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
Bibliography 281
Name index 298
Subject index 303
v
Figures
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
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
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
6 Historiography’s contribution to linguistics
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
8 Historiography’s contribution to linguistics
Theoretical considerations
1 Language → languaging
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
12 Language → languaging
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;
14 Language → languaging
(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).
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.
16 Language → languaging
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
Twentieth-century objects of linguistics 17
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.
18 Language → languaging
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,
Twentieth-century objects of linguistics 19
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.
20 Language → languaging
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)
22 Language → languaging
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:
Philology/philologism 23
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.
24 Language → languaging
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
Philology/philologism 27
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
Philology/philologism 29
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).
32 Language → languaging
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:
Languaging as an orienting behavior 33
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.
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).
38 Language → languaging
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
Languaging as an orienting behavior 39
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
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
(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
46 Developmental systems theory
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
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
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
Eliminating the nature/nurture dichotomy 51
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)
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).
54 Developmental systems theory
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
56 Developmental systems theory
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.
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:
Applied Oyama: Griffiths’ What Emotions Really Are 61
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,
62 Developmental systems theory
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).
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
64 Developmental systems theory
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
68
Opening on to the interior 69
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.
70 A twist in the cognitive turn
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
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
The brain in twentieth-century linguistics 75
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
76 A twist in the cognitive turn
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)
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
78 A twist in the cognitive turn
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
The brain in twentieth-century linguistics 79
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
The brain in twentieth-century linguistics 81
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).
84 A twist in the cognitive turn
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
“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)
86 A twist in the cognitive turn
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
88 A twist in the cognitive turn
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.
90 A twist in the cognitive turn
Dynamics Boundary
(metabolism) (membrane)
The exterior arrow indicates the The interior arrow indicates circular
unity producing its own border organization with operational closure
of the nervous system
environment
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
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
92 A twist in the cognitive turn
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
94 A twist in the cognitive turn
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
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
96 A twist in the cognitive turn
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
98 A twist in the cognitive turn
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
103
104 Standard story and the self-reproductionist script
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
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
108 Standard story and the self-reproductionist script
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.
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.
112 Standard story and the self-reproductionist script
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
Continuing continuity versus discontinuity 113
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
114 Standard story and the self-reproductionist script
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
116 Standard story and the self-reproductionist script
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.
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.
120 Standard story and the self-reproductionist script
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
122 Standard story and the self-reproductionist script
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.
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.
126 Standard story and the self-reproductionist script
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.
Done with Descartes 127
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
128 Standard story and the self-reproductionist script
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
Done with Descartes 129
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
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
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)
138 Emerging story and the self-realizationist script
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.
The self-realizationist script 141
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
142 Emerging story and the self-realizationist script
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.
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
148 Emerging story and the self-realizationist script
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
A continuist script for languaging 149
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
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
154 Emerging story and the self-realizationist script
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
158 Emerging story and the self-realizationist script
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
A range of evolutionary scenarios 159
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,
A range of evolutionary scenarios 161
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
A range of evolutionary scenarios 163
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
164 Emerging story and the self-realizationist script
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.
165
166 The ontogenic script begins
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
168 The ontogenic script begins
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
Getting off the ground & into the languaging loop 169
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.
170 The ontogenic script begins
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
Getting off the ground & into the languaging loop 171
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,
172 The ontogenic script begins
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).
174 The ontogenic script begins
• 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
When things go wrong 175
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
176 The ontogenic script begins
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).
When things go wrong 177
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
178 The ontogenic script begins
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).
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
180 The ontogenic script begins
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)
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
Ontogeny: the very beginnings 181
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?
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.
Ontogeny: the very beginnings 183
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:
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.
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.
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.
192 The ontogenic script begins
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:
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).
Case in point: the tight-fit/loose-fit distinction 195
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,
198 The ontogenic script continues
1
There is not room here to explore Brown and Levinson’s concept of face.
200 The ontogenic script continues
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.
The Sapir-Whorf hypothesis 201
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
202 The ontogenic script continues
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
The Sapir-Whorf hypothesis 205
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
206 The ontogenic script continues
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
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
210 The ontogenic script continues
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
212 The ontogenic script continues
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
214 The ontogenic script continues
THOUGHT
GENES LANGUAGE
FLOW
BLOOD MONEY
WATER CURRENCY
TIME
HISTORY
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.
216 The ontogenic script continues
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
218 The ontogenic script continues
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
220 The ontogenic script continues
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).
222 The ontogenic script continues
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
224 The ontogenic script continues
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
232 Revisit: Skinner, Chomsky and construction grammar
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.
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.
234 Revisit: Skinner, Chomsky and construction grammar
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).
Skinner’s exile 235
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.
238 Revisit: Skinner, Chomsky and construction grammar
2
1977, an exchange of letters between Murray and Skinner, 1977, personal communication.
Skinner’s exile 239
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 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”
242 Revisit: Skinner, Chomsky and construction grammar
(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
244 Revisit: Skinner, Chomsky and construction grammar
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.
3
Unpacking the terminology, along with Skinner’s use of it and Chomsky’s criticisms, goes
beyond the scope of this chapter.
246 Revisit: Skinner, Chomsky and construction grammar
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,
248 Revisit: Skinner, Chomsky and construction grammar
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
From Bloomfield to Chomsky 251
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.
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
Cognitive construction grammar 255
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).
256 Revisit: Skinner, Chomsky and construction grammar
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
258 Revisit: Skinner, Chomsky and construction grammar
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.
Cognitive construction grammar 259
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
260
Problem: theoretical disintegration 261
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
2
This sentence does not appear in the 9th edition of this text in the section on speech acts (2011:
215–16).
264 Revise: introductory linguistics textbooks
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)
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)
266 Revise: introductory linguistics textbooks
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
Solution: stepping up to the Austinian challenge 267
(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
268 Revise: introductory linguistics textbooks
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
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.
Solution: stepping up to the Austinian challenge 269
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.
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.
270 Revise: introductory linguistics textbooks
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.
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.
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)
272 Revise: introductory linguistics textbooks
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
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).
274 Revise: introductory linguistics textbooks
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,
276 Revise: introductory linguistics textbooks
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).
278 Revise: introductory linguistics textbooks
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.
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.
280 Revise: introductory linguistics textbooks
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
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
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,
MA: Harvard University Press.
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
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,
MA: Harvard University Press.
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
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?
Cognition 95: 201–36.
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
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
298
Name index 299