is rather that the set
of problems that currently confronts linguists is familiar to many sister disci-
plines, such that the coming to terms with those problems may be suggestive
to researchers in those related fields. What I have in mind here, for example, is
the tension that exists between understanding one’s native language as largely
imposed and the space we want to make for individual linguistic agency and
7
Here I violate Jane Austen’s stricture, which she pronounces at the end of Northanger Abbey, not
to introduce a new character on the last page.
choice. The ways that linguists can find to negotiate the public and the private,
with due attention to how we may view decision-making and even the notion
of responsibility, can have resonance for researchers in cultural anthropology,
political science, economics and yet again biology, now that biologists have
begun to explore questions relating to the evolution and constitution of human
morality.
Indeed, a case in point is Mark Hauser’s Moral Minds: How Nature Designed
Our Universal Sense of Right and Wrong, which is predicated on an uncritical
transfer of Chomskyan notions of language, the language faculty, syntax and
universal grammar to the study of morality. Hauser posits that moral judgments
will be the outputs of the moral faculty whose principles are as obscure to all
of us moral (i.e., normal) people as are the principles of the language faculty
to grammaticality judgments – that is, “if the arguments in linguistics hold for
morality” (2006: 156). But what if the arguments in linguistics do not hold
even for linguistics? Hauser’s account suffers the consequences, despite the
potential wealth of information he brings to the table concerning experimental
results of morality judgments, physiological responses to moral dilemmas and
comparative ethological studies. (Yes, Hauser’s methods and results have now
been called into question, but I am at this moment more critical of his funda-
mental move, which was to uncritically base his entire thesis on Chomskyan
linguistics.) The understanding that one particular theoretical project in the
discipline, namely Chomskyan syntax, is not scientific bedrock is as long over-
due as is the need for a view of our subject matter as something other than
export-only.
Bibliography
Aarsleff, H. 1976. An Outline of Language Origins Theory Since the Renaissance. In

Origins and Evolution of Language and Speech, eds. S. R. Harnad et al. Annals of
the New York Academy of Sciences 280: 4–13.
Abutalebi, J., J.-M. Annoni, I. Zimine et al. 2008. Language Control and Lexical
Competition in Bilinguals: An Event-Related fMRI Study. Cerebral Cortex 18.7:
1496–1505.
Aitchison, J. 1996. The Seeds of Speech: Language Origin and Evolution. Cambridge
University Press.
1998. On Discontinuing the Continuity-Discontinuity Debate. In Hurford et al.,
pp. 17–43.
Akmajian, A., R. A. Demers, A. K. Farmer and R. M. Harnish 2001. Linguistics: An
Introduction to Language and Communication, 5th edn. Cambridge, MA: MIT
Press.
Alter, S. 2005. William Dwight Whitney and the Science of Language. Baltimore, MD:
The Johns Hopkins University Press.
Anderson, S. 2008. The Logical Structure of Linguistic Theory. Language 84.4:
795–813.
Andresen, J. T. 1981. Linguistic Metaphors in Charles de Brosses’s Traité of 1765 and
the History of Linguistics. Linguisticae Investigationes V: 1.1–24.
1990a. Linguistics in America: 1769–1924. A Critical History. London: Routledge.
1990b. Skinner and Chomsky Thirty Years Later. Historiographia Linguistica.17:
145–65. (Reprinted in The Behavior Analyst 1: 49–60, 1991.)
1990c. On Genetic Encoding and Communication. Reply to Newmeyer. Language
and Communication 11: 29–32.
1992a. The Behaviorist Turn in Recent Theories about Language. Behavior and
Philosophy 20: 1–19.
1992b. The Contemporary Linguist Meets the Postmodernist. Beiträge zur Geschichte
der Sprachwissenschaft 2: 213–23.
1999. Pragmatism, Behaviorism, and the Evolutionary Script. The Peirce Seminar
Papers. Essays in Semiotic Analysis, vol. IV, ed. M. Shapiro, pp. 481–514.
2010. Historiography’s Contribution to Theoretical Linguistics. In Chomskyan (R)evo-
lutions, ed. D. Kibbee, pp. 445–71. Amsterdam/Philadelphia: John Benjamins.
Appleman, P. ed. 1970. Darwin. A Norton Critical Edition. New York: Norton.
Arbib, M. and G. Rizzolatti 1997. Neural Expectations: A Possible Evolutionary Path
from Manual Skills to Language. Communication and Cognition 29: 393–424.
Armstrong, D. 2008. The Gestural Theory of Language Origins. Sign Language Studies
8.3: 289–314.
281
282 Bibliography
Attwood, A., U. Frith and B. Hermelin 1988. The Understanding and Use of
Interpersonal Gestures by Autistic and Down’s Syndrome Children. Journal of
Autism and Developmental Disorders 18.2: 241–57.
Au, T. K., L. M. Knightly, S.-A. Jun and J. S. Oh 2002. Overhearing a Language During
Childhood. American Psychological Society 13.3: 238–43.
Austin, J. L. 1962. How To Do Things With Words. Cambridge MA: Harvard University
Press.
Bakhurts, D. and S. Shanker 2001. Jerome Bruner. Language, Culture, Self. London:
Sage.
Balter, M. 2007. Brain Evolution Studies Go Micro. Science 315: 1208–11.
Barkow, J., L. Cosmides and J. Tooby, eds. 1992. The Adapted Mind: Evolutionary
Psychology and the Generation of Culture. Oxford University Press.
Barrett, H. C. 2004. Dispelling Rumors of a Gene Shortage. Science 304: 1601–602.
Barrett, L., R. Dunbar and J. Lycett 2002. Human Evolutionary Psychology. Princeton
University Press.
Bates, E., P. S. Dale and D. Thal 1995. Individual Differences and Their Implications
for Theories of Language Development. In Handbook of Child Language, eds. P.
Fletcher and B. MacWhinney, Chapter 4. Oxford: Basil Blackwell.
Beadle, G. and M. Beadle 1966. The Language of Life: An Introduction to the Science
of Genetics. Garden City, NY: Doubleday.
Becker, A. L. 1991. A Short Essay on Languaging. In Research and Reflexivity, ed. F.
Stiner, pp. 226–34. London: Sage Publications.
Berlin, B. and P. Kay 1969. Basic Color Terms: Their Universality and Evolution.
Berkeley, CA: University of California Press.
Bialystok, E. and D. Shapero 2005. Ambiguous Benefits: The Effect of Bilingualism on
Reversing Ambiguous Figures. Developmental Science 8.6: 595–604.
Bickerton, D. 1990. Language and Species. University of Chicago Press.
2009. Adam’s Tongue. How Humans Made Language. How Language Made Humans.
New York: Hill and Wang.
Bloom, P. 2004. Children Think Before They Speak. Nature 430: 410–11.
Bloom, P. and F. Keil 2001. Thinking Through Language. Mind & Language 16.4:
351–67.
Bloom, P., M. Peterson, L. Nadel and M. Garrett, eds. 1996. Language and Space.
Cambridge, MA: MIT Press.
Bloomfield, L. 1933. Language. University of Chicago Press.
Bod, R. 1998. Beyond Grammar. An Experience-Based Theory of Language. Stanford,
CA: Center for the Study of Language and Information.
Bolinger, D. 1981 [1968]. Aspects of Language, 3rd edn. New York: Harcourt Brace
Jovanovich.
Boroditsky, L. 2001. Does Language Shape Thought? Mandarin and English Speakers’
Conceptions of Time. Cognitive Psychology 4: 1–22.
Boroditsky, L., O. Fuhrman and K. McCormick 2011. Do English and Mandarin
Speakers Think About Time Differently? Cognition 118.1: 123–29.
Bowerman, M. and S. Levinson, eds. 2001. Language Acquisition and Conceptual
Development. Cambridge University Press.
Brace, C. L. 2004. Bipedalism, Canine Tooth Reduction, and Obligatory Tool Use.
Response to Falk (2004), Behavioral and Brain Sciences 24.4: 507–08.
Bibliography 283
Brosnan, S. and F. de Waal 2003. Monkeys Reject Unequal Pay. Nature: 297–99.
Brown, P. and S. Levinson 1987 [1978]. Politeness: Some Universals in Language
Usage. Cambridge University Press.
Brown, R. 1973. A First Language: The Early Stages. Cambridge MA: Harvard
University Press.
Brown, R. L. 1967. Wilhelm von Humboldt’s Conception of Linguistic Relativity. The
Hague: Mouton.
Bruner, J. 1983. Child’s Talk: Learning to Use Language. New York: Norton.
1990. Acts of Meaning. Cambridge, MA: Harvard University Press.
Buller, D. and V. Gray Hardcastle 2000. Evolutionary Psychology, Meet Developmental
Neurobiology: Against Promiscuous Modularity. Brain and Mind 1: 307–25.
Burling, R. 1992. Patterns of Language: Structure, Variation, Change. San Diego:
Academic Press.
Buttelmann, D., M. Carpenter, J. Call and M. Tomasello 2007. Enculturated Apes
Imitate Rationally. Developmental Science 10.4: 31–38.
Calvin, W 1991. The Throwing Madonna: Essays on the Brain. New York: McGraw-
Hill.
Campbell, R. S. and J. Pennebaker 2003. The Secret Life of Pronouns: Flexibility in
Writing Style and Physical Health. Psychological Science 14.1: 60–64
Cantlon, J. and E. Brannon 2005. Semantic Congruity Affects Numerical Judgments
Similarly in Monkeys and Humans. Proceedings of the National Academy of
Sciences 102.45: 16507–11.
2006. Shared System for Ordering Small and Large Numbers in Monkeys and
Humans. Psychological Science 17.5: 402–407.
Cavalli-Sforza, L. 2000. Genes, Peoples, and Languages. New York: North Point
Press.
Chambers, J. K. 2003 [1995]. Sociolinguistic Theory, 2nd edn. Oxford: Blackwell.
Changeux, J.-P. 1985. Neuronal Man: The Biology of Mind, trans. L. Garey. New York:
Pantheon Books.
Changeux, J.-P. and P. Ricoeur 2000. What Makes Us Think? A Neuroscientist and
a Philosopher Argue about Ethics, Human Nature, and the Brain, trans. M. B.
DeBevoise. Princeton University Press.
Changizi, M. 2001. Universal Scaling Laws for Hierarchical Complexity in Languages,
Organisms, Behaviors and Other Combinatorial Systems. Journal of Theoretical
Biology 211: 277–95.
Chen, J.-Y. 2007. Do Chinese and English Speakers Think About Time Differently?
Failure of Replicating Boroditsky (2001). Cognition 104: 427–36.
Cheney, D. and R. Seyfarth 1990. How Monkeys See the World: Inside the Mind of
Another Species. University of Chicago Press.
Choi, S., L. McDonough, M. Bowerman and J. M. Mandler 1999. Early Sensitivity
to Language-Specific Spatial Categories in English and Korean. Cognitive
Development 14: 241–68.
Chomsky, N. 1957. Syntactic Structures. Berlin: Mouton.
1959. Review of Verbal Behavior. Language 35.1: 26–58.
1965. Aspects of the Theory of Syntax. Cambridge, MA: MIT Press.
1966. Cartesian Linguistics. A Chapter in the History of Rationalist Thought. New
York: Harper & Row.
284 Bibliography
1968. Language and Mind. New York: Harcourt, Brace & World.
1971. The Case Against B.F. Skinner, Review of Beyond Freedom and Dignity by B.
F. Skinner. New York Review of Books, December 30.
1972. Language and Mind (enl. edn.) New York: Harcourt Brace Jovanovich.
1975. Reflections on Language. New York: Pantheon.
1980. Rules and Representations. Oxford University Press.
1988. Language and Problems of Knowledge. Cambridge, MA: MIT Press.
1995a. The Minimalist Program. Cambridge, MA: MIT Press.
1995b. Language and Nature. Mind 104: 1–61.
2002. Nature and Language. Cambridge University Press.
2006. Language and Mind, 3rd edn. Cambridge University Press.
Chomsky, N. and M. Halle 1968. The Sound Pattern of English. New York: Harper &
Row.
Chomsky, N., M. D. Hauser and W. T. Fitch 2005. Appendix. The Minimalist Program.
A response to Pinker and Jackendoff (2005). Retrieved online.
Christiansen, M. and S. Kirby, eds. 2003. Language Evolution: States of the Art. New
York: Oxford University Press.
Chung, C. and J. Pennebaker 2007. The Psychological Functions of Function Words. In
Social Communication, ed. K. Fielder, pp. 343–59. New York: Psychology Press.
Clark, A. 1997. Being There. Putting Brain, Body, and World Together Again. Cambridge,
MA: MIT Press.
Cohen, J. 2007. Relative Differences: The Myth of 1%. Science 316: 1836.
Cohen, R. and D. Borsoi 1996. The Role of Gesture in Description-Communication:
A Cross-Sectional Study of Aging. Journal of Nonverbal Behavior 20.1:
45–63.
Colopento, J. 2007. The Interpreter. Has a Remote Amazonian Tribe Upended Our
Understanding of Language? New Yorker April 16, pp. 118–37.
Condon, W. and L. Sander. 1974. Neonate Movement Is Synchronized with Adult
Speech: Interactional Participation and Language Acquisition. Science 183:
99–101.
Connolly, W. 1995. The Ethos of Pluralization. Minneapolis: University of Minnesota
Press.
1999. Why I Am Not A Secularist. Minneapolis: University of Minnesota Press.
2002. Neuropolitics. Thinking, Culture, Speed. Minneapolis: University of Minnesota
Press.
2005. Pluralism. Durham, NC: Duke University Press.
Cooperrider, K. 2003. Observing Language, Observing Linguistics: Toward a Biological
Systems Epistemology of Languaging. Senior Honors Thesis, Duke University.
Corballis, M. 2002. From Hand to Mouth: The Origins of Language. Princeton, NJ:
Princeton University Press.
Cowie, F. 1999. What’s Within? Nativism Reconsidered. New York: Oxford University
Press.
Crick, F. 1957. On Protein Synthesis. Symposium of the Society of Experimental Biology
12: 138–63.
1970. Central Dogma of Molecular Biology. Nature 227: 561–63.
1989. The Recent Excitement About Neural Networks. Nature 337: 129–31.
Critchley, M. 1959. The Evolution of Man’s Capacity for Language. Evolution after
Darwin 2: 289. University of Chicago Press.
Bibliography 285
Crow, T. 1997. Is Schizophrenia the Price that Homo Sapiens Pays for Language?
Schizophrenia Research 28: 127–41.
2005. Who Forgot Paul Broca? The Origin of Language as Test Case of Speciation
Theory. Journal of Linguistics 41: 133–56.
2008. The “Big Bang” Theory of the Origin of Psychosis and the Faculty of Language.
Schizophrenia Research 102.1–3: 31–52.
Dąbrowska, E. 2004. Language, Mind and Brain: Some Psychological and Neurological
Constraints on Theories of Grammar. Edinburgh University Press.
Damasio, A. 1994. Descartes’ Error: Emotion, Reason, and the Human Brain. New
York: Putnam.
2003. Looking for Spinoza: Joy, Sorrow, and the Feeling Brain. Orlando: Harcourt.
Danziger, S. and R. Ward 2010. Language Changes Implicit Associations Between Ethnic
Groups and Evaluation in Bilinguals. Psychological Science 21.6: 799–800.
Darwin, C. 1968 [1859]. Origin of the Species by Means of Natural Selection or The
Preservation of Favoured Races in the Struggle for Life, edited and with an intro-
duction by J.W. Burrow. London: Penguin Books.
1979, 1970 [1871]. Descent of Man. New York: Norton.
Davies, I. and G. Corbett 1997. A Cross-Cultural Study of Colour Grouping: Evidence
for Weak Linguistic Relativity. British Journal of Psychology 88: 493–517.
Deacon, T. 1997. The Symbolic Species. The Co-Evolution of Language and the Brain.
New York: W. W. Norton.
de Laguna, G. 1927. Speech: Its Function and Development. New Haven: Yale
University Press.
de Waal, F. 1982. Chimpanzee Politics: Power and Sex Among Apes, with forward by
Desmond Morris. Baltimore, MD: The Johns Hopkins University Press.
2001. The Ape and the Sushi Master: Cultural Reflections of a Primatologist. New
York: Basic Books.
Dediu, D. and D. R. Ladd 2007. Linguistic Tone Is Related to the Population Frequency
of the Adaptive Haplogroups of Two Brain Size Genes, ASPM and Microcephalin.
Proceedings of the National Academy of Sciences 104: 10944–49.
Deleuze, G. and F. Guattari 1987. A Thousand Plateaus. Capitalism and Schizophrenia,
translation and Foreword by B. Massumi. Minneapolis: University of Minnesota
Press.
Derrida, J. 1967. De la grammatologie. Paris: Editions de Minuit.
1988. Limited, Inc. Evanston, IL: Northwestern University Press.
Descartes, R. 1996. Discourse on the Method [Discours de la Méthode 1637] and
Meditations on First Philosophy [Meditationes de prima philosophia 1641], ed. D.
Weissman. New Haven: Yale University Press.
Deutsch, D. and T. Henthorn 2004. Absolute Pitch, Speech, and Tone Language: Some
Experiments and a Proposed Framework. Music Perception 21.3: 339–56.
Deutscher, G. 2005. The Unfolding of Language. An Evolutionary Tour of Mankind’s
Greatest Invention. New York: Henry Holt.
Dick, A. S., S. Goldin-Meadow, U. Hasson, J. Skipper and S. Small 2009. Co-Speech
Gestures Influence Neural Activity in Brain Regions Associated with Processing
Semantic Information. Human Brain Mapping 30: 3509–26.
Dickerson, P., P. Stribling and J. Rae 2007. Tapping into Interaction: How Children with
Autism Spectrum Disorders Design and Place Tapping in Relation to Activities in
Progress. Gesture 7.3: 271–303.
286 Bibliography
Dolan, R. L. 2002. Emotion, Cognition, Behavior. Science 2989: 1191–94.

Donald, M. 1991. Origins of the Modern Mind: Three Stages in the Evolution of Culture
and Cognition. Cambridge, MA: Harvard University Press.
2001. A Mind So Rare: The Evolution of Human Consciousness. New York: W.W.
Norton.
Doolittle, W. F. and E. Bapteste 2007. Pattern Pluralism and the Tree of Life Hypothesis.
Proceedings of the National Academy of Sciences 104: 2043–49.
Dubner, S. and S. Levitt 2005. Keith Chen’s Monkey Research. New York Times
Magazine, June 5.
Elman, J. 1990. Finding Structure in Time. Cognitive Science 14.179–211.
2003. Generalization from Sparse Input. Proceedings of the 38th Annual Meeting of
the Chicago Linguistic Society.
Elman, J., E. A. Bates, M. H. Johnson et al. 1999. Rethinking Innateness: A Connectionist
Perspective on Development. Cambridge, MA: MIT Press.
Erard, M. 2007. Read My Slips: Speech Errors Show How Language is Processed.
Science 317: 1674–76.
Evans, N. and S. Levinson 2009. The Myth of Language Universals: Language Diversity
and Its Importance for Cognitive Science. Behavioral and Brain Sciences 32:
429–92.
Everett, D. 2009. Pirahã Culture and Grammar: A Response to Some Criticism.
Language 85.2: 405–42.
Falk, D. 2004. Prelinguistic Evolution in Early Hominins: Whence Motherese?
Behavioral and Brain Sciences 27.4: 491–503.
Fedorenko, E., M. Behr and N. Kanwisher 2011. Localizing Language in the Brain.
Proceedings of the National Academy of Sciences August 30, 2011.
Fillmore, C. and P. Kay 1999. Grammatical Constructions and Linguistic Generalizations:
The What’s X Doing Y? Construction. Language 75: 1–33.
Fillmore, C., P. Kay and M. C. O’Connor 1988. Regularity and Idiomaticity in
Grammatical Constructions: The Case of LET ALONE. Language 64.3: 501–38.
Fisher, S. 2006. Tangled Webs: Tracing the Connections Between Genes and Cognition.
Cognition 101: 270–97.
Fishman, J. 1960. Systematization of the Whorfian Hypothesis. Behavioral Science 5:
323–39.
Fitch, W. T. 2010. The Evolution of Language. Cambridge University Press.
Fleck, L. 1979 [1935]. Genesis and Development of a Scientific Fact, ed. Th. J. Trenn and
R. K. Merton, trans. F. Bradley and Th. J. Trenn. University of Chicago Press.
Fodor, J. 1983. The Modularity of Mind. Cambridge, MA: MIT Press.
1998. Concepts: Where Cognitive Science Went Wrong. Oxford University Press.
2000. The Mind Doesn’t Work That Way: The Scope and Limits of Computational
Psychology. Cambridge, MA: MIT Press.
Foley, W. 1997. Anthropological Linguistics: An Introduction. Malden, MA:
Blackwell.
Foucault, M. 1971. The Discourse on Language. Social Science Information 10.
1994. Ethics: Subjectivity and Truth. Essential Works of Foucault 1954–1984, vol. I,
ed. P. Rabinow, trans. R. Hurley and others. New York: New Press.
Freeman, S. and J. C. Herron 2004. Evolutionary Analysis, 3rd edn. Upper Saddle River,
NJ: Pearson/Prentice Hall.
Friederici, A. 2006. What’s in Control of Language? Nature Neuroscience 9.8: 991–92.
Bibliography 287
Fromkin, V., R. Rodman and N. Hyams 2003. An Introduction to Language, 7th edn.
Boston: Thomson/Wadsworth.
2011. An Introducction to Language, 9th edn. Boston: Thomson/Wadsworth.
Futuyma, D. 1986. Evolutionary Biology, 2nd edn. Sunderland, MA: Sinauer
Associates.
1998. Evolutionary Biology, 3rd edn. Sunderland, MA: Sinauer Associates.
Gabrieli, J. 2009. Dyslexia: A New Synergy Between Energy and Cognitive Science.
Science 325: 280–83.
Gamble, C. 1994. Timewalkers. The Prehistory of Global Colonization. Cambridge,
MA: Harvard University Press.
Gazzaniga, M., R. Ivry and G. Mangum 2002. Cognitive Neuroscience. The Biology of
Mind, 2nd ed. New York: W.W. Norton.
Geertz, C. 1962. The Growth of Culture and the Evolution of Mind. In Theories of the
Mind, ed. J. M. Scher, pp. 713–37. New York: The Free Press of Glencoe.
Gellatly, A. 1995. Colourful Whorfian Ideas: Linguistic and Cultural Influences on the
Perception and Cognition of Colour, and on the Investigation of Them. Mind &
Language 10.3: 199–225.
Gentner, D. and S. Goldin-Meadow 2003. Language in Mind. Advances in the Study of
Language and Thought. Cambridge, MA: MIT Press.
Gibbons, A. 2007. European Skin Turned Pale Only Recently, Gene Suggests. Science
316: 364.
Gibson, J. J. 1979. The Ecological Approach to Visual Perception. Boston: Houghton
Mifflin.
1982. Reason for Realism. Selected Essays of James J. Gibson, ed. E. Reed and R.
Jones. Hillsdale, NJ: Lawrence Erlbaum.
Gibson, K. 1994. Continuity Theories of Human Language Origins Versus the Liberman
Model. Language and Communication 14.1: 97–114.
Givón, T. 1998. On the Co-Evolution of Language, Mind and Brain. Evolution of
Communication 2: 45–116.
2002. The Visual Information-Processing System as an Evolutionary Precursor of
Human Language. In Givón and Malle, pp. 3–50.
Givón, T. and B. F. Malle, eds. 2002. The Evolution of Language out of Pre-Language.
Amsterdam: John Benjamins.
Goldberg, A. 2001. Patient Arguments of Causative Verbs Can Be Omitted: The
Role of Information Structure in Argument Distribution. Language Sciences
23: 503–24.
2006. Constructions at Work: The Nature of Generalization in Language. Oxford
University Press.
Goldberg, A. and R. Jackendoff 2004. The English Resultative as a Family of
Constructions. Language 80.3: 532–68.
Goldin-Meadow, S. 2003. Hearing Gesture: How Our Hands Help Us Think. Cambridge,
Goldstein, M. and M. West 1999. Consistent Responses of Human Mothers to
Prelinguistic Infants: The Effect of Prelinguistic Repertoire Size. Journal of
Comparative Psychology 113.1: 52–58.
Goldstein, M., A. King and M. West 2003. Social Interaction Shapes Babbling: Testing
Parallels Between Birdsong and Speech. Proceedings of the National Academy of
Sciences 100.13: 8030–35.
288 Bibliography
Gopnik, A. 2001. Theories, Language, and Culture: Whorf Without Wincing. In

Bowerman and Levinson, pp. 45–69.
González, J., A. Barros-Loscertales, F. Pulvermüller et al. 2006. Reading Cinnamon
Activates Olfactory Brain Regions. NeuroImage 32: 906–12.
Gordon, P. 2004. Numerical Cognition Without Words: Evidence from Amazonia.
Science 306: 496–99.
Green, J. 1987. Romance Languages. In The World’s Major Languages, ed. B. Comrie,
pp. 203–209.
Griffiths, P. 1997. What Emotions Really Are. University of Chicago Press.
2007. Evo-Devo Meets the Mind: Towards a Developmental Evolutionary Psychology.
In Integrating Evolution and Development, eds. R. Sansom and R. Brandon.
Hagoort, P. 2005. On Broca, Brains, and Binding: A New Framework. Trends in
Cognitive Science 9.9: 416–23.
Hakuta, K., E. Bialystok and E. Wiley 2003. Critical Evidence: A Test of the Critical-
Period Hypothesis for Second-Language Acquisition. Psychological Science 14.1:
31–38.
Harris, R. 1996. The Language Connection: Philosophy and Linguistics. Bristol, UK:
Thoemmes Press.
Hauser, M. 2006. Moral Minds: How Nature Designed Our Universal Sense of Right
and Wrong. New York: Harper Collins.
Hauser, M., N. Chomsky and W. T. Fitch 2002. The Faculty of Language: What Is It,
Who Has It, and How Did It Evolve? Science 298: 1569–79.
Hayek, F. A. 1988. The Fatal Conceit. The Errors of Socialism. Chicago University
Press.
Hendriks-Jansen, H. 1996. Catching Ourselves in the Act. Cambridge, MA: MIT
Press.
Hespos, S. and E. Spelke 2004. Conceptual Precursors to Languages. Nature 430:
453–56.
Hill, J. 1974. Possible Continuity Theories of Language. Language 50.1: 134–50.
Hockett, C. 1960. Origin of Speech. Scientific American 203: 89–97.
Holle, H., J. Obleser, S.-A. Rueschemeyer and T. Gunter 2010. Integration of
Iconic Gestures and Speech in Left Superior Temporal Areas Boosts Speech
Comprehension Under Adverse Listening Conditions. NeuroImage 49.1: 875–84.
Humboldt, W. von 1997. Essays on Language, ed. T. Harden and D. Farrelly, trans. J.
Wieczorek and I. Roe. Frankfurt am Main: Peter Lang.
Hunt, E. and F. Agnoli 1991. The Whorfian Hypothesis: A Cognitive Psychology
Perspective. Psychological Review 98.3: 377–89.
Hurford, J., M. Studdert-Kennedy and C. Knight, eds. 1998. Approaches to the Evolution
of Language. Cambridge University Press.
Huxley, Sir J. 1942 [1963]. Evolution: The Modern Synthesis. In Appleman, pp. 244–65.
Iverson, J. and S. Goldin-Meadow 2001. The Resilience of Gesture in Talk: Gesture in
Blind Speakers and Listeners. Developmental Science 4.4: 416–22.
Jablonka, E. and M. Lamb 2005. Evolution in Four Dimensions: Genetic, Epigenetic,
Behavioral, and Symbolic Variation in the History of Life. Cambridge, MA: MIT
Press.
Jackendoff, R. 1999. Possible Stages in the Evolution of the Language Capacity. Trends
in Cognitive Sciences 3.7: 272–79.
Bibliography 289
2001. Language in the Ecology of the Mind. In The Routledge Companion to

Semiotics and Linguistics, ed. P. Cobley, pp. 52–65. London: Routledge.
2002. Foundations of Language: Brain, Meaning, Grammar, Evolution. Oxford:
Oxford University Press.
Jackendoff, R. and S. Pinker 2005. The Nature of the Language Faculty and Its
Implications for Evolution of Language (Reply to Fitch, Hauser and Chomsky).
Jacob, F. 1966. Genetics of the Bacterial Cell. Science 152: 1470–748.
Jakobson, R. 1990 [1976]. The Speech Event and the Functions of Language. In On
Language, ed. L. Waugh and M. Monville-Burston, pp 69–79. Cambridge, MA:
Harvard University Press.
James, W. 1981 [1890]. The Principles of Psychology. Cambridge MA: Harvard
University Press.
1981 [1907]. Pragmatism. Indianapolis: Hackett Publishing.
Jänke, L., Th. Siegenthaler, S. Preis and H. Steinmetz 2007. Decreased White-Matter
Density in a Left-Sided Fronto-Temporal Network in Children with Development
Language Disorder: Evidence for Anatomical Anomalies in a Motor-Language
Network. Brain and Language 102: 91–98.
Jaworski, A. and N. Coupland 1999. The Discourse Reader. London: Routledge.
Jerne, N. 1985. The Generative Grammar of the Immune System. Science 229:
1057–60.
Jespersen, O. 1922. Language: Its Nature, Development and Origin. London: George
Allen & Unwin.
Johnstone, B. 1996. The Linguistic Individual: Self-Expression in Language and
Linguistics. New York: Oxford University. Press.
2000. The Individual Voice in Language. Annual Review of Anthropology 29:
405–24.
2002. Discourse Analysis. Oxford: Blackwell.
Joseph, J. 2002. From Whitney to Chomsky. Essays in the History of American
Linguistics. Amsterdam/Philadelphia: John Benjamins.
Kandel, E., J. Schwartz and T. Jessell 1995. Essentials of Neural Science and Behavior.
Norwalk, CT: Appleton & Lange.
Karmiloff-Smith, A. 1995. Beyond Modularity: A Developmental Perspective on
Cognitive Science. Cambridge, MA: MIT Press.
1998. Development Itself Is the Key to Understanding Developmental Disorders.
Trends in Cognitive Science 2.10: 389–398.
Karmiloff, K. and A. Karmiloff-Smith 2001. Pathways to Language: From Fetus to
Adolescent. Cambridge, MA: Harvard University Press.
Kay, P. and C. Fillmore 1999. Grammatical Constructions and Linguistic Generalizations:
The What’s X doing Y? Construction. Language 75.1: 1–33.
Kay, P. and W. Kempton 1984. What Is the Sapir-Whorf Hypothesis? American
Anthropologist 86: 65–79.
Kay, P. and L. Maffi 2000. Color Appearance and the Emergence and Evolution of
Basic Color Lexicons. American Anthropologist 101.4: 743–60.
Keller, E. F. and E. A. Lloyd, eds. 1992. Keywords in Evolutionary Biology. Cambridge,
Kendon, A. 1991. Some Consideration for a Theory of Language Origins. Man (N.S.)
26: 199–221.
290 Bibliography
Kohler, E., C. Keysers, M. A. Umilta et al. 2002. Hearing Sounds, Understanding
Actions: Action Representation in Mirror Neurons. Science 297: 846–48.
Krauss, R. 1987. The Role of the Listener: Addressee Influences on Message
Formulation. Journal of Language and Social Psychology 6.2: 81–98.
Kuhn, T. 1962. The Structure of Scientific Revolutions. Chicago, IL: University of
Chicago Press.
Labov, W. 1963. The Social Motivation of Sound Change. Word 19: 273–309.
1972. Sociolinguistic Patterns. Philadelphia: University of Pennsylvania Press.
1987. The Overestimation of Functionalism. In Functionalism in Linguistics, ed. R.
Dirven and V. Fried, pp. 311–32. Amsterdam, Philadelphia: John Benjamins.
Lai, C., S. Fisher, J. Hurst, F. Vargha-Khadem and A. Monaco 2001. A Forkhead-
Domain Gene Is Mutated in a Severe Speech and Language Disorder. Science
413: 519–23.
Lakoff, G. 1987. Women, Fire and Dangerous Things: What Categories Reveal About
the Mind. University of Chicago Press.
2004. Don’t Think of an Elephant! White River Junction, VT: Chelsea Green
Publishers.
Lakoff, G. and M. Johnson 1980. Metaphors We Live By. University of Chicago
Press.
1999. Philosophy in the Flesh. New York: Basic Books.
Lakoff, G. and R. Núñez 2000. Where Mathematics Comes From: How the Embodied
Mind Brings Mathematics Into Being. New York: Basic Books.
Langacker, R. 1987. Foundations of Cognitive Grammar. Palo Alto: Stanford University
Press.
LeDoux, J. 1996. The Emotional Brain: The Mysterious Underpinning of Emotional
Life. New York: Simon & Schuster.
Lenneberg, E. 1967. Biological Foundations of Language. New York: Wiley.
Levinson, S. C. 2001. Covariation Between Spatial Language and Cognition, and Its
Implications for Language Learning. In Bowerman and Levinson, pp. 566–88.
2003. Space in Language and Cognition: Explorations in Cognitive Diversity.
Cambridge: Cambridge University Press.
Levinson, S. C., S. Kita, D. B. M. Haun and B. J. Rasch 2002. Returning the Tables:
Language Affects Spatial Reasoning. Cognition 84.2: 155–88.
Lewontin, R. 2000. The Triple Helix: Gene, Organism, and Environment. Cambridge,
Li, C. 2002. Missing Links, Issues and Hypotheses in the Evolutionary Origin of
Language. In Givón and Malle, pp. 83–106.
Li, P. and L. Gleitman 2002. Turning the Tables: Language and Spatial Reasoning.
Cognition 83.3: 265–94.
Lieberman, P. 1984. The Biology and Evolution of Language. Cambridge, MA: Harvard
University Press.
1991. Uniquely Human: The Evolution of Speech, Thought, and Selfless Behavior.
Cambridge, MA: Harvard University Press.
2000. Human Language and Our Reptilian Brain: The Subcortical Bases of Speech,
Syntax, and Thought. Cambridge, MA: Harvard University Press.
2005. The Pied Piper of Cambridge. The Linguistic Review 22: 223–35.
2006. Toward an Evolutionary Biology of Language. Cambridge, MA: Harvard
University Press.
Bibliography 291
Locke, J. 1975 [1690]. An Essay Concerning Human Understanding, ed. P. Nidditch.

Oxford: Clarendon Press.
Luhmann, N. 1990. Essays on Self-Reference. New York: Columbia University Press.
MacFarquhar, L. 2003. The Devil’s Accountant. New Yorker March 31, pp. 64–79.
MacNeilage, P. 2008. The Origin of Speech. Oxford: Oxford University Press.
Maestripieri, D., M. Leoni, S. Raza, E. Hirsch and J. Whitham 2005. Female Copulation
Calls in Guinea Baboons: Evidence for Postcopulatory Female Choice?
International Journal of Primatology 26.4: 737–58.
Marr, D. 1982. Vision: A Computational Investigation into the Human Representation
and Processing of Visual Information. New York: Freeman.
Martin, L. 1986. Eskimo Words for Snow: A Case Study in the Genesis and Decay of an
Anthropological Example. American Anthropologist 88.2: 418–23.
Maturana, H. 1970. The Biology of Cognition. In Maturana and Varela 1980 [1972],
pp. 1–58.
1987. Everything Is Said by an Observer. In Gaia: A Way of Knowing. Political
Implications of the New Biology, ed. W. I. Thompson, pp. 11–36. New York:
Lindisfarne Press.
1989. Lenguaje y realidad: El origin de lo humano. Arch. Biol. Med. Exp. 22: 77–81.
Chile.
1991 [1988]. Reality: The Search for Objectivity, or the Quest for a Compelling
Argument. In Die Gedankenwelt Sir Karl Poppers. Kritischer Rationalismus im
Dialog, ed. N. Leser, J. Seifer and K. Plitzner, pp. 282–357. Heidelberg: Car
Winter.
Maturana, H. and J. Mpodozis 1992. Origen de las Espécies por medio de la Deriva
natural o La diversificación de los linajes a través de la conservación y cambio
de los fenotipos ontogénicos. Publicación Ocasional n. 46/1992. Santiago, Chile:
Museo Nacional de Historia Natural.
Maturana, H. and F. Varela 1980 [1972]. Autopoiesis and Cognition. The Realization of
the Living. Dordrecht: D. Reidel.
Maturana, H. 1992 [1987]. The Tree of Knowledge, The Biological Roots of Human
Understanding, revised edn., trans. R. Paolucci. Boston: Shambhala.
Mayr, E. 1963. Animal Species and Evolution. Cambridge, MA: Harvard University
Press.
1982. The Growth of Biological Thought: Diversity, Evolution, and Inheritance.
McGurk, H. and J. MacDonald 1976. Hearing Lips and Seeing Voices. Nature 264:
746–48.
McIntosh, A. R. and N. J. Lobaugh 2003. When Is a Word Not a Word? Science 301:
322–23.
McMurray, B. 2007. Defusing the Childhood Vocabulary Explosion. Science 317: 631.
McNeill, D. 2005. Gesture and Thought. Chicago, IL: University of Chicago Press.
2012. How Language Began: Gesture and Speech in Human Evolution. Cambridge
University Press.
Melser, D. 2004. The Act of Thinking. Cambridge, MA: MIT Press.
Meltzoff, A., P. Kuhl, J. Movellan and T. Sejnowski 2009. Foundations for a New
Science of Learning. Science 325: 284–88.
Mendoza-Denton, N. 2007. Homegirls: The Making of Latina Youth Styles. Oxford:
Blackwell.
292 Bibliography
Mingers, J. 2002. Can Social Systems Be Autopoietic? Assessing Luhmann’s Social

Theory. The Sociological Review 50.2: 278–99.
Morsella, E. and R. Krauss 2004. The Role of Gestures in Spatial Working Memory and
Speech. The American Journal of Psychology 117.3: 411–24.
Mufwene, S. 2001. The Ecology of Language Evolution. Cambridge: Cambridge
University Press.
Müller, (F.) M. 1965 [1861]. Lectures on the Science of Language, delivered at the
Royal Institution of Great Britain. Delhi, India: Munshi Ram Manohar Lal.
Neisser, U. 1976. Cognition and Reality: Principles and Implications of Cognitive
Psychology. San Francisco, CA: W. H. Freeman.
Nelson, K. 1996. Language in Cognitive Development: Emergence of the Mediated
Mind. Cambridge University Press.
2007. Young Minds in Social Worlds. Cambridge, MA: Harvard University Press.
Nevins, A., D. Pesetsky and C. Rodrigues 2009. Pirahã Exceptionality: A Reassessment.
Language 85.2: 355–404.
Newman, J. 2004. Motherese by Any Other Name: Mother-Infant Communication in
Non-Hominin Mammals. Response to Falk (2004), Behavioral and Brain Sciences
27.4: 519–20.
Newmeyer, F. 1998. On the Supposed “Counterfunctionality” of Universal Grammar:
Some Evolutionary Implications. In Hurford et al., pp. 305–19.
2003. Grammar Is Grammar and Usage Is Usage. Language 79.4: 682–707.
Nunberg, G. 1997. Verbed Off. In The Way We Talk Now (2001), pp. 151–53. Boston,
MA: Houghton Mifflin.
Núñez, R. and W. J. Freeman 1999. Reclaiming Cognition: The Primacy of Action,
Intention and Emotion. Thorverton, UK and Bowling Green, OH: Imprint
Academic.
Núñez, R. and E. Sweetser 2006. With the Future Behind Them: Convergent Evidence
From Aymara Language and Gesture in the Crosslinguistic Comparison of Spatial
Construals of Time. Cognitive Science 30: 401–50.
O’Grady, W, J. Archibald, M. Aronoff and J. Rees-Miller 2001. Contemporary
Linguistics: An Introduction. Boston: Bedford/St. Martin’s.
O’Higgins, P. and S. Elton 2007. Walking on Trees. Science 316: 1292–94.
Oozera, C., C. Noongwook, C. Alowa and I. Krupnik 2004. Watching Ice and Weather
Our Way. Washington DC: Arctic Studies Center, Smithsonian.
Oshima-Takane, Y., E. Goodz and J. L. Derevensky 1996. Birth Order Effects on Early
Language Development: Do Secondborn Children Learn from Overheard Speech?
Child Development 67: 621–34.
Oyama, S. 1985. The Ontogeny of Information: Developmental Systems and Evolution.
Cambridge: Cambridge University Press.
1988. Stasis, Development and Heredity. In Evolutionary Processes and Metaphors,
ed. M.-W. Ho and S.W. Fox, pp. 255–74. London: Wiley.
1989. Ontogeny and the Central Dogma: Do We Need the Concept of Genetic
Programming in Order to Have an Evolutionary Perspective? In Systems and
Development. The Minnesota Symposia on Child Psychology, vol. XXII, ed. M.
Gunnar and E. Thelen, pp. 1–13. Hillsdale, NJ: Lawrence Erlbaum.
1995. The Accidental Chordate: Contingency in Developmental Systems. South
Atlantic Quarterly 94.2: 509–26.
Bibliography 293
2000. Evolution’s Eye. A Systems View of the Biology-Culture Divide. Durham, NC:
Duke University Press.
Özyürek, A., S. Allen, R. Furman et al. 2008. Development of Cross-Linguistic Variation
in Speech and Gesture: Motion Events in English and Turkish. Developmental
Psychology 44.4: 1040–54.
Packman, A., C. Code and M. Onslow 2007. On the Cause of Stuttering: Integrating
Theory with Brain and Behavioral Research. Journal of Neurolinguistics 20:
353–62.
Pennebaker, J. 1997. Opening Up: The Healing Power of Expressing Emotions. New
York: Guildford Press.
Peters, A. 1983. The Units of Language Acquisition. Cambridge University Press.
Peters, J. D. 1989. John Locke, The Individual, and the Origin of Communication.
Quarterly Journal of Speech 75.4: 387–99.
Pica, P., C. Lerner, V. Izard and S. Dehaene 2004. Exact and Approximate Arithmetic in
an Amazonian Indigene Group. Science 306: 499–503.
Pinker, S. 1994. The Language Instinct: How the Mind Creates Language. New York:
William Morrow.
1997. How the Mind Works. New York: Norton.
2002. Why the Nature/Nurture Debate Won’t Go Away. Boston Globe. October 13.
2003. Language as an Adaptation to the Cognitive Niche. In Language Evolution:
States of the Art, ed. M. H. Christiansen and S. Kirby. New York: Oxford University
Press.
2007. The Stuff of Thought. Language as a Window into Human Nature. Viking/
Penguin: New York.
2008. The Moral Instinct. New York Times Sunday Magazine, January 13: 32ff.
2010. The Cognitive Niche: Coevolution of Intelligence, Sociality, and Language.
Proceedings of the National Academy of Sciences 107 (Supplement 2):
8993–99.
Pinker, S. and P. Bloom 1990. Natural Language and Natural Selection. Behavioral and
Brain Sciences 13.4: 707–27, with Open Peer Commentary 727–84.
1992. Natural Language and Natural Selection, reprinted in Barkow et al.,
pp. 451–93.
Pinker, S. and R. Jackendoff 2005. The Faculty of Language: What’s Special About It?
Pollick, A. and F. de Waal 2007. Ape Gestures and Language Evolution. Proceedings of
the National Academy of Sciences 104.19: 8184–89.
Poplack, S. 1979/80. Sometimes I’ll Start a Sentence in Spanish y Termino En Espanol.
Linguistics 18: 581–618.
Port, R. 2010. Rich Memory and Distributed Phonology. Language Sciences 32:
43–55.
Premack, D. 2004. Is Language the Key to Human Intelligence? Science 303: 318–20.
Pullum, G. 1996. Learnability, Hyperlearning, and the Poverty of the Stimulus.
Proceedings of the Berkeley Linguistics Society 22: 498–513.
Pullum, G. and B. Scholz 2002. Empirical Assessment of Stimulus Poverty Arguments.
Linguistic Review 19: 9–50.
Purves, D. and B. Lotto 2003. Why We See What We Do: An Empirical Theory of Vision.
Sunderland, MA: Sinauer Associates.
294 Bibliography
Putnam, H. 1967. The ‘Innateness Hypothesis’ and Explanatory Models in Linguistics.

Synthese 17.1: 12–22.
Quine, W. V. 1940. Mathematical Logic. Cambridge, MA: Harvard University Press.
Reddy, M. 1979. The Conduit Metaphor: A Case of Frame Conflict in Our Language
About Language. In Metaphor and Thought, ed. A. Ortony, pp. 284–324.
Cambridge University Press.
Rickford, J. R., T. Wasow, N. Mendoza-Denton and J. Espinoza 1995. Syntactic
Variation and Change in Progress: Loss of the Verbal Coda in Topic-Restricting As
Far As Constructions. Language 71.1: 102–31.
Riegler, A. 2005. Constructive Memory. Kybernetes 34.1/2: 89–104.
Rizzolatti, G. and M. Arbib 1998. Language Within Our Grasp. Trends in Neurosciences
12: 188–94.
Rizzolatti, G. and L. Craighero 2004. The Mirror-Neuron System. Annual Review of
Neuroscience 27: 169–92.
Rockwell, T. 2005. Neither Brain Nor Ghost. Cambridge, MA: MIT Press.
Rowe, M. and S. Goldin-Meadow 2009. Differences in Early Gesture Explain SES
Disparities in Child Vocabulary Size at School Entry. Science 323: 951–53.
Sampson, G., D. Gil and P. Trudgill, eds. 2009. Language Complexity as an Evolving
Variable. Oxford University Press.
Sapir, E. 1921. Language. New York: Harcourt, Brace & World.
1949 [1929]. The Status of Linguistics as a Science. Language 5: 207–14. Reprinted
in Selected Writings of Edward Sapir in Language, Culture, and Personality
(1949), ed. D. G. Mandelbaum, pp. 160–66. Berkeley, CA: University of California
Press.
Saussure, F. de 1985 [1916]. Cours de linguistique générale, edition prepared by T. de
Mauro. Paris: Payot.
1959 [1916]. Course in General Linguistics, translated and with an introduction by
W. Baskin. New York: McGraw-Hill.
Savage-Rumbaugh, S. 1987. Communication, Symbolic Communication, and
Language: Reply to Seidenberg and Petitto. Journal of Experimental Psychology:
General 116.3: 228–92.
Savage-Rumbaugh, S., S. Shanker and T. Taylor 1996. Apes with Language. Critical
Quarterly 38.3: 45–57.
Schiffrin, D. 1987. Discourse Markers. Cambridge University Press.
Schlaggar, B., T. Brown, H. Lugar et al. 2002. Functional Neuroanatomical Differences
Between Adults and School-Age Children in the Processing of Single Words.
Science 296: 1476–79.
Searle, J. 1995. The Construction of Social Reality. New York: The Free Press.
1998. Mind, Language and Society. New York: Basic Books.
Seidenberg, M. and L. Petitto 1987. Communication, Symbolic Communication,
and Language: Comment on Savage-Rumbaugh, McDonald, Sevcik, Hopkins,
and Rupert (1986). Journal of Experimental Psychology: General 116.3:
279–87.
Sheets-Johnstone, M. 1990. The Roots of Thinking. Philadelphia: Temple University
Press.
Silverman, L., L. Bennetto, E. Campana and M. Tanenhaus 2007. Speech-and-Gesture
Integration in High Functioning Autism. Cognition 11.3: 380–93.
Skinner, B. F. 1938. Behavior of Organisms. New York: Appleton-Century Company.
Bibliography 295
1945. The Operational Analysis of Psychological Terms. In Cumulative Record.

Acton, MA: Copley.
1957. Verbal Behavior. Acton, MA: Copley Publishing Group.
1966. The Phylogeny and Ontogeny of Behavior. Science 153: 1205–13.
1972. A Lecture on ‘Having’ a Poem. Cumulative Record: A Selection of Papers by
B.F. Skinner, 3rd edn., pp. 345–55. New York: Appleton-Century-Crofts.
1981. Selection by Consequences. Science 213: 501–504.
1985. Cognitive Science and Behaviourism. British Journal of Psychology 76:
291–301.
1990. Can Psychology Be a Science of Mind? American Psychologist 45: 1206–10.
Slobin, D. 1990. The Development from Child Speaker to Native Speaker. In Cultural
Psychology. Essays on Comparative Human Development, ed. J. Stigler, R.
Schweder and G. Herdt. Cambridge University Press.
1996. From “Thought to Language” to “Thinking for Speaking”. In Rethinking
Linguistic Relativity, ed. J. J. Gumperz and S. C. Levinson, pp. 70–96. Cambridge
University Press.
Smith, B. H. 1978. On the Margins of Discourse: The Relation of Literature to Language.
University of Chicago Press.
1988. Contingencies of Value: Alternative Perspectives for Critical Theory.
1997. Belief and Resistance. Dynamics of Contemporary Intellectual Dynamics.
2005. Scandalous Knowledge. Edinburgh: University of Edinburgh Press.
Steier, F., ed. 1991. Research and Reflexivity. London: Sage.
Stephan, K., J. C. Marshall, K. J. Friston et al. 2003. Lateralized Cognitive Processes
and Lateralized Task Control in the Human Brain. Science 301: 384–86.
Stephens, G., L. Silbert and U. Hasson 2010. Speaker–Listener Neural Coupling
Underlies Successful Communication. Proceedings of the National Academy of
Sciences 107.32: 14425–30.
Strand, E. 1999. Uncovering the Role of Gender Stereotypes in Speech Perception.
Journal of Langue and Social Psychology 18.1: 86–99.
Sundberg, M. L. and J. Michael 2001. The Benefits of Skinner’s Analysis of Verbal
Behavior for Children with Autism. Behavior Modification 25: 698–24.
Sundberg, M. L., J. Michael, J. W. Partington and C. A. Sundberg 1996. The Role
of Automatic Reinforcement in Early Language Acquisition. Analysis of Verbal
Behavior 13: 21–37.
Symons, D. 1987. If We’re All Darwinians, What’s the Fuss About? In Sociobiology
and Psychology, ed. C. Crawford, M. Smith and D. Krebs. Hillsdale, NJ: Lawrence
Erlbaum Associates.
Tallal, P. and A. A. Benasich 2002. Developmental Language Learning Impairments.
Development and Psychopathology 14: 559–79.
Tanenhaus, M., M. J. Spivey-Knowlton, K. Eberhard and J. Sedivy 1995. Integration of
Visual and Linguistic Information in Spoken Language Comprehension. Science
268: 1632–35.
Tanenhaus, M., J. S. Magnuson, D. Dahan and C. Chambers 2000. Eye Movements
and Lexical Access in Spoken-Language Comprehension: Evaluating a
Linking Hypothesis Between Fixations and Linguistic Processing. Journal of
Psycholinguistic Research 29.6: 557–80.
296 Bibliography
Taylor, T. 1997. Theorizing Language. Amsterdam: Pergamon.

Thierry, G., P. Athanasopoulos, A. Wiggett, B. Dering and J.-R. Kuipers 2009.
Unconscious Effects of Language-Specific Terminology on Preattentive Color
Perception. Proceedings of the National Academy of Sciences 106.11: 4567–70.
Thomas, M. 2002. Development of the Concept of “the Poverty of the Stimulus.” The
Linguistic Review 19: 51–71.
Tobin, Y. 1990. Semiotics and Linguistics. London/New York: Longman.
Tomasello, M. 1995. Language Is Not an Instinct. Review of The Language Instinct.
Cognitive Development 10: 131–56.
1999. The Cultural Origins of Human Cognition. Cambridge, MA: Harvard
University Press.
2001. Could We Please Lose the Mapping Metaphor Please? Behavioral and Brain
Sciences 24.6: 1119–20.
2003. Constructing a Language: A Usage-Based Theory of Language Acquisition.
2008. Origins of Human Communication. Cambridge, MA: MIT Press.
2009. Why We Cooperate. Cambridge, MA: MIT Press.
Tooby, J. and L. Cosmides 1992. The Psychological Foundations of Culture. In Barkow
et al., pp. 19–136.
Tsuang, M., W. Stone and S. Faraone 2001. Genes, Environment and Schizophrenia.
British Journal of Psychiatry 178 (suppl. 40): s18–s24.
Uexküll, J. von 1909. Umwelt und Innenwelt der Tiere. Berlin: Springer.
Venezia, J. and G. Hickok 2009. Mirror Neurons, the Motor System and Language:
From the Motor Theory to Embodied Cognition and Beyond. Language and
Linguistics Compass 3/6: 1403–16.
Vološinov, V. N. 1973. Marxism and the Philosophy of Language, trans. L. Matejka and
I. R. Titunik. Cambridge, MA: MIT Press.
Vygotsky, L. 1962 [1934]. Thought and Language, ed. and trans. E. Hanfmann and G.
Vakar. Cambridge, MA: MIT Press.
1986 [1934]. Thought and Language, revised and edited by A. Kozulin. Cambridge
MA: MIT Press.
Waddington, C. 1957.The Strategy of the Genes: A Discussion of Some Aspects of
Theoretical Biology. London: Allen & Unwin.
1960. The Ethical Animal. London: George Allen.
1975. The Evolution of an Evolutionist. Ithaca, NY: Cornell University Press.
Watson, J. B. 1931. Behaviorism, 2nd edn. London: Kegan Paul.
Whitney, W. D. 1869. On the Present Condition of the Question as to the Origin of
Language. Transactions of the American Philological Society 1: 84–94.
Whorf, B. L. 1956. Language, Thought, and Reality. Selected Writings of Benjamin L.
Whorf, edited and with an introduction by J. B. Carroll. New York: John Wiley &
Sons and The Technology Press of Massachusetts Institute of Technology.
Wickelgren, I. 2005. Autistic Brains Out of Sync? Science 308: 1856–58.
Wierzbicka, A. 1998. “Sadness” and “Anger” in Russian: The Non-Universality of
So-Called “Basic Human Emotions.” In Speaking of Emotions: Conceptualization
and Expression, ed. A. Athanasiadou and E. Tabakowska, pp. 3–27. Berlin, New
York: Mouton de Gruyter.
Wilson, R. and F. Keil 1999. The MIT Encyclopedia of the Cognitive Sciences.
Bibliography 297
Winawer, J., N. Witthoft, M. Frank et al. 2007. Russian Blues Reveal Effects of
Language on Color Discrimination. Proceedings of the National Academy of
Sciences 104.19: 7780–85.
Wittgenstein, L. 1997 [1953]. Philosophical Investigations, 2nd German/English edn.,
trans. G. E. M. Anscombe. Oxford: Blackwell.
Woodward, J. and F. Cowie 2004. The Mind Is Not (Just) a System of Modules Shaped
(Just) by Natural Selection. In Contemporary Debates in Philosophy of Science,
ed. C. Hitchcock, pp. 312–34. Malden, MA: Blackwell.
Name index
Aarsleff, Hans, 11 Brannon, Elizabeth, 166–67

Abutalebi, Jubin, 223 Bréal, Michel, 11
Agnoli, Franca, 223 Broca, Pierre Paul, 74
Aitchison, Jean, 112–13 Brosnan, Sarah, 166
Akmajian, Adrian, Linguistics, 280 Brown, Penelope, 199
Alter, Stephen, 157 Politeness, 266
Andresen, Julie Tetel, 156 Brown, Roger, 183–84
Arbib, Michael, 146–47 A First Language, 233
Armstrong, David, 146–47 Brown, Roger Langham, 188
Attwood, Anthony, 178 Bruner, Jerome, 171, 183, 191, 273
Au, Terry Kit-fong, 37 Child’s Talk, 179–80
Augustine of Hippo, 190–91 Buttelmann, David, 171
Austen, Jane, 279
Austin, J. L., 43, 235 Campbell, R. Sherlock, 184
How To Do Things With Words, 25, 240–44, Cantlon, Jessica, 166–67
263–64, 267–69, 278 Carroll, John B., 202
Cavalli-Sforza, Luigi, 182
Balter, Michael, 73, 96 Chen, Keith, 165–66
Bapteste, Eric, 116, 139 Choi, Soonja, 190, 193–95
Barkow, Jerome, The Adapted Mind, 127 Chomsky, Noam, 1, 7, 12, 13, 17–18, 20, 33,
Barrett, Louise, Human Evolutionary 75–77, 78, 81, 85, 105–06, 108–09,
Psychology, 158–60 111–12, 117, 134, 149, 158, 182, 201,
Beadle, George, 83 202, 230, 233, 234, 238, 239–40,
Becker, Alton, 32, 192 259, 272
Benasich, April A., 34 Aspects of the Theory of Syntax, 13
Berlin, Brent, 200, 209 Language and Problems of Knowledge,
Basic Color Terms, 209–10 87–88, 106–07
Bertolani, Paco, 166 poverty of the stimulus, 244–46
Bialystok, Ellen, 183, 223 Review of Verbal Behavior, 229–32, 234,
Bickerton, Derek, 104 235, 236–37, 244
Bloom, Paul, 112–15, 122, 127, 190–91, Syntactic Structures, 246–47, 249–50,
240, 244 251–52, see also competence
Language and Space, 219–20 The Case Against B.F. Skinner 238
Bloomfield, Leonard, 22–23, 84, 235–36, 256 Christiansen, Morten, 132
Language, 7, 14, 74–75, 247–49, Chung, Cindy, 184
251–52, 261 Cohen, Jon, 117
Bolinger, Dwight, 83 Cohen, Ronald, 185
Bopp, Franz, 156–57 Condon, William S., 38
Boroditsky, Lera, 216–17 Connolly, William, 96
Borsoi, Diane, 185 Corballis, Michael, 146–47
Bowerman, Melissa, 190 Corbett, Greville, 217
Brace, C. Loring, 163 Cosmides, Leda, 76, 127, 182
298
Name index 299
The Adapted Mind, 127 Genesis and Development of a Scientific

Courchesne, Eric, 177–78 Fact, 7–8
Courtenary, Baudouin de, 26 Fodor, Jerry, 78, 188
Cowie, Fiona, 78–79, 80 Modularity of Mind, 76
Crick, Francis, 49, 81–82 Foucault, Michel, 234
Critchley, Macdonald, 104 Friederici, Angela, 73
Crompton, Robin, 166 Fromkin, Victoria, 107–08
Crow, Timothy, 109, 175–76 An Introduction to Language, 262–63
Futuyma, Douglas, Evolutionary Biology,
Damasio, Antonio, 270 see Evolutionary Biology
Danziger, Shai, 279
Darwin, Charles, 239 Gabrieli, John, 175
as inspiration for linguistic theory, 1 Galton, Francis, 41
On the Origin of Species, 11, 118 Gardiner, Alan H., 233
second neo-Darwinian synthesis, 135 Geertz, Clifford, 32–33
survival, 54 Gellatly, Angus, 210
Davies, Ian, 217 Gibbons, Ann, 78
de Brosses, Charles, 156 Gibson, Kathleen, 104
de Waal, Frans, 104, 166, 167, 168, 170, Givón, Talmy, 198, 199
171–72, 276 Gleason, H. A., Introduction to Descriptive
Deacon, Terrence, 104, 155, 170, 211 Linguistics, 209
Symbolic Species, 160–62 Goffman, Erving, 36
Dediu, Dan, 65–66, 276 Goldberg, Adele, 4, 43, 66, 255
Descartes, René, as inspiration for linguistic Constructions at Work, 252, 253–55,
theory, 1, 123–24 256–59
Deutsch, Diana, 200 Goldin-Meadow, Susan, 35, 66–67, 184–85
Dick, Anthony, 35 Hearing Gesture, 269–70
Dickerson, Paul, 178 Goldstein, Michael, 38
Dolan, R. L., 59 González, Julio, 97
Donald, Merlin, Origins of the Modern Mind, Gordon, Peter, 206, 207
167–68 Griffiths, Paul, 42, 56–64, 166
Doolittle, W. Ford, 116, 139 What Emotions Really Are, 182
Dubner, Stephen, 165–66
Dugum Dani, 200 Hakuta, K., 183
Dunbar, Robin, Human Evolutionary Harris, Roy, 18
Psychology, 158–60 Hauser, Mark, 109, 111–12, 117, 134, 158
Moral Minds, 280
Eckman, Paul, 58, 78 Henthorn, Trevor, 200
Elman, Jeffrey, 36 Hespos, Susan, 190, 192, 193, 204
Elton, Sarah, 166 Hickok, Gregory, 147
Evans, Nicholas, 69, 208 Hill, Jane, 104, 133
Everett, Dan, 111, 208 Holle, Henning, 35
Humboldt, Wilhelm von, 13, 25–26, 187–88
Falk, Dean, 157–58 Hunt, Earl, 223
Fedorenko, Evelina, 72 Hurst, Jane, 64
Fillmore, Charles Huxley, Sir Julian, Evolution, 118, 120
Grammatical Constructions and Linguistic
Generalizations, 253 Iverson, Jana, 35
Regularity and Idiomaticity in Grammatical
Constructions, 253 Jablonka, Eva, 52, 68
Fisher, R. A., 121 Evolution in Four Dimensions, 135, 136,
Fisher, Simon, 64, 65 139, 144–45
Fishman, Joshua, 203–05, 214 Jackendoff, Ray, 41–42, 108, 109, 111, 112,
Fitch, W. Tecumseh, 109, 111–12, 117, 134, 122, 128, 150–51, 155, 158, 168,
147, 158 181–82, 253, 255
Fleck, Ludwik, 93, 194 Foundations of Language, 229–30
300 Name index
Jacob, François, 83 Lewontin, Richard, 3

Jakobson, Roman, 18–19, 23, 26, 29, 266–67 Li, Charles, 198
James, William, 4, 20–22, 32, 85–86, 105, Lieberman, Philip, 65, 70, 111, 163, 169
137, 194, 216 Lobaugh, Nancy J., 80–81
Jespersen, Otto, 146, 155, 156, 157, 233, Locke, John, 186–87
271–72 An Essay Concerning Human
Language Understanding, 16
Its Nature, Development and Origin, 26 London Philological Society, 11
Jessell, Thomas, 174–75, 256–57 Lotto, Beau, 129, 131
Johnstone, Barbara, 36–37, 261 Lowth, Robert, 279
Luria, Alexander, 210
Kandel, Eric, 174–75, 256–57 Lycett, John, Human Evolutionary
Kant, Immanuel, 221 Psychology, 158–60
Karmiloff, Kyra, 68 Lyn, Heidi, 150
Pathways to Language, 159–60, 173
Karmiloff-Smith, Annette, 4, 68, 78 MacDonald, John, 34, 79
Pathways to Language, 159–60, 173 MacNeilage, Peter, 70–71, 109–10, 146,
Kay, Paul, 200, 209, 210–11 147–48, 163, 170, 189–90
Basic Color Terms, 209–10 The Origin of Speech, 39
Grammatical Constructions and Linguistic Maestripieri, Dario, 157
Generalizations, 253 Maffi, Luisa, 210–11
Regularity and Idiomaticity in Grammatical Malthus, Thomas, Essay on Population, 118
Constructions, 253 Mandler, Jean, 190
Kempton, Willett, 210 Marr, David, Vision, 129
King, Andrew, 38 Martin, Laura, 200
Kirby, Simon, 132 Maturana, Humberto, 4, 6, 28, 32, 42, 52, 54,
Kohler, Evelyne, 109 86–87, 88, 196
Köhler, Wolfgang, 168 Autopoiesis and Cognition, 89, 92
Krauss, Robert, 35 Origen de las Especies por medio de la
Kruszewski, Mikolaj, 26 Deriva natural o…, 136
Kuhn, Thomas, Structure of Scientific Origin of Species by Means of Natural Drift
Revolutions, 7 or…, 135, 136, 143–45
“Scientific and Philosophical
Labov, William, 28, 29 Theories”, 42–47
Ladd, D. Robert, 65–66, 276 The Biology of Cognition, 155
Lai, Cecilia, 64, 65 The Tree of Knowledge, 139
Lakoff, George Max Müller, Friedrich, 104, 109, 155, 156–57,
Where Mathematics Comes From, 207, 158, 163–65, 188
208, 222 Mayr, Ernst, 115
Women, Fire, and Dangerous Things, 253 McDonough, Laraine, 190
Lamarck, Jean-Baptiste, 53 McGurk, Harry, 34, 79
Lamb, Marion, 52, 68 McIntosh, Anthony R., 80–81
Evolution in Four Dimensions, 135, 136, McMurray Bob, 183
139, 144–45 McNeill, David, 35, 146–47, 185, 267
Langacker, Ronald, Foundations of Cognitive Gesture and Thought, 262, 270
Grammar, 252 Meltzoff, Andrew, 273–74
Lavoisier, Antoine-Laurent, 126 Mithun, Marianne, 111
Lenneberg, Eric, Biological Foundations of Monaco, Anthony, 64
Language, 104 Morsella, Ezequiel, 35
Levinson, Stephen C., 69, 184, 199, 208, 209, Mpodozis, Jorge, 6
211, 223–24, 267 Origen de las Especies por medio de la
Politeness, 266 Deriva natural o…, 136
Space in Language and Cognition, 3, 34, Origin of Species by means of Natural Drift
220–23 or…, 135, 136, 143–45
Levitt, Steven, 165–66 Mufwene, Salikoko, 31
Name index 301
Mulcaster, Richard, 41 Savage-Rumbaugh, Susan, 15, 149

Schiffrin, Deborah, 235, 252
Nelson, Katherine, 37, 273 Schlagger, Bradley, 184
Young Minds in Social Worlds, 273 Scholz, Barbara, 251, 258
Nevins, Andrew, 48, 111, 193, 208 Schwartz, James, 174–75, 256–57
Newman, John D., 158 Searle, John, 44, 261
Newmeyer, Frederick, 128–29 Mind, Language and Society, 44–45
Nicholas, Evans,, 223 Seidenberg, Mark, 149
Nunberg, Geoffrey, 33 Seyfarth, Robert, 150
Núñez, Rafael, 219 Shapero, Dana, 223
Where Mathematics Comes From, 207, Sheets-Johnstone, Maxine, The Roots of
208, 222 Thinking, 156
Silverman, Laura, 178
O’Connor, Mary Catherine, Regularity Skinner, B. F., 76, 110, 111, 146, 155, 158,
and Idiomaticity in Grammatical 233, 234, 238–39, 266–67
Constructions, 253 Verbal Behavior, 7, 25, 105–06, 229–30
O’Higgins, Paul, 166 aesthetic feel, 230–31
Oyama, Susan, 4, 5–6, 41, 42, 48–56, 57, autoclitic function of verbal
61, 238 behavior, 235
The Ontogeny of Information, 40–41 comparison with Austin, 240–44
ethical story, 235–39
Pääbo, Svante, 117 evolutionary story, 239–40
Packman, Ann, 34 inside/outside story, 232–33
Pennebaker, James, 184 metalinguistic story, 233–35
Pettito, Laura, 149 style, 231–32
Pica, Pierre, 205, 207 terminology, 234
Pinker, Steven, 37, 41, 108–09, 111, 112–15, Slobin, Dan, 189, 224
117, 122–23, 127, 128, 149, 150–51, Smith, Barbara Herrnstein, 93, 266–67
155, 158, 168, 188, 240, 244, 253 Smith, John Maynard, 277
The Language Instinct, 107–08, 115, Spelke, Elizabeth, 190, 192, 193, 204
253, 271 Spencer, Herbert, 196, 197
Planck, Max, 105 Stephan, Klaas, 80–81
Pollick, Amy, 276 Stephens, Greg, 147
Port, Robert, 28, 82 Strand, Elizabeth, 79–80
Pruetz, Jill D., 166 Sweetser, Eve, 219
Prusiner, Stanley, 49 Symons, Donald, 129–30
Pullum, Geoffrey, 251, 258
Purves, Dale, 129, 131 Tallal, Paula, 34
Tanenhaus, Michael, 80
Quine, W. V., Mathematical Logic, 24 Taylor, Talbot, 70
Terrance, Herbert, 149
Rickford, John R., 251 Thierry, Guillaume, 211–12
Rizzolatti, Giacomo, 146–47 Thomas, Margaret, 246
Rodman, Robert, 107–08 Tomasello, Michael, 35, 109, 111, 155,
An Introduction to Language, 262–63 163–64, 171, 177, 180, 185–86, 191,
Rorty, Richard, 238 194, 246, 251, 253, 254
Rowe, Meredith, 66–67 Constructing a Language, 185
Cultural Origins of Human Cognition,
Sander, Louis W., 38 170–71, 208
Sapir, Edward, 198–99, 201, 223, 225, 233 Tooby, John, 76, 127, 182
Language, 14, 30 The Adapted Mind, 127
Saussure, Ferdinand de, 12–13, 24, 25, 31,
73–74, 94, 243 Varela, Francisco, 4, 6, 28, 32, 52, 54, 86–87,
Cours de linguistique générale, 16–17, 88, 89–97, 196
188–89, 262, see also langue Autopoiesis and Cognition, 89, 92
302 Name index
Varela, Francisco (cont.) Whitehead, Alfred North, 105, 233

The Tree of Knowledge, 139 Whitney, William Dwight, 11,
Vargha-Khadem, Faraneh, 64 156–57
Venezia, Jonathan, 147 Whorf, Benjamin Lee, 201–02, 204, 212–16,
Vološinov, V. N., 4, 27 219, 220, see also Sapir-Whorf
Marxism and the Philosophy of Language, hypothesis
25, 26 Wickelgren, Ingrid, 177–78
Vygotsky, Lev, 4, 20, 194, 270 Wiley, E., 183
Winawer, Jonathan, 212
Waddington, Conrad, 4, 16–17, 42, 49, 53–55, Wittgenstein, Ludwig, 190
135, 153, 196–97, 222 Wright, Sewall, 121
Ward, Robert, 279
West, Meredith, 38 Yuriko, Oshima-Takane, 37
Subject index
abstract objectivism, 25 orangutans, 166

actions of actions, 91–92 blind children, language development, 78, 173
activation, 3 bow-wow theory of speech, 156–57
adaptation, in self-realizationist framework, brain development, co-expansion of neocortex
137–38 and thalamus, 198
Adapted Mind (Jerome Barkow, Leda brain function
Cosmides and John Tooby), 127 historiographic review, 73–82
adopted culture, 127 location of language, 70–73
affect programs, 57, 58–59 Broca’s area, 70, 71–72
allopoietic unity, 97 monkey homologue, 109
amae, 62, 63
ambiguity, 231 canalization, 51, 53–54, 222–23
American Philological Association, 11 capuchin monkeys
amok, 60–61 sense of fairness, 166
analogues, 107 use of money, 165–66
analogy, 247–48 Cartesian-inspired linguistics, 1
Aspects of the Theory of Syntax (Noam cause/evoke distinction, 87–88
Chomsky), 13 Central Dogma of Molecular Biology (Francis
ASPM gene, 65–66 Crick), 49
assessment dimension, 264 child, as witness, 128
attending, 224–25 Child’s Talk (Jerome Bruner), 179–80
autism, 177–78 childhood vocabulary explosion, 183
autopoiesis, 6, 52, 68, 89–97, 137 chimpanbeings, 15
glossary, 97–99 chimpanzees
Autopoiesis and Cognition (Maturana and language abilities, 149–50
Varela), 89, 92 toolmaking, 166
autopoietic unity, 97 Chinese, character for the word crisis,
Aymara, 219 199, 200
chreod, 53, 221
babbling, 169–70 civilized script, 132–33
Basic Color Terms (Brent Berlin and Paul code theorism, 16–17, 22
Kay), 209–10 cognition (autopoiesis), 97
BCI (backgrounded constructions are cognitive construction grammar (CCG),
islands), 43 252–59
behavior (autopoiesis), 97 cognitive science, 69–70
behaviorism. See radical behaviorism cognitive tunings, 92–93
Skinner, B. F. cognitive turnings, 92–93
bilingualism, 223 color
Biological Foundations of Language (Eric Dugum Dani bright/dull terms, 200
Lenneberg), 104 human perception despite language, 200
Biology of Cognition (Humberto identified by single word, 204
Maturana), 155 Sapir-Whorf hypothesis, 209–12
bipedal locomotion, 162–63 comparative semantics, 267
303
304 Subject index
competence, 12, 13–14, 15, 20, 94 evolutionary homology, 107

complexity, 152–53 evolutionary scenarios
of languaging vs. improvement, 151–52 anthropology, 157–58
conative function, 19 continuity vs. discontinuity dichotomy,
Concerning Human Understanding (John 104–08, 112–16
Locke), 16 defining, 116–17
congruence (autopoiesis), 97 emerging story, 132–35
consensual coordination, 97 neurobiological, 160–62
Constructing a Language (Michael nineteenth-century linguistic heritage,
Tomasello), 185 155–57
Constructions at Work (Adele Goldberg), 252, overview, 162
253–55, 256–59 primate vocalization, 108–11
constructivism, 2–4 psychological, 158–60
constructivist interactionism, 51 standard story, 103–04
continuist script, 134–35 tree structures, 115–16, 124. See also
continuity theories, 104 self-realizationist script;
continuity/discontinuity conceptual complex self-reproductionist script
in evolutionary explanations, 116–17 Extreme Minimalism, 112
conventional dualism, 50
corpora, use of, 249–52 fairness, capuchin monkeys, 166
counterfactual reasoning, 257–58 fine motor skills, 171
Cours de linguistique générale (Ferdinand de First Language (Roger Brown), 233
Saussure), 16–17, 188–89, 262 first word, 146
crossmodal communication, 78–81 fitness, 118–20
Cultural Origins of Human Cognition flow, 215–16
(Michael Tomasello), 170–71, 208 flower-petal model of speciation, 139
food preferences development, 68
deleterious alleles, 142 form, 248–49
Descartes’s problem, 245 form–meaning pairings, 114
developmental dyslexia, 174–75 Foundations of Cognitive Grammar (Ronald
developmental explanation, 3 Langacker), 252
developmental systems theory, 5–6, Foundations of Language (Ray Jackendoff),
40–42, 48–56 229–30
reliability of systems, 172–73 FOXP2 gene, 64–65
ding-dong theory of speech, 156–57 functionalists, 265
discontinuity theories, 104
double vision, 85–86 Genesis and Development of a Scientific Fact
drift (autopoiesis), 98. See also ontogenic drift, (Ludwik Fleck), 7–8
phylogenic co-drift; phylogenic drift genetic assimilation, 53
dyslexia. See developmental dyslexia genetic code, 16–17
genetic drift, 98
ecological approach, 133 genetic imperialism, 50–51
emotion, Paul Griffiths’ clarification of, 57–64 gesture
emotive function, 19 autism and, 178
emulation/imitation, 171 during speech, 35
environment, 51–52 protolanguage leading to speech, 146
epidemiological culture, 127 Gesture and Thought (David McNeill),
epiphenomena, 48 262, 270
Eskimo, words for snow, 199, 200 gesturing practices, as linguistic
Essay on Population (Thomas Malthus), 118 interactants, 66–67
Eternal Form, 172–73 got-X? construction, 258
evaluation, criterion of, 43–44 Grammatical Constructions and Linguistic
Evolution (Sir Julian Huxley), 118, 120 Generalizations (Fillmore and
Evolution in Four Dimensions (Eva Jablonka Kay), 253
and Marion Lamb), 135, 136, 139, 144–45 grammatical constructions, as linguistic
Evolutionary Biology (Douglas Futuyma), interactants, 66
120–21 Great Chain of Being, 124
Subject index 305
Grimm’s Law, 12 problems with, 260–62

grooming at a distance, 160, 162 purpose of issuing utterances, 268–69
Growth Point, 270 irrelevant conditions, 231–32
Guinea baboons island constructions, 42–43
post-coital grunting, 157
Klinefelter’s syndrome, 175
handedness, and emergence of language in the knowledge (autopoiesis), 98
species, 174–75 knowledge sharing, 93–94
harmony of illusions, 93
hearing development, 68 la-la theory of speech, 157
Hearing Gesture (Susan Goldin-Meadow), language
269–70 differences within, 190
heterogeneous construction, 61 purpose of, 16, 18
higher cognitive motivational complexes, language (autopoiesis), 98
57, 62–64 Language (Edward Sapir), 14, 30
homologues, 107 Language (Leonard Bloomfield), 7, 14, 74–75,
homoplasy, 107 247–49, 251–52, 261
Hopi, 212–13 Language (Otto Jespersen), 26
horizontal/vertical, 216–17 language acquisition, 178–86
How To Do Things With Words (J. L. Austin), length of developmental linguistic system in
25, 240–44, 263–64, 267–69, 278 individual’s life, 183–86
human beings Language and Problems of Knowledge (Noam
closeness to chimpanzees, 117 Chomsky), 87–88, 106–07
defining characteristics, 167, 168 Language and Space (Paul Bloom et al.),
babbling, 169–70 219–20
imitation, 170–71 Language Instinct (Steven Pinker), 107–08,
skin thinness, 171–72 115, 253, 271
special nature of, 6 language learning impairment (LLI), 34
speed of evolution, 121 languaging, 5, 28–31
human cerebral asymmetry. See torque as an orienting behavior, 31–39, 145,
Human Evolutionary Psychology (Louise 153–54
Barrett, Robin Dunbar and John Lycett), autopoietic definition, 98
158–60 continuist script, 145–51
human population, 277 developmental problems, 172–73
humanzees, 15 evolving complexity vs. improvement,
hypostasis, 22–26 151–52
instantiating events, 168–69
ideas, progress of, 104–05 languaging living beings, 36–37, 91–92, 181,
Ideogrammes, 16–17 223–24
imitation, 170–71 definitions, 142–43
Immediacy of Experience Principle, 208 langue, 12–14, 15, 20
individualistic subjectivism, 25–26 LET ALONE construction, 253
ineffability, 35–36 linguistic domains, 92
intention-reading, 185–86 linguistic historiography, role in reimagining
interactants, 55–56 linguistics, 4–5
internal connectivity, criterion of, 42–43 linguistic interactants, possible
Introduction to Descriptive Linguistics (H. A. candidates, 64–67
Gleason), 209 linguistic relativity, 201–02
Introduction to Language (Fromkin and linguistics
Rodman), 262–63 banned discussions of origin of
introductory linguistics textbooks language, 11
basic unit of analysis, 263–68 interdisciplinary connections, 82
central feature of languaging, 270–71 mechanistic universe and, 124–26
incorporating Austin, 262–63 phenomena to be explained, 47–48
ontogeny, 271–75 reimagining, 4–5
phylogeny, 275–78 Linguistics (Adrian Akmajian et al.), 277
phenomena properly included, 269–70 listener, importance of, 14–15
306 Subject index
living beings open-ended counting routine, 134

adaptation in self-realizationist framework, operational closure, 98
137–38 orangutans
definitions, 135–36 bipedal locomotion, 166
organization (autopoiesis), 98
manifested/manifesting (Hopi), 212–13 orientation, 34
mapping metaphor, 35 Origen de las Especies por medio de la Deriva
Marxism and the Philosophy of Language natural o… (Maturana and Mpodozis,
(V. N. Vološinov), 25, 26 trans. Magro and Andresen), 136
materialistic theory, 74 Origin of Species (Charles Darwin), 11, 118
Mathematical Logic (W. V. Quine), 24 Origin of Species by means of Natural Drift
meaning, and context, 264–65 or… (Maturana and Mpodozis, trans.
mentalese, 188 Magro and Andresen), 135
metalanguage, 18 Origin of Species by Means of Natural Drift
metalingual function, 19, 23 or… (Maturana and Mpodozis, trans.
Microcephalin, 65–66 Magro and Andresen), 136, 143–45
Mind, Language and Society (John Origin of Speech (Peter MacNeilage), 39
Searle), 44–45 Origins of the Modern Mind (Merlin Donald),
Minimalist Program, 112 167–68
mirror neurons, 147–48
misunderstanding, 28–29 parade case, 245–46
modularity parole, 13
in biology, 82–84 partial constructivism, 3
of brain function, 76–78, 81 Pathways to Language (Karmiloff and
Modularity of Mind (Jerry Fodor), 76 Karmiloff-Smith), 159–60, 173
modules, 58–59 pattern-finding, 185–86
money, 215–16 perfect pitch, 200
monomorphic traits, 58 performance, 13
month sequence images, 217–18 perturbation (autopoiesis), 98
Moral Minds (Mark Hauser), 280 phatic function, 19, 29
mosaic traits, 78–79 phenomenological domain, 98
Mundurukú, 200, 205–06 phenomenology (autopoiesis), 98
counting, 200, 205–06, 207, 225 philologism, 20, 22, 26–31, 38–39
music, 168 philology, 20, 38–39, 276–77
myelination, 170 phylogenic co-drift, 138–39
phylogenic drift, 86–87, 138–39
nature/nurture dichotomy, 5–6, 40, 41, 50 Pirahã, 111, 134, 200, 208
navigation, 222 one/two/many counting, 206, 225
neurobiology, 69 Plato’s problem, 245
nucleocytoplasmic gap, 49, 51–52 Plotinus, 124
number poetic function, 19
human ability despite language, 200–01 Politeness (Brown and Levinson), 266
Mundurukú counting, 200, 205–06, polymorphic traits, 58
207, 225 pooh-pooh theory of speech, 155
Pirahã one/two/many counting, 206, 225 poverty of the stimulus, 244–46
Sapir-Whorf hypothesis, 206–09 precedence, 247
preformationism, 3, 40
observer primates
and domains, 136–37 vocalization, 109. See also capuchin
child as witness, 128 monkeys; chimpanzees; Guinea baboons;
how to observe myself as, 279 orangutans
status of, 84–88 evolutionary scenarios, 108–11
observer (autopoiesis), 98 primatology, 276
ontogenic drift, 86–87, 98, 138 priming, 256–57
Ontogeny of Information (Susan prions, 49
Oyama), 40–41 pronoun use, 184
Subject index 307
psychosis, 176–77 specify (autopoiesis), 98–99

putting the baby down hypothesis, 157–58 specify/trigger distinction. 88.
See also trigger
racism, 182 speech, and language, 34
radical behaviorism, 232–33, 238 stimulus-free learning, 244. See also poverty
Rambo behavior syndrome, 60–61 of the stimulus
rationalized evolutionary accounts of structural coupling, 52, 99
language, 126 structure (autopoiesis), 99
reconstructed culture, 127 Structure of Scientific Revolutions (Thomas
recursion, 48, 111–12, 133 Kuhn), 7
referential function, 18, 19, 24 subitizing, 207
Regularity and Idiomaticity in Grammatical Subject-Auxiliary Inversion (SAI), 258
Constructions (Fillmore, Kay, and superlunary, 57
O’Connor), 253 Symbolic Species (Terrence Deacon), 160–62
reification, 22 Syntactic Argument Structure (SAS),
representationalism, 85 distinction from cognitive construction
reproduction (autopoiesis), 98 grammar, 255
rhizome structures, 277–78 Syntactic Structures (Noam Chomsky),
right speech, 278–79 246–47, 249–50, 251–52
Roots of Thinking (Maxine Sheets-
Johnstone), 156 tableau, 124
Rubicon, 104, 165 taste preferences development, 68
temporal ranges, 267
SAE (Standard Average European), 212, thinking for both gesturing and speaking,
215, 219 189–90, 224
Sapir-Whorf hypothesis, 202 thought
color, 209–12 as contemporaneous with language, 188–89
levels of interpretation, 203–05 as preceding language, 186–87
number, 206–09 as succeeding language, 187–88
space, 219–23 language and, 186, 189–90, 195
time, 212–19. See also linguistic relativity tight fit/loose fit distinction, 190–95, 204
Scheherazade Effect, 157, 159, 160 time
schizophrenia, 176, 177 Hopi conception of, 202
science, philosophy and, 44–47 Sapir-Whorf hypothesis, 212–19
second-language acquisition, critical-period, tone languages, 65–66, 275–76
183–84 toolmaking, chimpanzees, 166
self-realizationist script, 136–45 torque, 175–76
self-reproductionist script, 118–23 transmitted culture, 127
semantic congruity effect, 167 Tree of Knowledge (Maturana and Varela), 139
sexuality, as a systemic phenomenon, 141–42 trigger (autopoiesis), 99
Shannon-Weaver information theory, 16–17 tunátya (Hopi), 213
silent scripts, 124–27 Turner’s syndrome, 175
skin color, 78, 121 two-tier evolution, 153
skin thinness, 171–72
Social Contract hypothesis, 159, 160, 162 unity (autopoiesis), 99
social couplings, 91–92 uptake, 267–68
Social Gossip hypothesis, 158–59, 160 use of money, capuchin monkeys, 165–66
socially constructed emotions, 57, 59–62
Société de Linguistique de Paris, 11 Verbal Behavior. See Skinner, B. F., Verbal
Space in Language and Cognition (Stephen Behavior
Levinson), 3, 34, 220–23 Verner’s Law, 12
space, Sapir-Whorf hypothesis, 219–23 virtuous circles, 196–99
spatial categories learning, 190–95 vision, 129–31
spatial frames of reference, 184–85 in linguistic processing of speech, 34.
species metaphor, 181–82 See also visual domains
Specific Language Impairment, 33–34 Vision (David Marr), 129
308 Subject index
visual cortex, 96 Women, Fire, and Dangerous Things (George

visual domains, 95–96, 151–52 Lakoff), 253
vocalization, in primates, 109 Woodward, James, 78–79, 80
written, vs. spoken, 242–44
Wernicke’s area, 70, 71
What Emotions Really Are (Paul XXX syndrome, 175
Griffiths), 182
Where Mathematics Comes From (George year sequence images, 217–18
Lakoff and Rafael Núñez), 207, yo-he-ho theory of speech, 155–56
208, 222 Young Minds in Social Worlds (Katherine
Williams syndrome, 33–34, 78, 173–74 Nelson), 273

Andresen J.T.-Linguistics and Evolution - A Developmental Approach-CUP (2013) PDF

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Andresen J.T.-Linguistics and Evolution - A Developmental Approach-CUP (2013) PDF

Diunggah oleh

Hak Cipta:

Format Tersedia

Linguistics and Evolution

Evolutionary linguistics – an approach to language study that takes into

Ju lie Tetel A n d re s e n is Associate Professor in the Department of English

Julie Tetel Andresen

Printing in the United Kingdom by TJ International Ltd. Padstow Cornwall

List of figures page vi

Introduction: Historiography’s contribution to

Part I Theoretical considerations 9

Part II A developmental systems linguistics 101

Part III What to do next 227

1.1 Saussure’s talking heads1 page 17

What is needed for a twenty-first-century linguistics is an understanding of

The related term developmental also requires a bit of explanation. In a book

In addition, the use of the term constructivism here is generally consistent

reimagined linguistics, one of Fleck’s central points is pertinent at the outset:

Darwin’s Origin of the Species (1859) made an immediate impression on phi-

evolutionary account of the subject matter of linguistics: (i) twentieth-century

A first historiographic review: twentieth-century objects of linguistics

A speaker-biased linguistics, furthermore, pulls our attention away from the

Figure 1.1 Saussure’s talking heads

idealized speaker–listener. This idealized speaker–listener was thereby relieved

The insufficiencies of narrowing linguistic investigation to only one function

We find ourselves in a bit of a conceptual and descriptive pretzel. This is

of hypostasis in the discipline of linguistics is particularly bedeviling because

written records of an Indo-European language, for instance, are of Sanskrit

and ­low-dimensional and which exists only as a set of statistical regularities

Stopping to explore the borderland is a fine thing to do as a linguist, but

bring such a reconceptualization into existence. My intervention in the history

Languaging as an orienting behavior

mind is a term denoting a class of skills, propensities, capacities, tendencies, habits; it

In other words, the term mind continues to have a place.

For Chomsky, Williams’ syndrome and Specific Language Impairments (SLIs)

(i) to scaffold the cognition of languaging living beings and to be an activity

(iv) to include, crucially, gesture as a dimension of its multimodality. The dra-

The term languaging living being facilitates the incorporation of under-

The shift from language to languaging requires a concomitant shift from a

thinking Oyama recommends that we abandon the old opposition of nature

more than useless; it is counterproductive in that it short-circuits the under-

Maturana’s “Scientific and Philosophical Theories”

Internal connectivity is often spoken of in terms of predictive power, that

body of data, recognize certain phenomena, follow certain methods in produ-

an objective reality if the consequence of doing so is desirable. For instance,

A parade of linguistic phenomena to be explained

two levels was native-speaker intuition about grammatical and ungrammat-

Eliminating the nature/nurture dichotomy once and for all

that it is possible to have an evolutionary perspective without the concept of

isolation is the desire to contain ontogenic variety within genetic boundaries.

environment of an organism constantly changes. The point is that the environ-

The idea that behavior is a consequence of the organism’s ongoing interactions

Transgenerational stability and change, which characterize evolution, depend

canalization is not linked to adaptation or to natural selection. Different kinds

Figure 2.1 Part of an epigenetic landscape. The developing organism is the

Figure 2.2 Interactions underlying the epigenetic landscape. The pegs in

some trait or behavior as a result of (genetic) nature or (environmental) nurture

Applied Oyama: Paul Griffiths’ What Emotions Really Are

we could use an institutional definition and say that a philosopher is anyone

Emotion as an affect program

cognitive processes and informationally encapsulated. For instance, we do

Socially constructed emotions

Higher cognitive motivational complexes

recognize as cool means-ends reasoning. Disclaimed action emotions easily

Beginning the search for linguistic interactants

A team of researchers at the Max Planck Institute for Evolutionary

List of figures page vi

Part I Theoretical considerations 9

Part II A developmental systems linguistics 101

Part III What to do next 227

1.1 Saussure’s talking heads1 page 17

and low-dimensional and which exists only as a set of statistical regularities