SOUND SYMBOLISM IN NATURAL LANGUAGES

By

NICKQCCOTOSTO

A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL

OF THE UNIVERSITY OF FLORIDA IN
PARTIAL FULFILLMENT OF THE REQUIREMENTS
FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

1991

DEDICATION

as

dedicate this

Don Tosta and

Tosta.

to

work

to

my Mom,

further dedicate this

Christopher

J.

my

Dad, Donald Ciccotosto, also known

Irene Ciccotosto, also

work

to Carol

Ciccotosto,

support

and

my

as Irene

wife, and

Costoff, her son.

These people have immeasurably enriched

no doubt

known

would not have accomplished

inspiration.

this

my

life

and there

study without their

is

ACKNOWLEDGEMENTS
I

would

like to express

my warmest

gratitude to

my

committee

members, Dr. Linda D. Wolfe, Dr. Christiana M. Leonard, Dr. Robert

Lawless, Dr. Ronald Kephart, and Dr.
all

this

encouraged and graced

with

much

perceptive criticism about

topic.

also

the

me

Norman N. Markel. They have

would

statistics

like to thank Dr.

department

Ronald Randies, chairperson of

the University

at

He gave

of Florida.

timely insight on the use of nonparametrics applied to linguistic

topics.
Illinois

thank

my

friend Dr.

Stanley R. Witkowski at Northern

University for his ever present

staging the entire series of experiments.

the first pilot
Finally,

studies
I

humor and
It

direction

was with

his

in

help that

on sound symbolism were carried out.

would

like to thank

my

mother, Irene, and father,

Donald, for their love and concern over the years. There are no
greater parents in the world.

My

wife,

Carol, deserves

thanks for her kind patience, interest, and concern

special

when

working too many hours on one facet of human existence.

especially want to thank

and Dawn, and

my

my

friends,

brother. Rick, and sisters, Nita, Angel,

Tom

McNulty, Jeff Rosenberg, Greg

McKinney, Brian Akers, and Larry Redman,

support.

was

for their interest and

TABLE OF CONTENTS
Page

ACKNOWLEDGEMENTS

iii

ABSTRACT

iv

CHAPTERS
I

SOUND SYMBOLISM AND BIO-CULTURAL

ANTHROPOLOGY: TESTING PROTO-LANGUAGE
HYPOTHESES IN NATURAL LANGUAGES
Introduction

Sound Symbolism and Proto-language

The Nature of Sound Symbolism
Sound Symbolism Hypotheses
Physiological

I I

10
15
18

Anatomical

29

Semantically Ancient

31

SOUND SYMBOLISM DATA AND ANALYSIS

The Universe of

the Linguistic Data

Coding the Linguistic Data
Hypothesis Testing Using Chi-Square
Hypothesis Testing Using Rank Ordering

I I I

39

39
41

44
63

SOUND SYMBOLISM AND PROSODY, SOUND SYMBOLISM

TERMINOLOGIES, AND SOUND SYMBOLIC EVIDENCE IN
NATURAL LANGUAGES
72
Introduction

Sound Symbolism and Prosody

Sound Symbolic Terminologies

72
74
80

Evidence of Sound Symbolism

Languages
I

in

Natural
106

OTHER SOUND SYMBOLISM EXPERIMENTS

141

Types of Experiments and their Limitations

"Size" Sound Symbolism Experiments
Artificial Lexicons in Sound Symbolism

146

Experiments

141

155

Natural Lexicons in Sound Symbolism

Experiments
"Goodness-of-Fit" Sound Symbolism
Experiments
Synaesthetic Studies into Sound Symbolism
Summary of Sound Symbolic Experiments

188

CONCLUDING REMARKS

191

165

174

180

Summary

191

Theoretical Weaknesses

195

Future Research

197

APPENDICES

A
B

D
E
F

WORD LIST FOR

16 CONCEPTS
SUPPORTING DICTIONARY REFERENCES FOR 16
GLOSSES
CODING PARAMETERS FOR ALL GLOSSES
INITIAL RANKINGS OF FEATURES AND GLOSSES
ACTUAL RANKINGS OF FEATURES AND GLOSSES
PHONETIC CHARACTERS

200
231

252
258
261

263

REFERENCES

265

BIOGRAPHICAL SKETCH

292

Abstract of Dissertation Presented to the Graduate School

of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

SOUND SYMBOLISM IN NATURAL LANGUAGES

By
Nick Ciccotosto

December,

1991

Chairperson: Linda D. Wolfe

Major Department:

major assumption

composing words
early

Anthropology

20th

in

arbitrarily

century

arbitrary

modern

linguistics is that

associate

Saussure's

This

dissertation

casts

doubt

theory by gathering evidence of sound symbolism from

this

virtually

meanings.

sound-meaning tenet has been neither

adequately examined nor challenged.

upon

with

sounds

all

known language

phyla.

Major sound symbolism

experiments are reviewed, and finally, a series of sound symbolism

hypotheses

is

proposed for a group of basic vocabulary words.

These glottochronological words, of a supposed arbitrary sound-

meaning nature,
relationships

are

routinely

among language

utilized

phyla.

by linguists

to

trace

genetic

composed of

Dissertation data are

1%

of 5000 world languages. Sixteen glosses contain 50 words per

meaning from 50 languages, and are taken from

17

sample representing

a lexical

human language

phyla.

The

set includes:

at

10 of the

least

NECK, TOOTH, MOUTH,

NOSE, COUGH, EAT, DRINK, VOMIT, BREAST, SUCK, DOG, SWALLOW,

SPIT,

FOOD, WATER,

of 229

languages,

articulatory

are

CHEW.

and

according to sub-phonemic distinctive

tallied

and acoustic features such as nasal, stop, spirant,

and others.

bilabial,

For the 16 concepts, a

hypothesis

found

These 800 glosses, taken from a pool

at

argues

that

total of

certain

63 hypotheses are proposed. Each

sub-phonemic features are

to

be

higher or lower levels than those in the remaining sample

of 750 words. Chi-square tests run on 63 hypotheses give 23

instances of association at significant levels, p<.05.
the

rank-order median

hypotheses

gives

test

of Kruskal-Wallis

similar results.

Jonckheere-Terpstra

test,

all

to

The application of

the

same

For the ordered alternative

predicted

features

based on three k-

samples are highly significant.

Such synchronically extensive sound symbolism

Sound symbolism, within

the

basic

behavioral

is

striking.

and physiological

meanings of these words, shows a heirarchy of sub-phonemic

features.
facile

Their evolutionary

entry

into

adaptive

value

communication network.

may allow

conspecifics

CHAPTER I
SOUND SYMBOLISM AND BIO-CULTURAL ANTHROPOLOGY:
TESTING PROTO-LANGUAGE HYPOTHESES
IN NATURAL LANGUAGES

Introduction

Sound symbolism,

and motor-acoustic features and meaning,

acoustic

study
light

for

because

anthropologists

an

is

understanding

its

mechanics may render a

humankind

explication of the lexicon possessed by

relates

to

may shed

faculty.
fuller

pre-sapiens

in

examines sound symbolism and argues

This dissertation

times.

important

accurate delineation

its

upon an underlying nature of the human language

Additionally,

it

one-to-one relation between

a nonarbitrary,

that

primitive cognitive levels such as those required of

neonates and early and pr&sapiens society. The crux of this type of

examination

that:

is

"There will always be layers of the vocabulary, representing a

more primitive stage of language in which the

sound and meaning is partly motivated.
there
.

systematic

of this

investigation

supplemented by psycholinguistic
universal
signs."

in

the

In this chapter,

of

hypotheses

vocabulary

about

in

between
need for a

relation
is

various

languages,

order to find out what

of these partly motivated

tests,

expressive function

(Fischer-Jorgensen

in

is

1978:80)

sketch sound symbolism and present a series

motivated

meanings and

their

representations

2
with

nonarbitrary

linguistic

discussed in Chapter

and

II

features.
it

The language data

represents

languages. The data set includes

are

800 words taken from 229

16 semantic categories

and their meanings) which are hypothesized

to

words

(i.e.,

sound

contain

symbolic elements. These words are part of the glottochronological

list

devised by Swadesh (1971) and refer to basic and proto-typical

ethnoanatomical,

My

domains.

physiological,

and culturally

specific

semantic

word sample includes: (ethnoanatomical) BREAST,

TOOTH, NOSE, NECK, MOUTH;

EAT, DRINK, CHEW,

(physiological)

SWALLOW,

COUGH, VOMIT, SUCK,

SPIT; (culturally specific)

WATER,

DOG, FOOD.
The data
categories

may

set

exposes semantically basic words and as such, the

reflect

universally

unmarked domains. That

unmarked domains contain words of

short

is,

form, phonetically

archaic in shape, which are basic in meaning, and which are learned

by

earliest

language

This data set

reasons.

Presently,

speakers
is

(Battistella

1990:23-68).

admittedly minimal, though for a number of

world culture exhibits

at

least

5,000

separate

languages. Given an upward limit on the actual size of a particular

language lexicon, an overestimate would be that any language

contains

more than 1,000,000 words. Even

so,

5,000 languages with

1,000,000 words each, means that 5 billion words are spoken on

earth.

Clearly this demonstrates

everywhere

at

distant

neuro-physiological
call

structures

an expansion of lexicons

time when phonemes,

morphology,

(Hewes 1983).

through a changing

became disentangled from primate

3
Statistically speaking, greater than two-thirds or

70%

of

all

languages contain a phonemic inventory of between 20 and 35

phonemes. Even

the range of

so,

human languages

Phonemic inventories range

Phyla)

phoneme

In turn, each

composed of binary

group,

acoustic features (Sapir

composed of

largely

accepting

the

view

word

(UsLwaiian-Austronesian
1984:7).

a mental construct of a given cultural

distinctive

there

11

articulatory

and motor-

This suggests that languages are

sound-meanings. The impetus for

is

an arbitrary connection

comes from

the

work

composed of sounds and reference

is

association
arbitrary,

between

of the great structural

to a concept.

languages would cease to change (Saussure

1959:67-71).

and completely replace their lexicon approximately every

would
in

(l&-5%
it

1,000 years) (see Gudchinsky

per

be unusual to find more than

1,000,000 items,

command

Still,

scholars

100,000

1964, for example),

million words per

language, this would represent a negligible one tenth of

language's lexicon.

entities

1,000 sound symbolic words

any given lexicon? For the supposed maximal

might

If the

between sound and concept were not predominantly

While languages are endlessly changing bio-cultural

years

all

Saussure (1959). In his groundbreaking work, he held that

linguist

is

arbitrary

signifier and signified

from

Phyla) (Maddieson

1929).

that

size

in

141 (IKung-K hoisan

to

actually produced in

500 (Pullum and Ladusaw 1986).

least

at

is

phonemes

1%

of a

although any language might contain

generally

behavioral

agree that an

average

speaker

and physiological mastery of 10,000-

30,000 words actively (Durbin 1969). For the neonate, child,

mentally handicapped, or the emerging bilingual

can be considerable smaller. Taking the

would mean

realistic

more than

3%

large

latter

speaker the total

figure

sound symbolic system could command

of a language's lexical system. This

may have

been demonstrated for Japanese (Hamano 1986), and

for English

in

Chapter

III.

1978;

Malkiel

this

a matter of

still

some

assumptions

is

that if a

words are compiled from a

and genetically distanced sample of world languages,

the expectation
years, then

is

statistical

basic glottochronological

geographically

argue

1990a).

The importance of these

number of

already

However, exactly how a language's sound

symbolic lexicon should be measured

debate (Ultan

more

as

that,

is

only

1%

more than 100,000

of the terms should be similar. Otherwise, since

contact

and borrowing

parallel

forces

are

not being in contact for

at

is

ruled

out,

internal

and cross-culturally

work. This more reliable intuition means that

sound symbolic words should appear significantly above limits

by glottochronologists in many languages. Further, there
"primitive"

less

so.

Sound symbolism may rank more

creative force in producing "new"

morphological

At

as yet

words.

uncovered levels of cognition and bio-mechanics,

"least

express exceedingly close association

concept.

as

words, than as a label for

sound symbolic processes approach

they

nothing

about a vocabulary rich in sound symbolic words versus

one appearing

aberrant

is

set

moves"

theories,

that

is,

between sound and

Contrary to what Saussure and disciples argued, sound

symbolic words are linguistically pervasive, proto-typical, and

if

time frames must be given,

old.

As LeCron Foster points

phonological

representation

once motivation

meaning

to

is

least

out,

hundreds of thousands of years

the

"arbitrary

relationship

between

meaning becomes questionable

and

discovered for assignment of a particular

a particular phonological

The subconscious

levels

unit"

(LeCron Foster 1978:83).

of language use are yet to be fully

because the extent and importance of sound symbolism

explicated
in

at

world languages. The function of sound symbolism as a citadel of

word-meaning formations

special

speculating

is

not well

studied.

Much

and many poorly designed studies have been done,

to

be sure, and few scholars suggest sound symbolism can expose

primordial words, for fear of reiterating some variation of the

"bow-wow," "sing-song," "ding-dong" language

disdained

Additionally,

theories.

linguists

origin

have omitted sound symbolism as

an arena of attention because of a focus upon sound changes and

the

primacy of words (Jespersen

etymological

Among

the

few

to

propose nonarbitrary sound meanings for

primordial words are

Hewes

1921/1947:410).

Mary LeCron Foster (1978) and Gordon W.

(1983).

So

far,

historically

documented languages

attest

sound

symbolism examples from 12 of the 17 language phyla. There

little

doubt much more evidence of sound symbolism

from the

lesser studied language phyla. Just as easily,

sub-field

emerging

"psycho-semiotics"

to

be labelled

(Markel

"generative

1991) to deal with

is

is

forthcoming

one can see a

phono-semantics"
the

or

under studied

mental

structures

which imbue language

its

affective

use within

dynamic contexts.

socially

Psycholinguists, linguists, and anthropologists have

implemented numerous types of experiments upon

sound

symbolism. Their investigations involve textual analysis and the

psychological testing of differing linguistic groups with the creation
of artifical lexicons and the use of sound symbolic words. This

research

has

never been

incorporated

into

anthropological

about language origins. Below, sound symbolism

this

is

theories

placed back into

context.

Sound Symbolism and Proto-language

The evolutionary advantages of vocal communication

in

primates are considerable. Calls warn others away from danger or

no small observation

toward food.

It

lengthening"

advantages to select individuals capable of their

efficient

is

production

that

they confer "life-

and understanding (Bickerton

most basic tenet of communicative function, when placed

context of

human

bio-social

evolution,

witnesses

This

1990:147).

humans

in

the

as

paragons of communicative efficiency. Humans are the only species

producing a vocal communication allowing themselves defense
outside of real evolutionary time. This

other
senses,

about dangers

is

to

say,

they can warn each

which are unknowable through the immediate

such as cancer and global warming (Pinker and

1990:712).

Bloom

Among

current speculation on language origins

is

the endless

though necessary reiteration that language evolution has had many

causes;

(Washburn

bipedality

restructuring

(Lieberman

1984),

(Falk

1990),

neural-reshuffling

vocal-morphological

1960),

increased

brain

gestural-motor

size

(Jerison

1976),

enhancement

(Ojemann and Mateer 1979), gender differences (Jonas and Jonas

1975), use of fire

(Goudlsbom 1983), increasing face

(Tanner and Zihlman

interaction

1976), and so forth.

however, most language origin arguments

versus punctuated

language

an

is

physiological
referential

scenarios.

"exaptation,"

splinter

to

face

Beyond

into

this,

gradualist

Stephen Jay Gould's school argues

a combination

events coalescing into

of otherwise spurious

remarkably sudden

system (Pinker and Bloom 1990). The classical school of

language origin antedates even Wallace's and Darwin's ideas on the

This school presents evidence of a gradualistic "language

subject.

design" apparent in nature, even

drinking,

breathing,

Lenneberg

1967;

the expense

at

of efficiently eating,

and swallowing (Hockett and Ascher 1964;

Lieberman

1984).

Scholars like to quibble over which selective pressures resulted

in

early

hominids leaving the

forest.

Our

distant

ancestors,

Bickerton argues, used their proto-words most likely

animal imitations, expressive grunts,

(Bickerton

1990:156).

Arising

as

language was adaptable because

intent of proto-words

and

this

was

"to

a
it

in

alarm

calls,

and chance associations

representational

system,

described nature. The only real

get the point across,"

says Bickerton,

echoes Wittgenstein's philosophy of language (C.H.Brown

1976). Wittgenstein

Simply
produce

put,

states,

means

this

sound, there

humans developed

"Whereof one cannot

that
is

where there

is

one

is

Chomsky claimed

not one there.

that

1968).

Instead

The more

of the

(1979), that syntax could have

concert with increasing fine motor control.

in

The primary function of language

as

intrinsically

function

and

to represent nature,

human behaviors

as

this

humanly produced words. The

meanings which words contain are only

of

is

connected to animal communication as a whole,

crucial to the intent of all

is

form

to

This view more closely follows the

Ojeman and Mateer

findings of

developed

innate.

of

"rules"

language being innate, Wittgenstein argues that the capacity

is

to

sudden and apparent "linguistic organ"

Wittgenstein attitude must prevail.

rules of language

silent."

no selection pressure

through the evolving neural tissue (Chomsky

typical

talk,

to

an animal species.

be found within a range

More

basic meanings

may

be inseparable from the sounds composing them because they

consistently

"get

the point

across."

Whether these basic meanings

are called

'flee',

'fight',

'food'

moot

point. This is exactly

is

'mate',

when she derived even more

or 'feed' versus 'run', 'hate',

distant

'love',

or

what LeCron Foster proposed

proto-words from the proto-

words of reconstructed language phyla (1978). She writes:

"Early linguistic symbols (phonemes), apparently parental to all
languages, are reconstructed from a group of languages

present-day

whose genetic

relationship to one another is extremely remote. The

symbols are found to be nonarbitrary. Their
motivation depends upon a gestural iconicity between manner of
articulation and movement or positioning in space which the symbol

reconstructed

represents.

Thus,

the

hypothesis

presented

here

implies

that

early

naming

language was not

in the conventional

sense but

means of another,
displaced in time but similar in spatial relationship" (LeCron Foster

representation of one kind of physical activity by

1978:78).

If a

handful of proto-words or sound-symbols can be

manipulated so as

generate elementary

to

propositions,

language

system can emerge with conspecific vocal partners. The advantage

"THERE"+"FOOD" would

of merely being able to indicate

tremendous
to

relate

rote

is

our early hominid relatives.

to

to

(or

Evidently,

be
capacity

this

name) objects and delay enactment of behavioral

to

well within the range of abilities demonstrated by our closest

and morphologically expressed cousins Pan (Gardner

genetically

and

Gardner 1971), Bonobo (Boehm 1989; Mori 1983), Pongo (Miles

1983), and Gorilla

and Linden 1981).

(Patterson

Bickerton's presentation of proto-language assumes the lexicon

of a

Homo

habilis

Homo

or

to be like a "miser's shoebox,"

erectus

each proto-word containing a meaning according to neccessity's

rankings

(Bickerton

contained

pronouns,

relative-time

particles

including

proto-syntax,

indicating

Proto-language

1990:158).

markers,

location

negators,

quantifiers,

(Bickerton

100,000 years ago

synonymous with

selective

symbolism element
theories fail to

meaning. This

in

are,

modal

auxiliaries,

and

identified for

any human

Without recourse

to

sound

language origin scenario, language origin

show how any sound

is

words,

question

Wittgenstein's views,

in

pressures.

may have

1990:185).

The neccessary semantic concepts

time before

also

is

ever connected to any

an absurdity because in order to be

at

an

10
overwhelming

of arbitrary

level

sound-meaning,

present

the

all

languages had to have undergone immensely long parallel

traditions.

The trouble with

that

a cursory dismissal of sound

to

have

totally

neccessity from their utterances. This

is

order to have arrived at fully arbitrary language now,

in

humans would have had

any

symbolism

is

foregone

all

emotion and

not the case with

clearly

language.

propose that the arbitrary

unreachable end for

all

sound-meaning hypothesis

is

an

languages and that sound symbolism

mechanisms underlie naming processes.

The Nature of Sound Symbolism

Why

should scholars of such differing ages as Socrates,

Aristotle, Plato, Condillac,

(Jakobson and

meaning

in

symbolism
realities

in

Waugh

Swift, Darwin, Wallace, Tylor, and Freud

1979) agree that some facets of words carry

and of themselves? The attractiveness of a sound

is

that

it

provides a bridge between extrinsic and inner

come

hominids. Such plausibility has

vogue. Presently,
holding vital

it

is

into and out of

becoming increasingly important

as

an arena

answers about language origins.

Take the largely autonomic, primate vegetative process of

coughing, as an illustration. Here, coughing

is

neurally at the medulla and

irritation

bronchio-alveolar,

tracheal,

is

initiated

laryngeal,

by
or

a reflex

pharyngeal

integrated

of the

mucosae

auricular nerve

and Bosma.1984). Additionally,

Bertozini,

Bernthal,

(Geoffrey,

can

stimulation

initiate

the coughing reflex

and

it

can be produced voluntarily as a discrete sign, a diagnostic event, or

unconsciously with symbolic meaning (Leith

1977:547). During a

cough, as the glottis closes, strong intrapulmonary pressure builds

and

with the respiratory muscle contractions,

finally,

suddenly opens to release an explosive discharge of

water, and foreign bodies
varies

from animal

the

glottis

mucous,

air,

(Ganong 1983:180). The sound of

some

animal, being species, age, sex, and in

to

cough

manners disease specific (Leith 1977). Nevertheless, the sounds of a

cough

in

all

species take place within a few frequency bands of

acoustic energy, not

or reiterates

process

Any animal who mimics,

of them.

all

duplicates,

cough would create the description of the autonomic

through

nervous

sympathetic

the

system.

There are miles of neural circuitry between the autonomic and

nervous

sympathetic

system,

attractive is just that

say.

In

"gets

it

what makes sound symbolism

but

the point across"

hominid neural evolution,

as Bickerton

points to a "least

it

moves"

pathway inexorably trained upon language development.

symbolism

is

known

to

Sound

provide a "least moves" route in a variety of

ways, the least of which

is

that

it

provides a

peripherally included vocal partners

mnemonic

assist to

such as neonates, other

erect us individuals, or foreign language learners (Wescott

Jakobson and

Waugh

of individual

genotypes and intelligences, and

of symbolic elements,

would

1979). If language

it

certainly

is

to

Homo
1971b,

include a wide range

still

needs mnemonic

incorporate a
assists.

list

12
In contemporary linguistics, there are arguments for "weak"

sound symbolism. That

meaning, say

"size,"

type to represent

finding one peculiar and neccessary

is,

diverse languages will

(Durbin

it

To

1969).

all

date,

utilize

one feature

evidence shows

this

type of a sound symbolism argument only as a general proposition.

Among

the

more

"weak" though universal sound

interesting

symbolism examples include

observation

the

languages the normal declarative order

English,
the
to

"I

Do

It").

universally

terms

appropriate

male/father

for

consonants

labial

consonants to the male ([mama]

vs.

to

1966:76).

the

common

In

regard

female/mother

and

female and apical

[dada]) (Jakobson

1960).

stronger sound symbolism argument supposes that

humans share

(e.g.

This word order represents better than any other

relationships,

social

most

Subject-Verb-Object

is

order of transitive events (Greenberg

actual

for

that

all

pool of semantically and evolutionarily

important events. In this case, the phonological, semantic, or

syntactic

language universals

are

linked

sound symbolism

through

on a language by language basis (Durbin 1969:8). That the front

vowel

represents "smallness" in most language

[i]

sound

semantic-phonological
e.g.).

symbolism
noun

(in

is

clearer

English

[-s]

or

[i]

vowel

example of

a connection
its

an example of a

symbolism("tiny">"teeny,"

Depending upon how the

with syntax.

is

this

is

used,

it

Bob>Bobbie,

might also connect

syntax-phonological

between [FRICATIVE] and

voiced counterpart

[-z]).

a pluralized

Here, the sound

symbolism expresses the concept of "more" with continued sound

13
instead of plosive and brief sound (use of an

[-s]

instead of a [-p],

e.g.).

Since sound symbolism

is

is

probably universal

in

language use,

it

necessary to regard the wider scope of language universals for

Although language universal research focuses upon

comparisons.

the regularities

was ignored

of syntax, phonology, and lexicon, the lexical domain

the

until

late

1960s (Witkowski and Brown

1978).

Since then, implicational universals have been found in folk color

terminology (Berlin and Kay 1969), folk botanical (Berlin 1972; C.H.

Brown

1977), folk zoological life-forms (C.H.

(Witkowski

1972),

ethnoanatomy (McClure

Breedlove, and Raven

(Berlin,

when

apparent

the

1973).

An

Brown

1979), kinship

and ethnobiology

1975),

implicational universal

is

occurrence of an item in widespread languages

implies the occurrence of another item or items, but not vice versa

(Witkowski and Brown

As an
First,

certainly

But,

illustration, an ethnobotanical lexical

no language exists which does not contain

involving the
is

1978:428).

of the

many languages

languages

spoken

by

human evolutionary

at

least

if

least

in order.

one word

cognitive experience.

contain more than one term for plants.

pre-literate

contain thousands of such

read then that

at

is

of a plant. Hence, naming the botanical universe

name

part

scheme

terms.

hunting-gathering

An

implicational

Some

societies

universal might

any languages have two words for botanical items,

one will be a term for

"tree"(e.g.,

large plant). If any

languages have three terms, the third term will be a "grerb," a small
plant relative

to

the

botanical

inventory of a particular

14
whose

environment,

are

parts

Given four

herbaceous.

chiefly

botanical words in a language, the fourth will be either "bush" or

"vine" or "grass"

(Witkowski and Brown 1978:434). One always gets

a term for "tree" before one for "vine", "grass", "grerb" an so on.

Biconditional universals are

known

as well for

human language

Using the semantic-differential approach, Osgood, May,

speakers.

and Miron (1975) found that people use the same qualifying

framework

in

applying connotation or affective meaning to words.

This biconditional
their

universal

implies

that

all

human speakers rank

emotional response to words and their sounds according to

potency

(good/bad),

evaluative

(active/passive)

dimensions.

(strong/weak),

and

For a biconditional

activity

universe,

the

presence of one concept or term will always indicate the other.

With regard

sound symbolism, language universals expose

to

human avenues

ancient

naming behavior. Like

of

symbolic words

of Wescott and Bickerton, sound

according

more

to

basic,

the earliest
primitive,

proto-words

the

may

rank concepts

hominid survival necessities. Hence, the

or universal

word may

be,

the

more sound

symbolism may be influencing emotional evaluations about such

word.

In

other terms,

basic

words may represent the

dimensions, or senses of primary sensory

early

and

activities,

survival value to

language users with sound symbolism. Strictly speaking, early

naming behavior should contain

signifier

and the event

to

close connection

be signified.

between the

15
Sound Symbolism Hypotheses

The
physical

vocalizations of primate communication are dynamic

events.

Their

imploded

include

productions

many complicated muscular and

fricatives,

acoustic

exploded grunts, coos,

screams, cries, hoots, gobbles, songs, clicks, geckers, whines,

growls, barks, pants, laughs, twitters, chirps, and

whistles,

"words."

The varied anatomies capable of such diverse modes of producing

sounds among primates point strongly that evolution selected for
vocalization

effects

in

differing

environments (Waser and Brown

1984).

Among humans,
less

physiological parameters of vocalization are no

complex. Voluntary production of sound requires coordination

of seven of twelve pairs of cranial nerves, seven major paired

muscles groups
midbrain,

in

the

larynx

alone,

widely integrated brainstem,

and cortical areas, and numerous recurrent thoracic and

lumbar nerves and muscles (Chusid

1970).

However, humans produce sound within acoustical physics laws

as

would any other primate. Namely,

stream of

air

is

Human

oral

anatomy consists of three

resonance chambers: the laryngeal, the

frequency and intensity

is

a decrease

oral,

and the nasal. Sound

mainly a function of the vocal folds

located in the glottal region.

air

and condensed

modulated through modification of ventilatory

resonance chambers.

tracheal

a rarified

An

increased muscle elasticity or a

pressure elevation can cause a rise in pitch. Conversely,

in

the vocal

folds elasticity or an

increased tracheal air

16
pressure elevation can cause an increase in intensity (Judson

Weaver

1942:77).

The voluntary
that

and

humans

act of phonation in

is

so extraordinary

accomplished singer can effect over 2,100 variations of pitch

an

by varying the length of the

1967:5).

Additionally,

movements of
astonishing

the

humans

1-1.5

folds

glottal

the

alter

tongue, mandible,

micrometer (Wyke

post-glottal

sound wave by

and velum with

lips,

speed and articulatory proficiency. John F. Kennedy, for

example, held the world record for an articulatory rate of 327 word
per minute in an outburst in a

December of 1961 speech

(McWhirter 1978:48). One can assume the topic was emotionally

loaded.

Although
mechanistically

initiated voluntarily, the act of

speaking

based

is

upon the precise subconscious integration of a large

number of feed-back

reflexes

which constantly adjust the large

numbers of muscles required with any type of phonation (Wyke

1967:3-4).

Three phonatory reflexes derive from mucosal,

and myotatic mechanoreceptors. The

cough

reflex,

first,

presented

above

articular,
in

the

produces occlusive glottal effects. Articular reflexes

occur very rapidly when the glottis

key of middle C, a human

articular reflexes

much slower and

glottis

produce what

is

is

opened and closed. For the

opens 256 times per second. The

called

phylogenetically

older

"phasic tuning."

myotatic

Finally,

mechanoreceptors

produce stretching adjustments, tonic tuning reflexes, allowing a

consistent

frequency emission

(Wyke

1967:13).

17
Considering the
breathing,

many

swallowing, vomiting, coughing,

drinking,

eating,

vegetative requirements of humans,

chewing, sucking, biting, and so on,

nerve combination
vegetative

here

is

existing

would

doubtful every muscle and

exist

(Judson and Weaver

functions

wholly because of such

1942:37).

Of importance

what anatomical, neurological, and physiological differences

distinguish

speech

the

Unfortunately, this
tissue

now

is

it

mechanism from

may never

the

mechanisms.

vegetative

be possible to do considering the soft

nature of the vocal apparatus in primates. Instead,

it

can be

argued that vegetative functions must have been closely connected

to

the

semantic conceptions of hominids and these

earliest

conceptions are

everyday

still

word present

at

psycho-semiotic level,

in

in

present three categories of sixteen words. For each

Table

meaning taken from

at

there

l.a.,

least

are

50 instances of

10 of the world's

this

particular

17 language phyla.

propose about each semantic gloss a number of hypotheses

arguing

manners of
phonetical
are

though

language.

Below,

shall

present,

nonarbitrary,
articulation

transcription

from are presented

are presented

in

though motivational, connection between

or places

of these
in

of articulation and meaning.

800 words and the languages they

Appendix A. Their supporting references

Appendix B. All phonetic characters

these words are presented and defined in

review.)

(The

Appendix F

utilized

in

for easy

Tablel.a.
Testing Glosses and Categories
Physiological

19
Because of

association

the

this,

between highly physiologically

hedonistic activities, such as chewing and swallowing, and socially

expressive ones of emotional value through the face and the mouth

cannot be ignored (Dellow 1976:9).

physiological sound symbolism origin

is

based upon the

assumption that part of the sound-producing mechanism

involved in the activity which
far

to

as

state

that

is

The reason

study of non-primate phonation

"[B]ecause the
a

initially

for the early focus

reinforcement

behavioral

produced

by

prominent

upon

lip

synesthetic

language

in

sounds

is

the

experience:

own speech-organs

most visible of another's speech-organs.

the

and human

are the outermost speech-organs, they are, for

lips

speaker, the most touchable of his

hearer,

closely

named. Wescott (1980b) goes so

speech suggests that labiality was

origins.

is

and, for a

When

the

senses of touch and sight overlap the sense of hearing, they not only
reinforce the

auditory

to

transfer."

latter

an

the evolutionary transition from a nonchannel of preferential information1980b:105)

but ease

auditory

(Wescott

Wescott's attitude
the

gesture-speech

is

origin

nothing less than a reworked version of

of language.

Its

most important

proponents have included Darwin, Wallace, Tylor, Paget, and

Johannesson (Critchley

1967:27-38).

In

one manner or another,

each of these scholars proposed that meaningful gesture and

language arose together in a mutual type of synergism (Hewes

1973).
utilized

Wallace,

in

particular,

held that a wide variety of languages

lip-pointing to express ideas such as

coming and going,

and other, up and down, and inwards and outwards.

self

20
At

the center of gesture-speech origin theories

the

is

assumption that the shape of the physiological components

certain

constituting

sounds (tongue placement,

may

baring, extreme exhalation, etc.)

to

shorthand

provide a

synonymy

other important

origin theory

is

behaviors.

better labelled

Two

symbolism.

sound

constrained

physiologically

assumptions

COUGH,

underlie the following hypotheses: First, that these words,

VOMIT, SUCK, EAT, DRINK,

physiological necessities for

SPIT,
all

Therefore,

behavior.

assume

Cough.
maneuvers

cough

as

when

they

became

became necessary

it

for

these

in

selection.

one member of a larger class of respiratory

acts

gas

muscles may be used

a fluid coupling

as

to

which

other sites in the

for:

1.

Ventilation, including

exchange, panting: thermoregulation,

sniffing:

Sound production, including phonation and singing,

2.

whistling,

snorting,

and Bronx cheer;

including coughing:

3.

upper

airways,

paranasal sinuses, nose,

sniffling:

sneezing:

Moving

material outward or

lower airways, larynx, forced expiration:

lower airways, larynx, clearing throat:

mouth,

are

represented affective arenas of

from the respiratory muscles

olfaction;

inward,

second,

CHEW,

system. This class contains three functions the energy of

the respiratory

breathing:

is

which respired gas

in

transmits energy
respiratory

that,

evolutionary

intense

to

still

and

become words, they became so

physiological processes to

response

SWALLOW,

primates;

semantic entities as words, they

teeth

be sufficiently close in manner

for

The gesture-speech language

protrusion,

lip

hypopharynx,

nose-blowing:
retaining

spitting:

nasopharynx,

secretions

in

the

nose.

21
snuffling:

nasopharynx, nose, paranasal

sinuses

1977:545-

(Leith

546).

Coughing appears rare when an animal possesses good

and

it

the appearance of coughing increasingly

likely

is

diagnostic

If this

While

it

should not be unlikely

COUGH

find

glottals

alternate

alternate

invariant

SICK,

so on.

COUGH

are:

Ho:

back vowels, and

stops, velars,

chance/normal distribution

in

the

sample. The

hypotheses are: ila: stops, velars, back vowels, and

find

glottals

as

could

suggestion has not yet been tested, the null

this

hypotheses for

sound

true,

words and concepts such

also have a polysemic relation to

The

is

most languages the distinctive features naming

HOT, DISEASE, and

became a

groups as they improved upon other

sign to hominid

social integration behaviors.

that in

health,

higher than chance/normal distribution

in

hypotheses suppose that because a cough

autonomic process,

it

provides reference

to

sample.

the

such an

is

itself

through

symbolism.

Vomit. There

are

numerous mechanisms which protect an

animal from ingested toxins. These include, in decreasing order of

temporal effectiveness:

1.

The smell

or taste

of potential foodstuffs

which may be avoided by innate or learned behaviors,

2.

The

detection of toxins by the receptors in the gut followed by a central

reflex

triggering

appropriate

responses;

nausea

to

prevent

further

consumption, inhibition of gastric motility to confine the toxin

to

the

stomach, and vomiting to purge the system of ingested, though not

22
absorped toxin (Davis, Harding, Leslie and Andrews

entirely

1986:66).

Vomiting

is

of great importance in

human

evolution considering

the vagaries of diet and health in a pre-scientific era.

memory and

powerful reinforcer of

Armelagos and Farb remark

behavior for

all

It

primates.

back vowels are noticed

that

is

world

in

languages for foods which can cause nausea (Farb and Armelagos

1980)

It

can be suggested, therefore, that when selection pressures

VOMIT,

developed a word for

features closely related to other

its

words for dangerous food items and visceral sensations, POISON,

ROTTEN, RANCID, ACRID, PUNGENT, NAUSEA, QUEASY,

Emetic responses

to

emotionally charged events also occur and

humans can speak of "sickening

(Ganong 1983:180). Likewise,
the

and so on.

it

sights"

and "nauseating fights"

can be suggested that because of

inflammatory contexts they are found within, taboo words,

especially

derogatory insults, contain features which are

synonymous with VOMIT. Wescott

swear words about

all

manner of

reports that for English, at least,

topics,

include velar and labial

consonants ("kike," "mick," "dyke," "nigger," "bugger," "fucker,"

"wop," "polack," "gook," "mex," "spic," "canuck," "redneck"

(Wescott 197 la: 124). This back and front pattern relates
superficially

Vomiting

least

is

VOMIT.

complex muscular event creating many points of

and noise, so presupposing features universal

examples of

at

with what could be considered fitting sound symbolic

phonetic features naming

stress

e.g.)

VOMIT

is

difficult.

Its

to

the

world's

complexity can be noted in

its

23
sequence of motor actions:

the elevation of the soft palate, 2.

1.

larynx and hyoid drawn forward,

mouth,

closure of glottis,

4.

opening of the cardia,

7.

5.

3.

relaxation of the esophagus, 6.

flaccid relaxation of the stomach, 8.

constriction of the lower end of the stomach,

respiration,

forced inspiration,

10.

waist, clenched fists,

at

The
nasals,

the

12.

VOMIT

are:

at

Ho:

velars,

glottals,

and back vowels find chance/normal distribution

sample. The alternate hypotheses are:

be found

posture,

characteristic

strained face, and so on.

null hypotheses about

stops,

inhibition of normal

9.

sharp contraction of

11.

diaphram and abdominal muscles, and

bent

and opening of the

salivation

Ha:

low frequency because the velum

nasal
is

shut

in

should

features

when

vomiting, so as to prevent vomitus from entering the nasal

passageways. Glottals, velars, and back vowels should be

frequency in the glosses for
crucial

areas of the process.

they imitate

is

it

spitting is

much

The normal person

contains

high

because they correspond

to

Stops should be high frequency because

suddenness and acoustic manner of vomiting.

Though

Spit.
activity,

the

VOMIT

at

like

generally thought of as a voluntary

coughing and

secretes

about 1.5

is

present at birth in neonates.

liters

of saliva per day, which

number of digestive enzymes, provides some measure

anti-bacterial

action,

of

and lubricates and cleans the mouth (Ganong

1983:392).
It

can be assumed that early hominids possessed some degree

of proficiency with spitting, and also put the secretion to important

bio-medical uses. Saliva

is

known

in

early

and present cultures as

24
the

means

to

cause fermentation of various grains for the

human

production of alcoholic drinks. In various

act of

the

can also be a segment of a threat display.

spitting

The bio-mechanisms of SPIT

exception

that the

is

The

airways.

null

much

are

gathered higher in the

liquid globule is usually

Ho:

hypotheses assume:

COUGH. The

like

stops,

fricatives,

dental-

and affricates should have a chance/normal distribution.

alveolars,

The

cultures,

alternate

hypotheses

are:

Ha:

stops,

and affricates should find higher rates

recapitulate the

articulatory

sounds which are made

in

points in

in

the

fricatives,

dental-alveolars,

the distribution.
act

They

of spitting and the

the course of violent

and abrupt

exhalation.
Eat.

Although a great deal

the nervous

system, this

is

of

is

now known about

little

aid in determining

EAT

semantic intent a proto-language word such as

The reason
the

steps

for this is that

even though

foundations

are

largely

would appear

selective pressure
qualities

EAT

what

might contain.

refers to ingesting food,

involved are diverse and complex. Eating involves

chewing, sucking, and swallowing. Each

It

eating centers of

in

turn

a behavior

whose

autonomic.
that

EAT may

announced a need

of foodstuffs

is

have become a word when

to

identify

the

good

or

bad

whose properties were not transparent

sensory detection. Proto-typically,

EAT may

to

any

mark an occasion

where non-poisonous foodstuffs might be ingested.

Of

all

the physiological

mysterious. Exactly what does

words proposed here,

it

refer to?

EAT

is

the most

propose these null

25

Ho:

hypotheses:

dental-alveolars,

fricatives,

should have chance/normal distribution.

fricatives,

dental-alveolars,

stops,

Alternately,

to

the front of the

mouth

propose:

Ha:

and front vowels should have a

higher rate of distribution. The words for

food

and front vowels

stops,

EAT may

refer to getting

(front vowels), the tools

of eating

sounds of chewing food (fricatives), or mechanics

(dental-alveolars),

of glottal closure in swallowing (stops).

Drink. The behavior of drinking

The difference between

two

the

is

is

that

closely related to swallowing.

whereas a normal swallow

occurs in one-thirtieth of a second, drinking can occur for durations

exceeding one second (Fink

drinks,

a liquid

is

1975:109).

Otherwise, when a person

introduced into the oral cavity and the larynx

is

elevated and glottis closes just as with swallowing.

The

should

stops

hypotheses
find
are

null hypotheses are:

be

are:

at

Ho:

velars,

chance/normal distribution. The alternate

Ha:

velars,

stops,

palatals,

higher than chance/normal distribution

elevated

and resonants should

in

sample.

the

Velars

because the manipulation of the velum prevents liquid

from entering the naso-pharynx.

Palatals

sensations of a mouthful of liquid.

glottal

resonants, and

palatals,

shutting.

Resonants

mime

the

represent

the

kinesthetic

Stops indicate the necessary

of the tongue while

action

drinking.

Chew. As mentioned

earlier,

chewing

is

a hedonistic event for

hominids. Evidently, the pattern of mastication

pool of motoneurones in the brainstem and
nature (Lund

is

is

generated by a

not proprioceptive in

1976:145). The ability to gently crack a peanut or

26
crush

tiny

blackberry seed arises from other proprioceptive facial,

and

oro-pharyngeal,

Since chewing

laryngeal
is

motoneurones.

a reflex present at birth, its similarities to

features in the production of speech

fact,

one scholar recently stated that

and

consonants,

have not gone unnoticed. In

in

the production

of vowels

"incorporating noncylcical gestures at specific points in an

ongoing cycle of movements closely resembles the incorporation of

food transport and swallowing movements into the cyclical jaw
movements of chewing, suggesting that the pattern in speech is
taken over from eating, with modifications specific to manipulating
the shape of vocal tract resonators in place of ingesting food"
(Kingston 1990:738-739).

Chewing
food.

is

only one stage in a series of behavioral steps to eat

Not only does chewing involve many

muscles,

it

cranial nerves and

appears that humans chew soft foods more slowly than

hard foods (Lund

1976:146). With these bits of information on

chewing, the following null hypotheses are made: iio: features

found

at

chance/normal

rates,

dental-alveolar,

front

and fricatives. The alternate hypotheses are made:

found
velars,

at

above chance/normal

fricatives.

rates,

vowels,

Ma:

dental-alveolar,

velars,

features

front

vowels,

These hypotheses are made because chewing

involves articulation of the two dental arcades, in the anterior

portion

of the oral cavity, bordered by the velum posteriorly,

and

with sufficient force to cause breaking noises to be routinely heard.

Suck. There
post-natal period

is

for

little

doubt that sucking

primates.

Some

studies

is

crucial in the early

suggest that "sucking

is

27
a

adaptive response that

functionally

nutritive

reinforcement contingencies in

and DeLucia

(Siqueland

lips

feeding

the

situation"

tactile,

muscular, and gustatory stimuli

by triggering a flow of saliva to assist in the

on the nipple. The thrusting and closure of the infantile

seal

and

at first

be influenced by

1969:1145).

may have

child

initiate sucking,

labial

may

gum

pads upon the peri-areolar tissue

is

responsible for

milk removal, and importantly, the true physical sucking

minimal factor

in

is

milk secretion (Dellow 1976:14).

Surprisingly, the effective reinforcement of sucking can be

achieved with a wide variety of stimuli including visual, auditory,

tactile,

olfactory,

other words,

reinforce

and kinetic (Siqueland and DeLucia 1969:1146). In

humming

or rocking

therefore,
direct

be used

to

feeding behavior in a neonate over and above more

autonomic controls of the nervous system.

to

may

an infant

that

sucking reinforcement

in

can be suggested,

It

early

hominids was related

communication with an infant with multi-modal sensory

elements,

the

ultimate purpose

being to train

and exercise effective

motions of the facial and oral musculature.

Sucking behaviors are also an important part of healing
procedures

practiced

cultural areas.

by shamans and doctors

When SUCK was coded

symbolism could have

its

reference.

affricates,

The

null

set

the

limits

in

finally into
to

the

widespread

word form, sound

features

appropriate to

hypotheses are: rlo: palatals, fricatives,

and nasals should find chance/normal distribution

sample. The alternate hypotheses are: rla:

palatals,

nasals.

in

the

28
and affricates should find a higher rate

fricatives,

The

SUCK

glosses.

of sucking creates a negative pressure inside the oral cavity,

act

explaining

mimic

in

the

features

palatal

the sounds

made

in

and affricates

chosen. Fricatives

sucking. Nasals are hypothesized at a

higher rate because an infant can breathe and

suck simultaneously

and the nasally produced consonants may have reinforcing and

calming

qualities.

Swallow. When
collecting

oral

swallow

is

by the voluntary action of

initiated

contents on the tongue and propelling them

backwards into the pharynx, a wave of involuntary contractions of

the pahrynegeal

muscles push the material

at

rate

cm/s into the stomach. Inhibition of breathing and

are

vital

parts

of the swallowing reflex

Swallowing

is

birth

closure

(Ganong 1983:393).

closely corresponds

mother's milk shortly after birth (Dellow

such a major portion of

1976:7).

human experience

that

to

that

of

This behavior

is

even when fasting,

normal human swallows approximately 2,400 times per day

(Ganong

1983:393).

Since
of

glottal

present in utero and the amount of amniotic fluid

swallowed shortly before

the

of about 4

its

SWALLOW

refers to a virtually

autonomic process, parts

sequence could be coded into the phonetic rendition of a word

with sound symbolism. The null hypotheses are:

and glottals should be

alternative

hypotheses

at

are:

Ho:

glides,

velars,

chance/normal distribution. The

Ha:

glides are also

known

as

semi-

vowels since the acoustic energy and articulatory form

splits

and consonant definitions. So, because of the similarity

to

vowel

29
swallowing, glides should be found
distribution.

at

higher than chance/normal

Velars and glottals should also be found

higher rates

at

because the act of swallowing inhibits respiration and closes the

Humans must manipulate

glottis.

both the glottis and velum to

prevent food or water from entering the nasal pharynx or the

trachea.

Anatomical
The
anatomy

in

universal presence of words labelling parts of the

all

languages

strongly

suggests

that

human

ethnoanatomical

terms are members of a proto-lexicon. Which body parts were

named

first

function

response to selection pressures

in

of body

is

a mystery.

One

terms might have been to represent associated

behaviors with specific areas of the anatomy.

might have been

to

Another function

extend self-reflective reference upon the outer

world. Widespread occurence of this type of metaphor

seen in

is

world languages. Such extensions include "mouth of the

river",

"neck

of the woods", "shoulder of the road", "foot of the mountain", and so

on (Lehrer 1974:135). In some languages the more basic body

terms extend to

name even more

specific

bodily locations, such as

"the neck of the hand" for "wrist" and "neck of the leg" for "ankle."

The
physical

basis for sound symbolic

similarity

naming of anatomy

with place of articulation and part so named.

This naming behavior presents a "least moves,"

and activity of an area

to

The

null

allowing

memory

be the same.

Breast. For neonates, the

behavior.

rests in the

human

hypotheses about

breast

BREAST

is

an active area of

are:

Ho:

nasals.

30
bilabials,

occurence
bilabials,

vowels, and

front

the sample.

in

should

stops

The

have chance/normal

alternate hypotheses

Ha:

are:

nasals,

and front vowels should be higher than chance/normal

the sample because they are found in

suckling. Since feeding

is

the

same area most used

in

in

a continuous process, less than

chance/normal distribution of stops should be seen.

Tooth. The properties of human teeth include hardness,

smallness, and presence in the front of the mouth. Wescott argues
that

TOOTH

terms for

also contain dental-alveolar elements

(Wescott 1971:424). With

dental-alveolars,
distribution

the

in

and bilabials

in

teeth

also

Nose. The nose

turbulence.

It

create nasal

made
is

contains

The

assume

this

rate

that

the

alternate

the

third

null

it

place of articualtion.

air

resonance chamber necessary to

the

is

hypotheses

hypotheses are:

teeth,

dichotomy obvious.

prominent structure of the

are:

Ho:

nasals,

Ha:

Bilabials

resonants,

in

the

nasals and resonants

higher than chance/normal rates because the nose

their

than

softness of the lips

and bilabials should find chance/normal distribution

The

Ha:

an anatomical center of unceasing

sounds. Likewise,

face for humans.

chance/normal

Though covering

sample.

the

in

Ho:

are:

should be less than

Bilabials

distribution.

the lips clearly are not teeth.

the

find

sample. The alternate hypotheses are:

chance/normal distribution

versus

should

and stops should be higher

dental-alveolars

chance/normal

stops,

mind, the null hypotheses

in

this

is

should be higher

also
in

sample.

should find
part

of

frequency

31
chance/normal

than

visually

similar to

Neck. Many

distribution

because they represent the

with the

the protruding possible

activities take place in the neck.

nose

lips.

the

is

It

most

obvious source of phonation, coughing, hiccuping, choking,

swallowing, and drinking. The null hypotheses are: lio: velars,

and back vowels should find chance/normal distribution

stops,

the

Ha:

sample. The alternate hypotheses are:

in

stops, and

velars,

back vowels should find higher than chance/normal occurrence

the

sample. These features are the most representative of the more

autonomic processes
than

since

the

in

the neck.

In

addition,

hunter era,

paleolithic

Mouth.

It

is

not so clear what

languages. Though

is

it

many

alveolars,

things.

bilabials,

distribution in

the

must be assumed

MOUTH

it

is

realized

broken.

refers to in

The

its

meaning

is

null hypotheses are:

Ho:

stops, dental-

and velars should find chance/normal

Semantically

Any

Ha:

sample. The alternate hypotheses are:

distribution.

Dental-alveolars,

velars,

circumscribe the mouth and also should be present

chance/normal

many

variable cross-culturally

because they inflate the oral cavity, should be found

chance/normal

the

generally thought of as the cavity after the

and before the neck,

like so

it

humans have

crunching cracking sound a neck makes as

lips

in

Stops,

at

higher than

and

bilabials

at

higher than

rates.

Ancient
"once upon a time" theory about

human language

origins

must include the necessities of finding water, food, and defense

against predators. If sound

symbolism did play

a pivotal

part of the

32
proto-language naming system in early hominids,

number of sensory data

of transforming a

into

it

did

so because

consistent acoustic

Semantically ancient examples of sound symbolism are based

form.

upon the connection between the most distinctive feature

named and

of the object

WATER

example,

be named

with

water.

became

stops

A human

Water.
There

little

is

and

soft

is

proficient in

or

fluid,

metaphor. For

acoustical

a referent

so

it

attribute

would not be expected

cannot live for more than a week without

doubt than the earliest savanna dwellers

finding hidden water as

Ho:

a matter of survival

The

alveolars,

approximants, glides, front vowels, and stops should

hypotheses are:

chance/normal distribution

find

The

alternative

hypotheses

are:

in

Ha:

the

labio-velars,

distinctness

water as a

vowels

mime

drinking

Food.
a

dental-alveolars

It

is

be

oral

gesturing

labio-velars,

higher than

is

all

and stops

WATER.

Both

and are incongruous with

approximants, glides, and front

chance/normal frequency

since

they

behaviors.

hard to imagine what actual food,

semantic universal in world languages. Does

that
it

The

fluid.

should

in

dental-

sample of world languages.

should be less than chance/normal distribution for

represent

it

FOOD

represents as

mean something

merely eaten, and thereby include medicinal herbs? Or does

mean something

appropriate

set

it

dental-alveolars.

necessity.

null

that

that

is

eaten every day and carries an

of preparative

behaviors

about

itself?

Although

could be hypothesized that the taste of a food might determine

it

its

33
name,
taste

it

hard to invent or even imagine any one food that might

is

same

the

millions of genetically

for

Nonetheless,
it

if

variable individuals.

a very sweet food like

honey were

named,

to be

might be named more for the front of the mouth where those

taste

receptors are found, rather than the back of the mouth.

For

example, the English "honey" and Greek "mellis" both contain front

vowels and nasal consonants.

a food

If

were

or used to induce

bitter

vomiting, like the gourd called "kolosinth" by the English, a front

and back consonantal symbolism might be produced (Norwood

1978:9).

FOOD

For

Ho:

are:

vowels should find chance/normal distribution

front

The

alternate

find

that

hypotheses

in general, the null

hypotheses

are:

ila: nasals and

in

Dog.
humans.

identified

It

It

is

FOOD

uncertain

in

when

much

the

the wolf

sample.

the

front vowels

higher rate than chance/normal in the sample.

humans

nasals and

same way

as

should

argue here

BREAST.

was domesticated by early

can be assumed that since the use of

and the

fire

production of lancelate tools, the wolf ceased to be a threat to

human communities.
spread

to

all

Importantly,

continents.

Human

wolves are
cultures

contamination
cultures

member,

in

in

having

to

humans because

they also eat feces and reduce levels of

the

immediate human environment.

they are food,

scientific

humans

almost universally contain

myths concerning wolves. Dogs are important

when domesticated

like

servant and

subject,

and god.

work horse,

pet

In

various

and family

34

A DOG
null

is

most readily identified by the sounds

hypotheses are:

Ho:

velars,

stops,

it

makes. The

back vowels, and

glottals

should find chance/normal distribution in the sample of world

languages for

DOG. The

vowels, glottals, and

distribution

in

stops

the sample.

synonymy with NECK,

Below

alternative hypotheses are:

Ha:

velars,

back

should find higher than chance/normal

The proto-word

for

the place of the bark

are tables l.b., I.e., and l.d.

is

DOG may
near the

have

NECK.

which recapitulate these

unwieldy hypotheses. Each table presents the 16 glosses and the

types

of hypotheses

argued about each. There are 63 predictions

away from an average

feature frequency for

all

16 glosses.

Table l.b.
Glosses and Consonantal Articulation Hypotheses
Features:

35

36

37

38
and discuss more than a dozen synonymous sound

identify

symbolism terms and introduce some order

found scattered
natural

in

the

literature.

Finally,

to

such references

Chapter

III

presents

language examples of sound symbolism for world languages.

symbolism throughout

These are

illustrative

the world,

types of sound symbolism, and functions of sound

of the extent of sound

symbolism.
Chapter IV

critically discusses the

most important sound

symbolism experiments carried out over the past 70 years. The

diversity
results

of these experiments

from Chapter

II.

is

not easily compared

with the

Nevertheless, the concurrence they lend

is

impressive.
Finally, a

summary and concluding remarks

are given in

Chapter V. Weaknesses of the dissertation design are outlined and

promising areas of future research are

listed.

CHAPTER II
SOUND SYMBOLISM DATA AND ANALYSIS

The Universe of

The hypotheses proposed

a

unitary event,

languages.

To

representative
is

necessary.

in

the Linguistic Data

Chapter

human language

Chapter

the depth of the hypotheses outlined in

test

as

I,

sample of the 5,000 languages spoken among humans

When

predicted feature

below chance

regard

though as an entity expressed as over 5,000 regional

testing

any gloss of

assumption becomes apparent. This

significant to

or

pattern

a universal
levels

of sound

domain when

of occurence.

meaning hypothesis holds

is

that

In

sample, one major

this

presence of any

the

that

and meaning becomes

its

frequency

short,

the

above or

falls

arbitrary

sound-

both words and their sounds

should

only find average levels of association regardless of meaning.

The data base

categories include:

consists of 800

monolexemes

The

BREAST, TOOTH, NOSE, NECK, MOUTH; COUGH,

VOMIT, SUCK, EAT, DRINK, CHEW, SWALLOW,

FOOD. Each

for 16 concepts.

SPIT;

WATER, DOG,

contains 50 examples or words, and each word

a different language. For each category of

and

comes from

50 words, no more than

languages come from one of the 17 language phyla considered. So, for

each meaning and

its

50 instances of globally sampled words,

39

at

least

40
10 language phlya of 17 language phyla are represented. The

language phyla considered include:

4 Austronesian,

Austro-Asiatic,
7.

Dravidian,

Amerind,

11.

8.

Indo-Pacific,

9.

5.

1.

Afro-Asiatic, 2.

Eskimo-Aleut,

6.

Niger-Khordofanian,

Australian,

3.

Indo-European,
10.

North

South Amerind, 12. Uralic, 13. Nilo-Saharan, 14.

Khoisan, 15. Austro-Thai, 16. Sino-Tibetan, and 17. Altaic. Language

phyla such as Na-Dene, Paleo-Siberian, Georgian, Basque and others

were excluded from

sources

and

ambiguities

The

demonstrate
is,

because of the lack of representative

list

surrounding

their

phyletic

assignments.

creation of this data base assumes that a balanced sample

of geographically

That

this

or

historically

languages

their

separated

languages

composed of varying

differences should

show

should

structural

apt use of the

components.

"language"

category because, by definition, languages are changing entities

never possessing the exact phoneme usage frequencies or phonetic

inventory.
distinctive

inventories.

present;

This should be so even though they use the same

features

All

recognizing and creating

in

told,

their

phonemic

what Saussure (1959) argues should be

namely, there should be few strong connections between

sounds and meanings,

their

signifiers

and the concepts they signify.

The fragmentary documentation of geographically separated

languages made collection of

all

800 words from 50 languages

impossible. This would have been ideal because a range could have

been obtained for

total

numbers of phonemes present

in

the

sample. Unfortunately, the data set holds words from 229 sampled
languages, with no one language providing more than a total of 16

41
words for

16 concepts. Thus, no single language's phonemic

all

range and sound frequencies could influence decisions very much.

By sampling from 229 languages and phonologies

association

between sounds and the meanings would be impressive.

In point of fact, less than one percent of the

identical

in

all

represented loans
plenty

is

words were

possessed features with others. These words were

from the same language phyla and

there

instead of 50, any

or cognates

it

in

phyletic

boundaries.

different

cultures,

whether they

from a mother language. Clearly,

of distance between

meaning

uncertain

is

sounds used to represent

the

especially

when comparing

across

However, when predicted patterns of sound-

meaning relationships are consistently observed, the arbitrary

sound-meaning

hypothesis

is

not

supported.

Coding the Linguistic Data

Each word
descriptive ways.

in the

sample (N=800) was coded

(The entire

set

of words

is

in a variety of

presented in Appendix

and each specific language's supporting reference

B.) First,

all

phones were

tallied.

A mean word

is

Appendix

in

length for each

EAT

category was found. Interestingly, the shortest word was

phones per word) and the longest was

SWALLOW

(5.2

(3.6

phones per

word). Perhaps the longer average reflects the less cultural and

more autonomic behavior "swallowing".

In

addition,

words contained between 4-5 phones. Below

is

over

90%

table 2. a.:

of

all

42

Table

2. a.

Data Sample Descriptive Tallies

Words:

43
longest three.

might be hypothesized that the longer words

It

represent longer

slower phenomena,

or

for the shorter terms.

would be

It

interesting

to

length and

meaning connection

could be analogous with examples

These conjectures

systems.

because

this

data

is

is

from the standard mean

EAT

and the longest

(3.6)

If

true,

universal

undoubtedly be tested further

will

a large

The standard

So, one standard deviation

1.6.

both the shortest word

contains

easily

(4.4)

languages.

alloprimate communication

in

not significantly different.

deviation of this sample

such a guess by

human language

as

test

new

simply replicating the same size sample with

a

might be true

reverse

the

SWALLOW

(5.2).

Analysis of the data set was done further for a comprehensive

number of

articulatory

consonant or vowel,
Tallying

features.
either:

the

a).

is

is

and acoustic features. Each sound, whether

identified

a binomial decision.

Yes, contain or

maximum number

given feature

is

according to

50, or

b).

its

distinctive

language and

its

word

No, does not contain a feature. Hence,

of words for each

100%. The gloss

category possessing any

COUGH,

for instance, gives

49

out of 50 languages with an obstruent in that meaning. These coding

parameters for vowels included

rounded or unrounded

all

front,

back vowels distinguished by high, middle or low tongue

central,

height.

Consonantal coding was done for the following front to back

places

of articulation:

alveolar,

palatal,

bilabial,

labio-velar,

labio-dental,

velar,

uvular,

interdental,

and

glottal.

dental-

Consonants

were also coded for the following manners of articulation:

fricative,

affricate,

nasal,

glide,

trill,

lateral,

approximants.

stop,

44
and resonant. These six coding tables are given

obstruent,

Appendix C according
Not

glosses.

ethnoanatomical, physiological, and cultural

to

the coding parameters

all

in

were used

in

testing

hypotheses. The vowels, for instance, are tested only according to

whether they are front or back. The extra coding parameters are
scope of the data and for further

available to demonstrate the full

by interested

testing

scholars.

Hypotheses Testing Using Chi-Square

In Chapter

hypothesis

over

63 hypotheses contrasted sound symbolism and

sound-meaning

arbitrary

is

known

5,000

I,

the

null

hypothesis.

languages,

all

It

arbitrary-meaning

argues that in the pantheon of

phones will be randomly represented

meanings. There should be no particular agreement among

all

separate

languages and the sounds

Further,

when

languages,

the

My

interlanguage

nonparametric
is

is

this

The

is

to

sounds.

compared among

should be as small.

synchronic.
century.

statistical

It

takes

No words

tests

words from
represent the

necessary are

because the underlying population distribution

not uniform

(Wynne 1982:330). The 800 word

inventories.

The

little

information

assumptions difficult to

available
test.

of a

data set

229 languages and therefore, 229 distinct phonetic

represents

normality

meanings attached

similarity

800 word data sample

proto-forms of any phyla.

in

of words

single category

languages as they are known

sample

The

relations.

about most makes

45
However, the 800 word sample does represents 229 languages
this

possible phonetic

known

for

human language

in

of the

its

Then, by obtaining frequency counts of categories (words)

entirety.

for

variation

90%

sample probably reaches

and the phonetic range for

variables

qualitative

certain

(distinctive

features),

two-by-two

contingency table or Chi-square can measure significance of any

relationship.
In the 63 Chi-square tests below, a test word, (e.g.

compared

to

all

is,

as a sample,

of 50 examples of a certain feature for

is

features,

is

other words (15 other glosses) according to a

qualitative feature. That

number

COUGH),

compared

which might

to

the

total

total

COUGH
its

might contain a

total

50 languages. This

number of other languages and

750 features

for

750 languages. Chi-

square results from the calculation of Phi shown below (Driver

1966:322-324):

ad-bc

W(a+b) (a+c)

(b-Hd) (c+d)

X=(D2o N

Since the degrees of freedom equalled

1,

skewing. There

is

was applied

for

distribution

the Yates correction

some debate

over whether the Yates correction for continuity

case, the

values very

is

is

recently

necessary.

In

our

so large (800), that applying this correction lowers Chi

little.

The nature of Chi-square only allows non-

46
directional

COUGH
in

associative

predicts

it

either direction

will contain
will

Although the hypothesis about

findings.

result

more STOPS than average, deviation

in

significant Chi-square.

All 63 hypotheses discussed in Chapter

contain the same

predictions:

Hypothesis:

Null

Ho: u=U, (given a word of n=(50) and u occurence of a feature, a

larger sample

(800)-n(50)=(750) and

occurence of a feature

should be similar);

Hypothesis:

Alternate

Ha: u

is

not equal to U.

The
given.

The

test statistic is

significance level sought

hypothesis asks that

meaning

is

Chi-square, and the corrected Yates value

zero,

if

the

true

is

p<.05. At this level, the null

correlation

between a feature and

what would be the probability of obtaining, by an

error of sampling, a value

as

high or higher than that obtained from

observed sample. Since there are repeated

the

results
at a

must be qualified.

.05

If

100

probability level, 5 cases

insignificant by

is

tests

being made, the

were made with Chi-square

would be

chance factors. In the

tests

tests

likely to be significant or

presented below, for 63

hypotheses about 3 cases should be expected to yield results solely

according to chance associations. As the results will show, this error
is

negligible due to the dramatic

are the

number of

significant tests.

16 glosses and the Chi-square tests for each:

Below

47
Table 2.b.l,
Breast

48
Table 2.b.2.

Tooth

49
Table 2.b.3.

Nose

50
Table 2.b.4.

Neck

51
Table 2.b.5.

Vomit

52
Table 2.b.6.

Cough

53
Table 2.b.7.

Mouth

54
Table 2.b.8.

Suck

55
Table 2.b.9.
Eat

56
Table2.b.l0.

Drink

57
Table 2.b.ll,

Chew

Table 2.b.l2.

Swallow

59
Table 2.b.l3.
Spit

60
Table 2.b.l4.

Food

61

62
Table

2.

b.l6

Water

63

WATER

contacts. Finally,

way,

features. In this

would be

It

it

does not tend

much

is

knowledge of

If

for water

stop features
its

from

and those which have had

ice

little

there is a reference to water because of

would the cultures with knowledge of

liquidity,

stop in

ice.

BREAST.

compare terms

interesting to

which are aware of

cultures

like

use stops as naming

to

ice

its

include more

average for their water term? (English contains a

than

water term,

but also a labio-velar /w/).

/t/,

Hypothesis Testing Using Rank Ordering

Since an 800 word sample

type of statistical analysis

is

large and bulky,

is

more than one

useful to bring out significance.

number of ranking nonparametric

hypothesis for social scientists.

are

tests

One

available

to

test

of the most widely used

Kruskal-Wallis one-way analysis of variance by ranks. This

useful

when

are

there

When

test

the
is

test

is

equivalent to the

and equates the Chi-square distribution tables

Another nonparametric
considered here
Terpstra

in

is

null

1990:226).

(Daniel

1954).

the

only two categories and two

populations are given, the Kruskal-Wallis

test

large

more than two categories comparing more

than two populations or samples.

Mann-Whitney

In

test

the

is

for

known
ordered

in

the

test,

to the types of data

literature

as

as

the Jonckheere-

(Terpstra

alternatives

Kruskal-Wallis

a particular direction

test useful

in

1953)

the Chi-square,

(Jonckheere
the deviation

from the null hypothesis cannot be

64
measured (Holander and Wolfe 1973:122). With the JonckheereTerpstra

the

test,

samples drawn. Since

three

hypotheses are ordered and

alternative

this

test

is

at

least

used with three or more

samples of observations, the distinction between one-sided and

two-sided tests
a

is

very powerful

simplified

understanding

nonparametric

alternative

available

results

1990:235).

not maintained (Daniel

of z-score

to

It

is,

therefore,

which creates

test

any researcher with a rudimentary

and normal distribution

statistics

(Odeh

1972:471).
In Chi-square analysis, each of the 16

word categories has a

number of hypotheses. Presumably, each word

mean average

has a

category

is

bilabial,
stop,

dental-alveolar,
nasal,

in

same universe of

palatal,

labio-velar,

glottal,

back vowels, front vowels, glides,

approximants, and resonants. The

tests

the

nonparametric ranking analysis, each word

In

fricative,

affricate,

category (n=50)

ranked against each other according to each of the 15

features;

tested

from the mean average of a larger

different

number of words (N-n=750) drawn from

words (U=800).

as

initial

ranking needed for both

given in Appendix D. The actual rankings are given in

Appendix

E.

Actual rankings average the

and are not merely

ties

between categories

through 16 rankings found

in

the initial

rankings.

Kruskal-Wallis
rank

test.

Any

sums of ranks

null
(that

Testing.

The Kruskal-Wallis

hypothesis formed with

is,

the

sums of

it

test

is

assumes

a medianthat

the k

the ranks in each sample) to be

about equal when adjusted for unequal sample sizes (Daniel

65
1990:227). According to the 63 hypotheses outlined in Chapter

and tested according

that

each Chi-square

to

Chi-square in

shows or

test

fails

show

to

between word and feature frequency.

association

we can

chapter,

this

only say

significant

the case

In

to

follow, the testing feature (e.g. bilabial, velar, et cetera), not

individual

hypotheses

not mean,

the

as

about words

Kruskal-Wallis

considered.

is

can

test

testing

median,

whether the hypotheses,

tell

grouped by feature, are significant or

In

not.

In order to test using the Kruskal-Wallis design, the hypotheses

outlined at the end of Chapter

tables

La.,

l.b.,

Low

Mid, or

in

and

I.e.

must be used. This time, as the

show, each feature

is

predicted to be High,

frequency in each of the 16 glosses. The 63

now become 240

hypotheses

hypotheses,

with

177

the

unstated

Middle or average values considered hypotheses. Further,

this

test,

predicted,

some

of the features have only

Mid and High

while four, bilabials, dental-alveolars,

stops,

in

using

values

and nasals

have three values predicted. Below are the predictions made for 16
glosses and

15 features on

The Kruskal-Wallis
is

measurement

that is

two and three values (k=2, k=3

test statistic is

a weighted

sum

given below. In summary,

of the squares of

deviations of the sums of ranks from the expected

sum of

using reciprocals of sample sizes as weights (Daniel

2

12

^i

1-1 n.

this test statistic

ranks,

1990:227).

H=N(N^.^--3(N^1)
The use of

e.g.).

involves making the null

it

66
hypotheses that given nl, n2, or n3 population comparisons (Hi,

Mid, or Lo samples,
alternate

hypotheses

i.e.),

their

medians

argue the medians

will be identical.

are

frequency medians predicted for

are

Mid

predictions.

the d.f. score

for
test

is

1.

When

my

data

set.

High or

Low

The remaining 177

k=3, the degree of freedom

The significance

from one

different

another in the predicted manners. There are 63

The

tables are the

is

same

2,

as

for k=2,

those used

Chi-square. The table below gives the computed Kruskal-Wallis

statistics:

Table

2.c.

Kru skal-Wallis Results and Significance

67

68
languages,
strong

would have

it

be remarkably obvious to create such a

to

showing.

Jonckheere-Terpstra
Kruskal-Wallis

Testing. While the Chi-square and

measure differences between selected

statistics

test

samples of words or features, neither indicates whether the

difference

ranking

in

is

tests,

one useful

test

the

is

populations are required. In

equal,

inequality
is

in

but

With

Jonckheere-Terpstra

n2 which
test

The advantage of

partial

prior

is

this

information in

null

is

least

at

test,

hypothesis predicts

three

all

an

predicts

lesser or equal to n3. In short, the

Mann-Whitney

is

that

Wilcoxon

or

takes into account the

it

previous

postulated

In the tables listing the hypotheses in

that

many

are

For the alternate hypothesis, nl

one-sided

test

this

hypothesis

alternate

direction.

particular

lesser or equal to

test.

the

the

it,

there

known Jonckheere-

little

Terpstra test for ordered alternatives.

populations

Though

predicted direction

the

ordering.

Chapter

1,

it

can be seen

only bilabial, dental-alveolar, stop, and nasal features contain a

k=3 and qualify

for this type of testing.

hypotheses of the dissertation can be

Additionally,

summed and

hypothesis efficacy can be figured. This type of

score,

which given probability

tables,

elicits

Entering such a table, the p-level desired

score,

is

level can be less than

3,

their sizes are 3,

.05.

the

a grand

test

score of

creates a J-

a significance level.

is

matched with the k-

and the k-score's three or more sample

k-score for bilabial

all

12,

sizes.

and

1.

For instance, the

The

probability

69
The formula
below.

tallies

It

all

giving a score of
in

for obtaining the Jonckheere-Terpstra test

pairwise comparisons from each population,

when one population element

another, and one-half point in the case of a

whether

at

least

given

is

one of the population means

one of the other population means (Daniel

is

tie.

is

greater than that

measures

It

less

than at least

1990:234).

J=XUij
i<j

The k-scores
unusual.

As

for each of the five tests are non-symmetrical and

a result, tables

This

statement.

score into a probablity

devastating.

do not exist which can translate the

When sample

size

is

is

J-

unfortunate, but not

large enough,

the J-score can be

converted quite readily into the standard z-score, which carries a

normal distribution. In the z-score, the mean

variance

given

1.

The formula

to

is

always

and the

convert using the obtained J-score

below.

"

(N^-Z.^jn. )/4

V
This

[N^(2N-F3)-zJ'^jnf(2n. +3)]/72

test is useful

because

hypotheses to predict order

arbitrary

sound-meaning

in

it

relates the ability of the

a data

tenets,

set,

should

which according

not

have

order.

to

is

70
The

scores are given below in Table 2.d. with their significance

levels.

Jonckheere-Terpstra Res

71

The following chapter places these

widespread

results into the context of

sound symbolism examples from world languages.

Given such comparison, the unusually marked

results

of this

chapter appear so only due to lack of structured research into sound

symbolism

phenomena.

CHAPTER III
SOUND SYMBOLISM AND PROSODY, SOUND SYMBOLISM
TERMINOLOGIES AND SOUND SYMBOLIC EVIDENCE IN
NATURAL LANGUAGES

Introduction

Within

They

are

this chapter,

important

the difficulties

shed light upon

consider because they

to

which

three related areas are examined.

arise

when

scholars choose to specialize

domains and forget the overall unity of

research

phenomena.

First,

evidence suggesting sound

encompasses prosody

symbolism

viewed. As a long labelled "supra-

is

segmental" feature of linguistic pattern, prosody

all

linguistic

is

essential

to

languages. Philosophers from Plato to Freud and linguists

from Ben Johnson

to

Roman Jakobson have

held that prosodic

functions are intrinsic to the lineal nature of sound use in

communication purposes. Prosody not only occupies

role

the language

in

children,

it

speakers.
serious

is

Yet,

play during

basic in
until

attention

pivotal

language acquisition for

allowing meaning transfer between

recently,

prosody has received

little

by language scholars.

Many works have

scratched out schemes which place

prosody within a sound symbolic domain or sound symbolism

72

73
within prosodic

Among

conclusions.

Each paradigm reaches vastly different

one.

more notable

the

Fonagy (1979),

include:

La Metaphore en Phonetique Genette (1976), Mimologiques:

.

Voyage en
and

Waugh

Cratylie. Ertel (1969)

Psychophonetik

Jakobson

(1978) The Sound Shape of Language

Wescott

(1980c) Sound and Sense: Linguistic Essays on Phonosemic

Subjects

and Thass-Thienemann (1967) The

Language

Subconscious

Second, prosody

is

vast and

adequately reviewed anywhere.

its

literature has not

Neither has

been

body of

its

knowledge ever been truUy compared with sound symbolism

studies.

So, even though this cannot be

and define a plethora of sound

used without
this

symbolism researchers,
a

much

that

tighter

when

the

here,

will

list

symbolism terms, currently

much agreement among

immense arena claimed by

done

scholars.

recognizing

In

numerous sound

the

propose that prosody

is

grouping of sound symbolism rules.

sub-set of
I

predict

elements of a universal prosody are identified

and codified, they will be indistiguishable from sound

symbolic ones.

sound

Lastly, evidence of

major language phyla

presented.

is

expose sound symbolism

absence
certainly

is

it

due

to

lack

symbolism from 12 of

in

all

the 17

claim research will

known language

phyla.

Its

of published research data, though

appears present in scans of relevant dictionaries.

74
Sound Symbolism and Prosody

The bio-acoustic universe

is

human

sounds, animal calls, and

composed of environmental

speech.

Sounds have always

emotive meanings for humans. Any survey of the

carried

metaphors ascribed and debated about sounds

cultural

Each

languages demonstrates this pervasiveness.

particular

one of these domains

is

described in

numbers of semantically polar

exhaustive

includes

list

bio-accoustically

the

perceived

all

in

cultures with varying

adjectives.

far

from

following contrasting beliefs

sound:

about

sound may be described

and thereby taught to be understood as small or large, dry or

wet, light or dark, lightweight or heavy, fast or slow, hard or
soft,

smooth or rough, weak or strong, sharp or

dull,

female or

male, quiet or loud, angular or round, clear or abstruse, near

or far,

empty

or full, gay or sad, pure or mixed, short or long,

few or many, sweet or sour, even or odd, squat or

low, thin or wide, major or

flat,

and so on (Fonagy 1979). Even

that

high or

tonal or atonal, nervy or calm,

so,

evidence remains anecdotal

any sounds innately evoke emotions.

Though

the acoustic features lending themselves to such

binary description

acceptance
this

tall,

and

it

among
carries

(Jakobson and
refers

to

are

not well

scholars

that

understood, there

is

general

prosody plays a major part in

"sound suggestiveness" and "intrinsic value"

Waugh

1978:198).

In

most definitions, prosody

suprasegmental manipulation of the forms of

75
So defined

utterance.

suprasegmental, the prosodic process

as

takes place on a level which overlies a basic structure, usually

phoneme. Any number of suprasegmentals can be created

the

and labelled prosodic. However, the most commonly cited ones

such that the pitch, loudness, tempo, duration, and

function

rhythm are linked, either innately or voluntarily,

meaning (Barry

connotative

1981:321).

at least four functions. First, the "globally

Prosody has

and tonal pattern direct a hearer's attention and act

rhythmic"
as

to

1981:337). Prosodic tonality and

semantic guides (Barry

tempo modulation

aid

in

dividing

acoustically

inseparable

"connected speech" into semantic units. "Connected speech"

common

to

all

word

of one

languages and involves the ordinary blending

into another.

This

phenomenon

is

witnessed

the difficulty of aurally learning a foreign language,

more

easily

learned

signalling.

For

whether

speaker

sad,

this

or despondent,

elements processed

in

when

is

it

literally.

second prosodic function

is

function,
is

is

known

a person

agitated,

as speaker attitude

hears and discerns

angry, calm,

seductive,

happy,

by voice quality. Though the prosodic

to

achieve this aim can include pitch,

tempo, and loudness, an accurate discernment of speaker

attitude
social

by conspecifics has been shown

context. That

is,

broadly comparable for

to

interact

even though emotional

all

humans, the

traits

states

within
are

used to identify

each are highly malleable to change according to particular

76
Nevertheless, keeping

instance.

mind, for English speakers,

it

social

qualifier in

situation

has been shown that mild anger

produces an increased tempo of speaking, whereas depression

When

produces a decrease (Markel, Bein, and Phillis 1973).

emotions of speakers according

rate

listeners

"harshness",

it

been evident that

has

soft,

or

"softness"

to

empathetic

emotions such as grief and love are expressed through peakpitch profiles.

contempt,

are

harsh, hostile emotions, such as anger and

The

expressed

peak-loudness

through

(Costanzo, Markel, and Costanzo

1969:269).

profiles

Additionally,

length of utterance seems connected to an expression of

Consequent

friendship (Markel

1988).

now doubts

context and prosodic elements

social

synergistically

interact

to

to

these studies,

convey speaker

Third, perceptual focussing

is

no one

attitude.

a function of prosody.

Localization in the tonal accent, determined by pitch

movement, forces

a centralization

upon the type of

information being conveyed (Barry

1981:330).

With

this

prosodic function, for example, most languages utilize high

and/or rising
to

indicate

intonations

to

mark questions and

statements (Bolinger

1964).

the

Otherwise,

converse
speaker

such as an irritated parent might indicate the imperative

command
acts

as

to a child

such as "Get

a double function in

that

in
it

this

house

NOW!"

in

Focussing

determines the

communicatively most important elements within the sense

77
unit and

same time

the

at

links the unit to

context (Barry

its

1981:337).

show

Finally, experiments

with

presented

syntactically

listener's

attention

semantic

switch

follows

when

that

ruptured

the

subjects are

sentences,

binaural

prosody, while the

the

syntacto-

merely caused hesitations and omissions

(Darwin 1975)(Barry 1981). This "guide" function of prosody

is

suspect in the emergence of proto-syntax. This

the

may have been

language scenario, prosody

earliest

to say,

is

in

the

Consequently, conspecific sound meant emotion and

syntax.

meaning emplacement within

context.

social

Certainly,

vocal

pauses marked an upward physiological constraint of vocal

length

utterance and must have played

part

semantic

in

"guidance."
Cross-cultural similarity in the use of the fundamental
or emotion

frequency to convey

affect,

known

and experimental evidence (Ohala

in

1984:2).

anecdotal

Neonates prefer

intention,

their

own

well

is

mother's voice over others

(DeCasper and Fifer 1980). "Baby-talk" or "motherese"

occupies higher and harmonic regions of

consistently

frequency
1987).

and

1964)(Fernald

Perhaps one of the oldest perceptions

proto-language

and

amplitude (Ferguson

may

be that

TONALLY HIGHER

conjecture

is

extended,

MOTHER

is

in

FEMALE

and

any hominid
and

SMALLER

in acoustical production. If this

the

earliest

human

Kuhl

culture

and

language

began

affective

intent.
is

It

emotional

mother-infant

with

secret all

little

activity

with

interaction

mammalian

tonality

communicating

orders communicate

and other prosodic features.

Within humanly conceived sound symbolic words, high tone

tends to be associated with words connoting or denoting small,

diminutive, familiar, near, familiar, near, or narrow, and the

reverse meanings for low tone (Ohala
terms,

for

vowels,

this

means

1984:4).

the front

In

phonemic

vowels represent the

higher frequency versus the back vowels. For consonants, this

means

the

voiceless

ones represent the higher versus the

voiced ones. As shown further, this

testing

an important focus of

symbolism experiments.

sound

in

is

humans, vowels are most easily recognized and are

In

always intonated. Intonation of utterance

is

universal,

if

only

because Nature creates animals of differing shapes and

and possibly intonation

capacities

the

is

most

common

denominator (Bolinger 1964). For example, an evolutionary

pattern

producing, accepting, and perceiving a high front

unrounded
differing

/i/

vowel by a female or male, child or

size and

health

is

borrowing, descent from a

(Ohala

1984:2).

the

of

too widespread to be explained by

common

linguistic

Indeed, Liebermann

"supervowel" because

it

source, or chance

pointed out that this

group of articulatory parameters forming

called

adult,

is

this

identified

intonation

with

be

unerring

79
among

accuracy

(Lieberman

pantheon of cultural groups and actors

1984:158-161).

Intonation

selected

evolutionarily

shows

crucial

is

it

deemed

thusly

is

to

behavior.

It

partly an innate
is

so

and

because evidence

the socialization processes in

alloprimates

by allowing the inherent variability of the individual a place

in

communicative adaptiveness. Over-specialization gets a

genera wiped out and no species can perfectly create high

frequency vowels invariably.

process entailing the use of

sound for communication of affective intents must include a

multitude of constraining factors.

Some

of these include the

health of the animal, a social context, an age of the animal, a

sex for the animal, and an emotional state of the animal.

one of these can

often,

alter the

Any

formation of a vowel intonation.

Too

language or communication schemes assume "once upon

a time"

that

animals created a sonal frequency, and that this

became an auditory frequency. All

without the

slightest

Prosody

is

scheme. Partly,

this,

the

assumption goes,

variability.

not yet a subset of any sound symbolic

this

is

due

to

lack of cross-cultural data on

prosody and the lack of a unifying framework with which to

study sound symbolism. Even

Any two
commonly

so,

all

vowels are intonated.

phrase utterance occurs within a temporal and

iconic scheme. Plus, the use of prosody

is

linked

with intent within a social context, and the use of sound

symbolism

is

connected with clarifying intent within a social

80
context containing shared perceptual routines.

seems absurd

In

any case,

it

argue that when small front vowels indicate

to

semantic "smallness" in a particular culture,

this

be labelled

"sound symbolism," while claiming the use of a high frequency

register,

including the same vowels, and evincing affective

connotations,

belongs for study within prosodic subfield. The

troublesome blur

between sound symbolism mechanics and

prosodic ones belongs in part to faulty logic. Use of sound

symbolic phonetic devices implies a shared cognitive tradition.

This tradition owns functions identical to those of prosody.
Often, sound symbolism

is

treated as if

must only occur

it

within a vacum, something a categorical definition of prosody

could

never sustain.

Sound Symbolic Terminologies

Sound symbolism
including:

"iconic

is

labelled with a swath of terms

symbolism" (Wescott 1971b), "psycho-

morphism" (Markel 1966), "phonosymbolism" (Malkiel

1990a),

"phonetic

"synaeslhesia"

symbolism" (Sapir 1929) (Newman 1933),

(Karkowski, Odbert, and Osgood 1942), "sound-

meaning correlation" (Heise 1966), "onomatopaeition" (Kahlo

1960),

"vocal-gestural

"phememism"

(Foster

symbolism"
1978),

1983),

1930),

"animal talk"

"ideophone symbolism" (Samarin

(Fisher

(Paget

"mimicry" (Bladon

1970),
1977),

(Langdon 1978),

"magical imitation"
"expressiveness"

Fudge

1970),

symbolism" (Wescott

1987).

(Henry

1936;

"holestheme-phonestheme

and

Such colorful nomenclature regards types of sound and

meaning within language mechanics

as

sometimes partially

and entirely motivated. These terms can refer

sound symbolism:

lexical,

syntactic,

to

types of

morphic, psychological,

and phonological. Otherwise, they can appear as combinations

more

of two or

types.

delimit most below.

simple

organization on a expressive scale ranging from minimally to

maximally arbitrary
though

difficult

is

to

construct

cross-culturally,

has been done for a single language elsewhere

it

(Bladon 1977). Even in the case of the least arbitrary,

mimicry, the given defintions are paradoxical. Nevertheless,

comparison, each possesses

communicative intent
a

manner more

meaning

to

certain

semi-inclusive

be interpreted
than

in

purely

functions

among

in

enabling

conspecifics in

arbritary

sound-

units.

Mimicry. Mimicry

is

the least arbitrary

form of language

use and generally the best possible imitation of a particular

sound source by a conspecific (Bladon 1977). Individuals

always vary in their capacity to mimic with vocal dexterity
fluctuating

widely among a speaking groups.

difference exists,

An

however, between imitating a cat using a

high-toned rasping falsetto voice and reporting a

what a
delivery

cat says.

speed,

important

The former can use vocal

staccattoed

presentation,

pitch,

name

for

amplitude,

reduplication,

and so

82
on

(in

below

English, [miauw], [hesss]

i.e.].

The

latter

are described

onomatopes and represent an abbreviated

as

of

recall

an obvious auditory feature of the thing described (in English,

[kaet], [pus]

i.e.).

Mimicry

is

and speech therapists. Consequently,

poets,

linguists,

not easily transcribed orthographically.for

well studied scientifically.

Still,

it

is

not

is

it

extensive in the collective

psyche and oral history of a culture's forms of dramatic

The great

recitation.
activity,

or emotion,

art
is

Evidence abounds

mimicry, whether of human voice,

to

well
that

mimicry capabilities and

known among

primates.

humans possess extraordinary

Widespread communities

talents.

astound the public yearly by hosting pig-calling, eagle-calling,

alligator-calling,

duck-calling,

turkey-calling

or

festivals.

The

only requisite for a person to become a rich and famous

performer in Western soicety

is

an uncanny ability to

duplicate other people's voices and

semantically

One
speaker's

inappropriate

to

that

something which

say

persona's

is

voice.

of the few studies done on this topic reports on a

ability

to

create onomotopoetic

words so

to

describe

auditory

phenomenon. Wissemann (1954) asked subjects

describe

various

to

sounds which included rattling chains,

snapping wood, sploshing water, shattering glass, clanging

bells,

and the

like.

Interestingly

enough, the longer sound did

not necessarily elicit the longer name.

syllables

corresponded

to

the

Instead,

the

number of

number of divisions heard

in

83
the

Syllables

noise.

differentiation

and

created
stress

expressed the sound's

important

highlighted

sonal

dimensions (Brown 1958:116). Abrupt onset of sound, such as

in

snapping, breaking, pounding, and the

named with

usually

like,

a voiceless stop consonant (e.g., [p],

[t],

was

[k]

Gradual onset noises became labelled with fricative

consonants

(e.g.,

[s], [z], [h],

(Brown 1958:117). Further,

Wissemann's subjects agreed upon a

vowel

utilization in

common scheme

labelling colors and sizes.

produced frontally were used

to

refer to

for

Vowels

bright small noises,

low back vowels the reverse (Brown 1958:118).

This study raises the possibly that mimicry or a process
similar

to

echoism underlies naming principles for sensory

experiences.

man

Roger Brown inquired:

created his

first

words

in

"Is

it

accordance with these same

imitative rules and that these rules, being

made

translation

of the

first

possible that primal

"natural"

words easy?" Such an

to

all

men,

earliest

language scenario presents mimicry as only part of a creative

manipulative naming
order,

system

in

dynamic communicative

loaded with changing social needs, for numerous

primate genera. For example, higher rank in early hominid

vocalizations,

observations,

in

comparison with other alloprimate

might have been signalled by greater than

normal use of vocalizations given and received from

conspecifics (Gouzoules, Gouzoules and Marler 1986).

84
Onomatopes. Onomatopes
acoustical

imitations.

As

"words" and not mere

are

"words," they seldom

qualified

unchanging spelling forms and show considerable

possess

difference

dictionary

in

They represent

definitions.

source and are phonemically characterized

sound

speech sounds. For

example, sonogram comparisons could show that the English

voiced

alveolar-palatal

fricative

resembles the sound of a

/z/

bee buzzing. The /z/ and the sounds of the word "buzz" are

phonemes

phoneme

to

and their

/b

lb

I.

If

English. In Yucatecan

in

is

no

/z/

use in an onomatope for the sound a bee makes

is

Mayan

imploded, the feature reversal of the English

make

children

will not contain a 111 if

phonemes

Mayan, there

into those

cross-culturally.

it

is

word

for

what a bee

says,

it

an onomatope. The codification of

"words" for a speaking group varies

Onomatopic production

is

distinct

from

mimicry, though, and languages contain rules for compressing

an imitation of what an animal/process actually emits into a

shared word.

languages

is

regarding

it.

This acoustical compression phenomenon

little

studied and few

statements can be

of

made

"Morpho-phono-symbolics" or similarly, "phonosemantics" are empty jargon.

No

one knows how speakers go

from imitating the bark of a dog, for example,

word

for

its

bark.

To

give

to creating

some examples from

the Indo-

European family, English speakers' dogs can say [wuf],

Germans' [vau], Frenchs' [wal, Icelandics' [gelta], Rumanians'

and

[latral, Croatians' [lajati], Lithuanians' [lotil,

Palis'

[bhussati]. In the Altaic language family, a Turkish dog says

[haul, and a Japanese [war)]. For the Niger-Khordofanic

language Mbukushu a dog says [kudha].

an

Tahitian,

Austronesic language, allows dogs to say [aoa]. North Amerind

languages differ as well for dog barking. In Hopi

Crow [bahuk], Ojibwa

for

[miki],

[waha],

is

it

and Micmac ['psagagwj.

Mon, an Austro-Asiatic language,

a dog's bark

Finally,

Iki^]

is

(Bladon 1977:162; for others see dictionaries in Appendix B).

The common sense adage

round

untrue.

is

may

barks

differ.

that

dogs bark the same world

Even among packs of

Which

same sub-species

the

types of dogs and what area of the

geographic world do the dogs bark in are two variables

influencing onomatopic construction of "bark."

dismisses a tidy

summary

All

this

quickly

of a mechanical dog bark. In short,

simply naming the vocalization of an animal

is

complex

event.
Other onomatopes relate to sounds that a culture
recognizes as emotionally significant. In English these include
"tee-hee,"

Certain
styles,

"boo-hoo," "ugh,"

"tut-tut,"

"no-no" and so on.

onomatopes also have echoic reference

such as "blah-blah,"

"yammer," "stammer,"

"la-dee-dah,"

"babble,"

to

speech

"hem and haw,"

"stutter,"

and so on. Of particular importance

to

this

"mutter,"

"sputter,"

dissertation is a

group of onomatopes regarding vegetative process such as

hiccuping, sneezing, coughing, laughing, and so on. Cross-

86
onomatopic

cultural

similarities

expose the operation of sound

and gestural symbolism. In the experiments following,

"semantic" compression of sound value

is

this

further examined.

It

should be noted that even with the most automatic event, say

coughing, the cross-cultural expressions

some ways, but

identical

non-identical

in

ways.

predictable,

other,

in

are

Synaesthesia labels a subject's connotative

Synaesthesia.

regard for sounds as they associate with unusual senses. In

early

Greece,

(Pecjak

Homer equated

More modern

1970:625).

music, report

colors,

emotions, and sounds

subjects,

in

response to

major chords "wet" and minor "dry" (Karkowski,

Odbert and Osgood 1942). Similarly, Naval submarine

radiomen during World

share

information

called "bright"

2,

in

response to the need to

about sonar recordings, developed

lexicon.

specialized

War

In

this

creative

vocabulary sounds were

"shiny" and "dark". Large objects, explosions, or

were given low frequency phonemes.

processes

When

events

approached the ship, they were called small, bright, and high

(Solomon

1958,1959).

Sapir (1929), discussed at length in Chapter IV, using

nonsense

CVC

consonant),
the vowel,
tests

words

(i.e.

demonstrated
the

words created of consonant-i- vowel+

that

the

more

anteriorly

smaller in relative perceived

size

produced

(1929).

Other

have associated high tones with sharp objects, and low

tones with round objects (Davis

/b/,/p/,/m/,//,/6/

1961).

Bilabial

phonemes

associate with rounded shapes and

(e.g.

87
velar stops (e.g.

/k/yg/./g"/

with angular shapes in English

1935).

(Firth

Synaesthesia experiments are described in detail in the

Compared with sound symbolism, synaesthetic

next chapter.
are

definitions

archaic

conceptions of sense perceptions and

Just as any

neurologist would say there are

sense receptors,
includes

fuzzy because they were formulated upon

sound dynamics.

more than

five

any audiologist would says sound perception

transduction

of mechanical

energy through

air,

bone, chemical, and electrical mediums. Sound lends

synaesthetically with light, touch,

space,

water,

itself

and the like

presumably because of somatosensory overlapping modes of

sensory processing in the brain.

Phonaesthetics.
to sounds.

safe

are

Good

Phonaesthetics label an emotional nature

or bad, hot or cold, fast or slow, dangerous or

varied affective connotations which

types

of sounds

can acquire in orderly fashion within a culture. Examples

include: a.) [-ses] found in words (such as dash, gash, clash,
lash, flash,

etc.)

associates with violence, b.) low

unrounded sound ,/a/,

(in

mud, dud, cud,

an unspecified heaviness and dullness,

a

pejorative

connotation for English

Ideophones operate
"big"

or

in

c.)

e.g.)

mid back

associated with

[sm-]

cluster carries

speakers (Markel

Niger-Khordofanian languages

1966).
to

label

"harmonically ideal", and "thin" or "discordant"

speaking styles (Wescott

1980a;

Samarin

1967;

Sapir

1975).

Phonaesthetic devices vary considerably between

cultures.

Nonetheless,

no comparative studies have been done

upon universal world poetry, song, or recitation

crucial value of an idea of linguistic

trope.

The

"beauty" in any language

Language speakers

directed

is

underestimated.

to

vary their speaking registers from earliest utterances.

each of these registers carries

is

its

own

are

critically

That

of appropriateness

rules

well known. The ability to interact successfully within a

social milieu

tied

is

with knowing the rules of the "pleasant"

speech game (Farb 1974). Perhaps because the rules are so

because they are so subjective, scholars have

or perhaps

fluid

failed

to

develop a scheme appropriate for the study of

phonaesthetics.

Still,

phonaesthetic

devices

are

different

little

from sound symbolic ones. Sounds which are made during

pleasant

Many
on,

activites

become synonymous with pleasantness.

of these include sucking, making love, smacking, and so

and are described

symbolism

in

Linguistic

natural

in

the following

section upon

sound

languages.

icons. Linguistic iconism denotes the use of

sounds as icons, nonarbitrary intentional signs acting as

designations bearing an intrinsic resemblance to the thing

designates
linguistic

(Wescott

1971b:416).

Instances

it

highlighting

iconism in the world's languages include:

a.)

quickness in English, stop consonants convey the iconic

impression of brevity and discontinuity as

between "chirp," "yelp" versus

"chirr,"

in

"yell".

the

The

contrast
rapidity with

which they are made, iconically recapitulates

"quick".

terms of meters per second, they are the fastest

In

sounds humans can

produced

rank as

their

make.;

quietnessvoiceless

b.)

consonants imply inaudibility or a vocal incapacity and are

most effective when coupled with high front vowels

smallness.

Such English exemplars include

"whistle,"

"sizzle,"

process;

in

later

the

c.)

naming event. This

commonality--frequently

also

referent

diminished activity of a

parallels

temporality--later

of suffixing for past tense

when

"hiss,"

"whisper," and "shush". Again, diminished

volume with speech terms

referent

"tick,"

imply

to

importance

is

events

are

reflected

evident in the commonality

morphemes (Greenberg 1964);

used

terms

rises.

are

shorter

d.)

average

than

These short basic terms are

learned earliest by children (Brown

and Witkowski

1982:73).

Such a

list

of linguistic icons

is

hardly complete.

An

exhaustive study of their pervasiveness has not been done. As

a whole,

they

demonstrate that vocal behavior parallels non-

vocal behavior as far as some semantic intents are concerned.

Iconism

is

similar to

abbreviated behavior display.

like

expressions,

Vocal icons. Vocal iconism

it

such,

it

is

very

sound symbolism devices. Like behaviors and

meanings get

Instead,

As

albeit
is

in

greatly reduced

forms.

not strictly linguistic iconism.

refers to the use of gestural

specificity of vocality.

For example, dentality can be a vocal icon. Since

consonantal feature involves articulation

this

with the teeth,

it

90

Many

connotes steady projection of something from a base.

world languages contain names of various projections from

the

earth

or

the

body

utilizing

include Proto-Indo-European

dental

*ed-

"to

consonants.

and *dent- "tooth;"

bite"

Effik -ot "head," eto- "tee;" Mixtec tu- "tail,"

Following

this conjecture

Hockett proposed that the

v]

thuk "horn,"

t'e

and duti- "mountain" (Wescott 1971b:422).

"woods,"

[f,

Instances

rise

about vocal icons and the teeth,

phonemes

of the labiodental

were caused by the advent of agriculture (Hockett

1985:284).

He remarked

that

nascent agriculture centers

phonemes diffused from

these

and represented the

shift

to

the

chewing configuration required of grinding cereals instead of

the

for cutting

scissor-bite required

meat.

Such a

iconic and presumably, the terms for grains of

shift

all

became

types should

overlap significantly with those of teeth, at least as far as

phonemes.

sharing

In

some languages, minimal

minimal semantic
that,"

shifts.

or "six-seven,"

articulatory shifts indicate

For English, instances include

or "four-five". In proto-Semitic (*einay)

and (0ala:0u) "three-four" and (sidGu) and (sab'u)

(Wescott

name

for

body

parts often include just those parts so

have compiled evidence that hundreds of languages

"tooth"

with dental consonants

Similarly, "lip"
are

"six-seven"

1971b:421)

Names
named.

"this-

is

named with

necessarily redundant.

made with

labial consonants.

the

teeth.

Vocal icons

For example, the word for tongue

91
in

all

languages will include movement of the tongue.

would be of

interest

to

is

test

What

through electro-mylography

whether muscles of named anatomical parts invariably

respond when so named.

so,

If

vocal iconism

may be

considered an adjunct to other identified synergistic

languages

(Argyle

1973).

Psycho-morphs.

A Psycho-morph

meaning can be established,

may

not

1966:2).

accompany

clusters

identified

instance,

but,

units

and

/sm-/

speakers

Hamp

example.

for Markel's

appearing culturally

However,

like

so

many speech

of the psycho-morph

awareness.

select

negatively

inherited,

given

for

/gl-/,

can include the

The speaker
understood,

attitude.

regard

the

For

/sm-/

cluster

1960).

The mechanisms

influence

connotative meaning

this

for English

psycho-morph with

English

(Markel and

and

psycho-morphs are not

language specific.

behaviors, the active processes

occur below the normal level of speaker

Unconscious

attitudes

toward

psycho-morphs

speaker selection of appropriate word choice when

competing alternatives (Markel

Psycho-morphs impute

non-morphemic

with cognitive mechanisms not well

associates,

in

"a

occurrences of the unit" (Markel

all

Non-morphemic

phoneme

level

is

of one or more phonemes for which a connotative

unit

the

body

linguistic

1966).
units,

other that at the

of the word, actively disturbing a level of

speaker's cognitive mind.

word

retreival

Within a culture, psycho-

92
morphs demonstrate

self-reflexive

culture's

injected into actual language use. Attitude

iteration

Even

of a speaker.

expressive words,

up

use and use

of attitude. For Markel, the psycho-morph

number of processes expressing

of a

is

processes,

inner psychic world

the

the selection of large groups of vocabulary,

of negative and positive connotation, link

Feelings reiterate use, use reiterates feelings.

findings

recapitulate

ethologist.

views of virtually every

itself,

these

"mentalist"

reality.

Ideophones. Ideophones are

their exact definition

Africanist Clement

linguistically

marginal units,

being a matter of some debate.

Doke

first

described a group of

common

grammatically deviant expressive forms

languages

In

1990).

Animals, including humans, overlay their inner

upon extrinsic

worlds

Bantu

to

and conveying sensory impressions as ideophones

(Doke 1935:118-119). He argued ideophones were

part of speech
their

only one

is

frequency of use in hypertense speakers (Markel

in

is

special

much

lexical

as

separate

an adjective or adverb. Since then,

status

has

been largely dismissed (Wescott

1980a).
Other linguists have added
ideophone.

Samarin reports that

to the
at

growing corpus of the

least

synonymous with ideophony (Samarin

labels

it

twenty-five

1971).

terms

Westermann

Lautbild,"si word that depicts a reaction to sensory

impressions and expresses a feeling

form"(Smithers

1954:73).

in

Linguist Gerard

suitable acoustic

Diffloth

93
characterizes

ideophones as grammatical units which can

by themselves as complete sentences. Their

function

morphemic constituents

are

phonic features (Diffloth

1972).

Ideophones contain unusual sounds, form exceptions

and

the rules of length, tone,

applying to other

and are commonly reduplicated (Smithers

elements,

Two

stress

examples are

illustrative:

a.)

ss,

11,

"guffaw,"

"sluggish,"

"chatter,"

dd,

tt,

to indicate intensity such as in "puff,"

etc.]

1954:83).

intensity--English

ideophones can involve consonantal doubling [mm,

pp,

to

gg,

"yell,"

and "quarrel". Verbs with

voiced consonantal doubling are rare in Old English and as

well as Old Norse. There are six

when

known

in

each language. But

usage increases in Middle English, they are used in

their

words expressing actions, gestures, or movements of a

or vacillating kind, or those that are repeated

sluggish, inert,

(Smithers

ideophony

1954:85);

b.)

sound duplication--another event of

palimphony or sound-repetition. Types abound

is

in English including "pop,"

on.

"crack,"

"plop," "boob," "dud," and so

Disyllabic examples are also well represented in "hot-

head,"

"tid-bit,"

"kick-kack,"

"sad-sack,"

"sing-song,"

"rag-tag,"

and "hobo". Echo-compound words can be seen as well in

"hodge-podge,"
wootsy"(e.g.
nilly,"

"hurly-burly,"

bilabial

series);

"pell-mell,"
"rag-tag,"

and

"super-duper,"

and "ding-a-ling," "chit-chat"(e.g. apical

"hootchy-kootchy"

(Wescott

and

"hurdy-gurdy"(e.g.

1980a:200-202).

"tootsy-

series);

velar

"willy-

and

series)

94
Discussions of ideophony present an interesting dilemma
for scholars.

In

order to define ideophony, an assumed

and homorganically

learned

same place of

(i.e.

for

articulation

same meaning) ascribed form of sound symbolism must be

This

present.

has

symbolism

debate sound
but

never been seriously investigated. Linguists

as

learned

mechanisms can exploit verbs,

its

or

inherited

suffixes,

universal,

infixes,

prefixes,

and other distinctive phonological features as

linguistic

tools

for the speaker.

These

are identical in

tools

scope with ideophonic types. Ideophones might best be

subsumed

sound symbolic systems, instead of distinctly

as

types

different

of sound

Vocal-gesture.
is

the

symbolism.

With vocal-gesture, human

communication

shorthand for complex of bodily gestures. Proponents argue

vocal

apparatus became pre-adapted

language syntax (Hewes

1973).

The

an

as

result

was

ancillary
that

body

sounds

representing events in the real world would be formed in

sonally

produced gestures representing those events. For one

scholar,

vowels are called posture sounds because they

indicate the emotional

state

of the

sounds of movement (Johannesson

as

speaker.

1952:10).

Consonants are

process such

choking would mimic the sound and muscular event of

choking

itself.

and so on
catch',

'to

is

Indeed,

the

velar

sound made by

attached to the primitive meaning

hold in mouth', and

'to

close'

'to

/k/,

eat',

/g/,

/q/,

'to

(Johannesson 1952:18).

95
The term articulatory

Grammonts

(1897),

gesture

discussed by Schuchardt

is

and Jespersen (1918).

(1901),

An

elaborate examination of this form of sound symbolism

found

Paget (1930). His studies of Polynesian, Semitic and

in

many

languages produced

Sinitic

intriguing

word [gar] was considered appropriate

because

word

is

to

examples. The

devour

the verb to

contained a swallowing motion. For the English

it

the sounds

roll,

fit

because they rolled

Sound-gesture paradigms are tautological. Words are

posited

be gestures because gestures must have preceded

to

complex

vocality.

Even

so,

the theory might predict

velarization of sounds describing

a physiological

such as

act

vomiting, where the velum must be closed during the actual

process.

It

could predict labial sounds for processes involved

Much

with sucking, eating, drinking.

argued

in

dissertation,

this

the

like

the

hypotheses

sound gesture hypothesis

argues that certain behavioral routines find analogy in other

behavioral

expressions.

Johannesson continued Paget's work, but arbitrary

reasoning emerged in his articulatory premises
tracks

a large

list

of root

morphemes

for six

when one

distantly related

languages compiled as evidence. For example, he

velar+vowel+/\/ or
for primitive

however,
his

"it

attempts

/r/

argues

described a rolling or curved motion

hominids. Then,

in

the

same paragraph

was therefore very natural

to

first

describe the

that

Homo

insists,

sapiens

surrounding nature also

in

made use

96
of the lips or teeth as starting point instead of the throat or
the

(Johannesson

palate"

There

is

no doubt

1952:15).

found

that gestural value

speech registers presages sound symbolism.

at

or

five

least

more "languages"

alongside vocal language (Argyle

to the

sensible

as

simultaneously

1973).

Wild speech

this

explained.

still

room

for

By

understand.

Gordon Hewes

"mouth-gesture theory

point,

symbolism research

is

to

left

little

is

explaining everything, nothing

remarks on

is

oratory can be likened

vocal-gesture leaves no

understanding because there

argued that

Nevertheless, a theory as

"ballet of the tongue."

intuitively

is

some

exist

wildly articulated and prosaic

certainly

It

in

and sound

leave most of the postulated

transformation from a gestural to a vocal language

(Hewes

unexplained"

Sound-meaning
statistical

another.

correlation.

Correlations occur with the

comparison of one event, incident, or behavior

They

interaction
certain

1973:10).

Evidence of

are not proof of causality.

that

to

correlations present are valid

only so far as

measuring constraints are constant. For linguistics,

statistical

measurement can be

One

dubious

type of measurement scheme designed explicitly to

examine the "meaning of meaning"

differential."

affair.

is

known

Published in the book The

Meaning (Osgood,

Suci,

Tannenbaum

as

the

"semantic

Measurement of

1957), the semantic

97
differential

has been used as the basis of hundreds of social

experiments.

science

Briefly, the semantic differential offers the subject of an

experiment the choice to describe a word with 20 scaled

antonymic adjectives. For example, asked

subject could choose the conceptual

to

describe father a

side of the

scale indicating

"happy" instead of "sad", "hard" instead of "soft", and so on. In

order to investigate every possible pairing of n

scales

it

is

necessary to generate a test consisting of n{n-l)/2 items.

measurement of relationship
agreement

in

The

direction

interaction

"weak-strong,"

is

The

simply the percentage of

of alignment.

among

these 20 concepts ("good-bad,"

and others.) creates clustering by factor

Eight dimensions of meaning analysis are identified

analysis.

and described. These are called evaluative, potency, oriented

activity,

stability,

aggressiveness
evaluation,
studies

tautness,

dimensions.

potency,

novelty,

and

receptivity,

"The three prime factors--

and activity are also identified

in

where the "concepts" are not words or anything

linguistic,

but rather such

things

as

and representational pictures in an

underwater sonar signals

art

gallery"

(Carroll

1959:67).

The authors of
activity,

and potency are pervasive

characterization
attributes

the semantic differential argue evaluation,

because

they

in

adjectival

correspond

to

fundamental

people hold and to the organization of basic

98
and conceptual processes. Evaluation concerns an

perceptual
individual's
it

is

to

the extent

negative or painful and positive or pleasant. Activity

refers
in

approach or avoidance of a stimulus

the

to

necessity

or

nonnecessity of making

movement

regard to the stimulus. Finally, potency suggests the

amount of adjustment needed

(Carroll

question

to

deal

1959:74-75).

number of sound symbolism

scaling

adjectival

with the concept in

studies have used

so subjects rank sounds in words with the

use of the semantic differential. These studies lend insight

into the prelinguistic basis for a certain

classes of linguistic

presumably because such behaviors are founded

behavior,

upon perceptions of reward value, of demands for certain

kinds of adjustment to stimuli and of the transmission of

information about such perceptions (Carroll

1959:75).

Navy

rating the

study involved 342 males in the U.S.

most commonly used English words on

scales.

One
1,000

semantic differential

Ratings were averaged for each visually presented

word. Factor scores of the three dimensions of evaluation,

activity,

and potency were found with regression equations

(Heise 1966:16). The study did find that words in which

certain

phonemes occur tended

to

have attitudinal meanings

and that the attitudinal meanings predicted from their

phonemic content correlate significantly with the actual

attitudinal

meanings (Heise 1966:14). However, the results did

not match those of other tests which involved "artificial"

99
words constructed of separate phonemes. This study

is

discussed in more detail in the next chapter.

In the case of this test, the use of the semantic
differential

only as good as the experimental design.

is

Drawing the
presented

test

list

from a single frequency stratum

not

results

generalizable

words. Since word length

may

the results

words.

inversely related to frequency,

also be specific to relatively

also likely that a

is

It

is

beyond frequently used

short English

phoneme may have more

that

one attitudinal meaning depending upon the influence of

other linguistic events. For example, the stressed
is

/i/

phoneme

seen in more "active" words than the unstressed schwa

(Heise

1966:26).

Sound meaning

correlations are powerful as evidence of

semantic intentions for words,

features.

/a/

They

are

as

sounds,

strong as

phonemes, and

the experimental

which produce them. They also imply

that

designs

sound symbolism

events are causal. Most tests do not follow up such unstated

For

implications.
specific

reason,

sound symbolics are considered

types of sound and meaning associations.

Animal
for

this

talk.

Most languages include

a special

set of

use with animals playing important cultural roles.

Yuman

Cocopa, a
animal talk

and which

is

in

the

turn

language spoken

in

words

In

Arizona and Mexico,

kind of speech humans attribute to animals,

is

used by humans to address animals

(Langdon 1978:10). American English and Yucatec Mayan

100
follow similar themes about animal

talk:

(1)

it

does not

normally function as a means of communication between two

or

more

adult

members

between actors playing the

perhaps in narrative exchanges

roles

community except

of the speech

of mythic animals (Burns

1983), and (2)

it

includes

unique linguistic features which do not occur in the normal

(Chandola

language

1963:203).

Animal commands often imitate

animal.

directed

the vocal repertoire of a

The involvement of mimicry, reduplication of

phonemic segments, unusual phonological vocalizations, and

imperative or denotive intonations lead

some scholars

propose that animal talk marks an important stage

preconditions for the development of complex
preconditions include:

a)

the

imitate environmental

magical behavior and

illusion

that

to

in

Such

language.

tendency for early hominids to

vocalize more than the great apes, b)

to

to

human

ability

and desire

sounds, c) a hominid tendency to

act

a difficult goal

under frustration
is

to

create

the

nearly or already achieved.

Thus, the imitation of the sounds of a desired phenomena

could be one form of magic, d) the tendency of
to

the

many animals

approach on hearings human imitations of their sounds, of

sounds of their own, or other unusual, non-threatening

sounds, and e) the conscious production of such sounds by

hunters

to

attract

prey

(Fischer

1983:313).

Superficially, animal talk sounds have

speaker's

meaning

in the

phonological inventory as functional tools for extra-

101
species

communication. Nonetheless, the logic underlying

animal

talk

an

represents

ancient

interplay

environment and culture, cause and

imploded

made

to

cause a horse to turn and

behavior sequence, but

the soil.

trot

the

is

which

is

a higher order

in

merely causes

initially

Hooves make sharp sounds when

in

An example

(e.g., "chick-chick-chick")

affricate

lateral

effect.

between

striking

it

to

move.

against rocks

buried

Sharp sounds cause most animals to reflexively

move. Another example

is

staccatoed

and sharply intonated

whistle to cause a dog to return. Again, the animal talk

imitates the canine

with

"whine"

call

it

uses to establish contact

conspecifics.

Involving mechanisms of sound symbolism, the words of

animal talk also create a psychological distance between the

animal world and the human world by defining both
according to roles and expectations. Sounds purposely used to

move animals

to

action

move humans. Animal

referential.

command

No

talk

to

fetch the

that

newspaper with a

balance a bank statement.

but not

Everyday speech,

parody, uses animal talk on an expressive

one says "giddy-yap"

fool might be called "giddy",

legs.

from combinations

expressive, not heavily

is

dog can learn

as part of social
level.

different

are

The "wolfs whistle"

to

a person, but a staggering

someone who seems

have four

associates sex on a physical and not

romantic level in miming another species' contact

orders are growled

to

out with

commands

call.

Military

of animal talk under

102
dramatized physical settings. Male fighting identity, closely

modeled
or

after

"sea-bees."

the

men

animal world, makes

"grunts",

"squids",

Physical sports like wrestling, football, and

boxing create staged animal talk names (Raging Bull La Matta,

King Kong Bundy, Boom-Boom Mancini, Bronco Nagurski,

commands

cetera),

(Kill 'em.

Get 'em, Hut-Hut,

et

and the

e.g.),

mythic identity with animal totems (Rams, Bears, Eagles, and

so on).
Identity with the

human

facets of animal origins across a

swatch of mass society occurs utilizing

manner of sound

this

symbolism. Say the sound and an actor becomes as

animal.

of sound and behavior are marketed

Qualities

modeled

for children

competition.

this

Is

as the

it

the

and

source of inspiration for

manner

so different from the

complex language began? While animal

control animals,

if

talk

is

in

which

useful to

metaphor defining a

also serves in a larger

conceptual and cultural boundary for humans. As a specific

type of sound symbolism,

may

or

may

not be the origin for

sound symbolism.

all

Phememism. At
in

it

present,

debate about the earliest historical forms of proto-

lively

languages.

The

earliest

linguistic

phonemes, and are posited

whose genetic relationship

remote.

and

numerous scholars are engaged

Still,

their

reconstructed

as
to

symbols are

parental

to

one another

symbols tend

motivation depends upon

still

considered

modern languages,
is

to

gestural

extremely
be nonarbitrary
iconicity

103
between manner of articulation and a movement or a
positioning in

which

space,

the

language was not naming in the

scholars propose that early

conventional

symbol represents. Here, some

sense but representation

of one kind

activity

by means of another, displaced

spatial

relationship.

in

of physical

time but similar in

Based upon LeCron Foster's work with primordial

language, the

phememe

is

defined as a minimal unit

combining distinctive features of sound and meaning (LeCron

Foster

1978:78).

A phememe

is

taxonomically organized

according to features of a sounds articulation. For example,

sounds involving

lip

movements carry meanings including

peripherality, while sounds articulated with tongue and teeth

interaction

between the alveolar ridge bring forth internality

with their meanings (LeCron Foster 1978:111). This concurs

with results of Chapter

BREAST

nonarbitrary

much

larger

than

connotations.

phememes

phememe

Importantly,

more

as

Phonemes
a result.

are

NOSE,

comparative

intuited

reconstruction

phememes
skills

are the

its

deduced

result of

LeCron

with linguistics, not the result of

of proto-language
(i.e.

might

only a few.

hypotheses proposed

primordial language

structurally

A phoneme

10 or more structural units necessary for

perception, a

Foster's

and the association between

and bilabiality. Compared with phonemes, phememes

carry

contain

II

by Paget (1930).

claims

103

Her

ancient forms

PL) and these words, morphs, or

of

104
reflexes

plausibly existed

argument says

that just

articulatory

feature

completely

distinct

into

or

replace

particluar

configurations.

as

shift

New

phonemic

rather

articulatory

Old English

in

III

50,000 years ago. The phememic

e.g.),

than

transition

substitution

is

an

of a

configuration, (Teutonic

/d/

so semantic transitions eliminate

features

whole meaning

than

rather

meanings, then, like new phonemes, are

both similar and different from those which they replace

(LeCron Foster 1978:86). With an expanded and changed

language arising somewhere after the Mid-Pleistocene, 50 to

75,000 years ago, primordial language's semantic and

enormously

syntactic

base

intricacy.

The phonological base changed,

doing, but

it

increased

in

sophistication
as

it

is

and

forever

no longer expanded because the separation of

meaning from sound meant phonological expansion was

unnecessary (LeCron Foster

1978:86).

Sound symbolism, under

phememic

the

hypothesis,

is

based upon gestural cues. Sound carries meaning insofar as

references

motor sequences within the hearer's mind,

iconically

representative

Conceivably, there
present today in

is

all

of expressed

behavior and

it

activity.

substratum of this cognitive process

languages.

In

addition,

language learning

could be activated with the aid of the partly innate and

acquired

knowledge of

Phonesthemes.

the

phememe.

phonestheme

is

or sound cluster shared by a group of

defined as a

phoneme

words which also have

105
in

common some

element of meaning or function. Generally,

phonesthemes are not found within words which are

etymologically

provenance

is

linked

(Householder

limited

when words

Etymologic

1946:83).
are

traced

proto-forms

to

through hundreds of feature shifts and reflexes. So, a

phonestheme must be regarded more

morph than

a distinct entity.

as

Summary
in

this

wired"

of

This term

form of a psychorarely

is

sound symbolic terms. In

terms each refer in some

set of

way

brief,

used.

the elements

intrinsic

to

"hard

between sound and meaning through

connections

unspecified

as

pathways.

neural

Some

are

merely

structural

suggestions of the avenues language travelled in order to have

expanded

lexicon.

its

Others indicate culturally prescribed

modes of connecting sound with meaning.

facile

or minimal

units

Still,

others identify

containing meaning and sound.

All societies contain language arts, revered and

traditionally

transmitted,

evolutionary past.
is

difficult

tempoed

to

Poetry

of

humankind's distant

as

part

is

ubiquitous in

human

cultures.

It

imagine language not containing rhymes,

insights,

and verbal games. The ultimate sonal,

acoustic, or phonetic

"quanta" of meaning, according to

Jakobson, has never been identified by any scholar. Sound

to

express

manners.

Though

symbolism describes an innate propensity

and

sound together

certainly

elaborated

misunderstood.

in

nonarbitrary

upon

in

meaning

every culture, this propensity

is

The scholars presented above have attempted

106
to

describe the transition from

mandatory, genetically

evolved connection of sound and intentional content

arbitrary

over the

language

expanded geometrically

culture

one million years.

last

Below,

symbolism

human

as

to

consider the factual

present languages.

in

evidence for sound

In

fairness,

all

sound

symbolism requires considerably more research and

unifying framework
still

acts

as

as

mechanism of proto-language, which

support for the

structural

communication known

as

human

super-

language.

Evidence of Sound Symbolism

Of

Natural Languages

in

the 17 major language phyla, published evidence of

sound symbolism for

12 exists. Unfortunately, like the

sound symbolism terminologies presented

discussions

of

above,

evidence

this

least

at

neither

is

comparable from study

to

study.

exhaustive nor systematically

I

present studies of each of

these language phyla and discuss the types of sound

symbolism pointed out by

their

researchers.

should document sound symbolism in

my

However, since

discussion of sound
In

this

regard,

all

Future research

language phyla.

audience in English speaking, the largest

symbolism evidence

will

involve English.

attempt to show that English, like Japanese,

contains a vast sound symbolic system. This system

be

analyzed

sufficiently.

is

yet to

107
There are a variety of reasons for presenting
evidence of sound symbolism here.
diverse kinds of sound

First,

in

this

describing the

symbolic words and concepts used

throughout the world, a clearer picture of the affective intent

of language

seen.

is

Second, the widespread existence of sound

symbolism negates the adage

that

it

is

spurious.

and similarities of languages and

many

cultural

their

relations

their

sound symbolic examples. Language

process.
that

are

differences
to

the

Third, the

physical

world are exposed

is

in

presenting

selective

Sound symbolic words often involve sensory events

both universal and sometimes specific to culture.

focussing upon

how

a certain

environment into word

perception

are

units,

language partitions
important facets of

its

By

sensory

human

highlighted.

Afro-Asiatic. As a language phylum, the Afro-Asiatic

languages number about 250. They are spoken by about 175

million people across Northern Africa, the Middle East, and the

Northwest corner of Central Africa.

Important sub-divisions

include Egyptian, Semitic, Chadic, Berber, Omotic and Cushitic.

Hausa, like

and

utilizes

many African

a tonal

languages, contains ideophones

system. These linguistic events interact

with a variety of reduplication types to create sound symbolic

meanings (Newman 1989:248). For example, the Hausa verb

'to

become dim'

[dusi-dusil.

is

[duseel; the Hausa adverb 'nearly blind'

Other adverbial ideophones have a light-heavy

rhythmic pattern and low-low tone:

'movement with big

is

108
gown': [buyaa-buyaal;

two objects rubbing

(Newman

bustling about':

together':

[hayaa-hayaa];

[kayaa-kayaa], and

so on

1989:251-252).

Austro-Asiatic. This language phylum

numerous language groups spoken

largest

'noise of

of the group

is

comprised of

Southeast Asia. The

in

Mon-Khmer, containing Vietnamese.

is

Other familiar languages include Cambodian, Laotian, Malay,

and Mon.

known from

Expressives are

a variety of Southeast Asian

and Malaysian languages. These are described

in

some

detail

by Diffloth (1979). For the language of White Hmong,

expressives

are

created

by reduplicating the verb, adding a

sentence final intensive particle,

or

adding a post-verbal

morpheme (Pederson 1986:472). White Hmong expressives

are used in

image-rich and flowery language. While used in

normal conversation, they are more frequently utilized

literature

(Pederson

1986:479).

The meaning of

a particular expressive

the specific verb with which

meaning of

the

V+PVE

is

it

is

from

its

Hmong

is

dependent upon

combined, though, the

total

seldom decomposable (Pederson

1986:481). For example, the cluster [pi-]

following White

in

expressives:

nest on the ground', [plig

[plig

used in the

plawg];

'a

bird rising

plog]; 'someone jumping into

the water', [plib

pleb]; 'wood crackling',

popcorn popping

in

a big pan', [plij

is

[plij

ploj];

plej];

'a little

'bullet impact,

109
bamboo

and

bursting',

'heavy raindrops' (Pederson

plooj];

[plij

1986:481).

Hmong

White

clear that the

It is

[pi-]

represents

cluster

suddeness, but also, that the use of expressives entails a

syntactic extension of

symbolism

This

formation

of

is

meaning with

the

use of sound

an event long considered essential to the

proto-syntax

proto-language.

in

Another Austro-Asiatic language showing a type of sound

symbolism
this

of

is

Gta^ or Dideyi, of the South

language, echo-words are used, chiefly

women (Mahapatra

1976:815).

Munda

grouping. In

among

Echo-words

speech

the

formed

are

by duplicating a stem and inserting an alternate

usually

vowel. This lexical class broadly designates thing, manner,

quality, or action

of a general nature in relation to the specific

idea of the base

word (Mahapatra 1976:823). From

semantic

point of view, echo-forms derivable from a single base form

can be classed into four types:

variety,
3.)

2.)

[i-]

forms, indicating diminutive or tender forms,

category, and 4.)

Some

[kesa],

'a

'a

to

or

forms, indicating an inferior

[i-]

original

form (Mahapatra

1976:823-824).

unsurprising examples of echo-words in

[kitir]],

larger and

[bill],

[a-]

compared

include:

forms, indicating gross

forms, indicating variety different from a related

[u/a-]

quality

[a-]

1.)

'a

small and

stronger ghost';

weak

[kisi],

ghost' versus [kitor)],

'a

'a

small piece of cloth' versus

large thick piece of cloth'; [bala],

snack' (Mahapatra

Gta^

'a

main

dish' versus

1976:424). In other uses, echo-

no
indicate a sense of vagueness and

words are formed

to

uncertainty:

'to eat'

like';

[ko],

[cor]],

'to sit'

and [cog-cag-e],

and [ko-ka

ce],

'after

and the

'he will eat

sitting,

etc' (Mahapatra

1976:827). These examples are identical to what Sapir found

in

1929 studies.

his

Austronesian. Members of

this

language phylum stretch

half-way around the world, from islands in the Indian ocean

At

to the far reaches of the Eastern Pacific.

20%

languages are Austronesian. Such diversity

world's
the

least

of the

due

is

to

geographical isolation imposed by island culture.

Javanese has the largest number of speakers, (over 60

the

million),

longest literary tradition (+1200 years),

one of the major

literatures

number of unusual morphic

expressives,

krama

nick-names,

language

is

structures

is

also contains a large

for

plants,

hortatives,

spoken upon the

contains

it

It

words (Uhlenbeck

-courtesy

Malagasy

of Asia.

and

words involving
and

animals,

1950:265).

isle

of Madagascar and as a

an extensive sound

symbolic system.

Bernard-Thierry outlines four major categories of sound

symbolic words:

1.)

cries

describing their actions;

of animals, their names, and verbs

2.)

noises

made from

natural

and noises made from physical properties inherent

objects;

3.)

stuttering,

on;

and

patterns

muttering,

4.)

of peculiar speech registers

forces

in

common

including

sobbing, blabbing, crying, yelling,

and so

physically and emotionally loaded words such as

HI
on

anger,

shaking,

shivering,

excitement, gaiety,

sadness,

and so

1960:241-242).

(Bernard-Thierry

Single syllable sound symbolic words are a rarity in

this

suffixed,

are

sort

Examples of

infixed,

interest include:

hounds' [kinaonaona];

mewing' [meo];
[rehokal;

Most

1960:243).

Malagasy (Bernard-Thierry

or reduplicated.

prefixed,

'dog bark' [vovo];

'tiger roaring'

words of

often,

of

'crying

[kaonkaonal; 'housecat

'death cry of cattle in a slaughterhouse'

'light rain'

[dadadadal; 'heavy

rain'

[dradradradra];

'stomach growling' [goraraika]; 'howling winds' [popopopo];

'baby cries' [jaja]; 'laughing' [hehihehy]; 'giggling' [kikikiky];
'babbling' [bedidedy]; 'fury' [afonafona]; 'heart pounding
fright'

[tepotepo] (Bernard-Thierry 1960).

Dravidian. Dravidian languages are centered

India and claim

language phylum

in

South

175 million speakers. As a whole, this

is

Among

not well studied.

the

more

important languages include Tamil, Malayalam, Kota, Telegu,

and Tulu. Dravidian languages show an extensive system of

sound symbolism. Their forms involve reduplicative morphs

including

identical

repetition,

vowel alternation, consonant

apophany, tonal contrast, and interfixal replacement

various

cases

symbolism

is

(Emeneau 1978:204-205). The extent of sound

such that

it

is

a diagnostic trait defining the

whole of the Indian continent. There

that

many sound symbolic forms

neighboring

in

Indo-Aryan

is

increasing evidence

spread from Dravidian

languages

(Emeneau

1969:274).

to

For Kota, the sound symbolic forms are a basic

CVC

shape

with only a few derivative suffixes. In other ways the roots

may

vowel nasalization,

modified non-systematically with

be

added phoneme length, or a

CV

(Emeneau 1969:275). Some of

instead of
the

more

CVC

pattern

interesting

examples

of sound symbolism in Kota include: 'noise of lamenting'

[dododo];

'lullaby' [jojo]; 'suddenly'

with fatigue' [danak in-]; 'death

bumping

in

[kavakn];

rattle'

[kor kor]; 'noises of

sexual intercourse' [dop dap]; 'heart

furiously' [litk titk]; 'smack lips while eating'

[civk civk];

twitter'

whisper' [gucgucn];

'to

'to

laugh' [gilgiU;

become limp

'to

'to

beating

[mak mak];

talk secretly, in a

pour water' [bodbodnj;

dog

bite'

[labakn] (Emeneau 1969).

Niger-Khordofanian. Languages of

phylum

are found

speakers

share

at

in

the

sub-Saharan Africa. Over 100 million

least

500 hundred languages spread from

Senegal to South Africa. Well

known languages

are Yoruba, Zulu, Xhosa, Kikuyu, Fula, Dyola,

Igbo, Igala,

A
tonal

in

large

this

phylum

Mande, Dogon,

number of Niger-Khordofanian languages

nature and utilize vowel contrast in

are joined,

in

are

sound symbolism.

sound symbolism as well, and when

ideophones

result.

previously and are noted in a large

Among

in

Ewe, and Tiv.

Consonants are used

both

Niger-Khordofanian

Ideophones were discussed

number of languages.

these include Bini, Diola-Fogny, Zulu, Akan, Gbeya, Ijo,

Igbo, and

others

(Doke 1935; Samarin 1967; Samarin 1970;

12

113
Samarin

1971;Wescott

each ideophone

may

1980a;

Wescott

1980d).

Unfortunately,

be defined differently according to each

language.

Though
have created
event there
an

at

are

scholars disagree about ideophonic structure and

25 types to describe

least

notable

amazing tendency

to

similarities.

special

sound symbolic

Ideophone words display

play with the phonological

of the language they are within.

they contain

this

inventory

They display reduplication,

phonological inventory, and they reflect

meanings of a dazzling variety (Samarin 1970:160).

specific

Ideophones are found in large measure with the

narrative register of social

1990:10).
in

the

discourse (Shanks and Velanti

Phonologically, they seem to provide a mirror image

minds of the speakers of the sound of

Nature. Again, as a

mnemonic

that

word

in

bridge, the use of ideophonic

words brings sensations into juxtaposition with meanings

iconic

in

fashion.

Ideophones are not confined

to the

Niger-Khordofanian

phylum. The other African language phyla Afro-Asiatic and

Nilo-Saharan contain ideophones, but

it

may

lacking in the Khoisan languages (Samarin

"Africanized" Creoles

show ideophonic
West African

be that they are

1970:159).

based upon European languages also

structures.

These include: Sierra Leone Krio,

Pidgin, Gullah, Jamaican,

Saramacan, Ndjuka, and

Sranan based upon English; Crioulo based upon Portuguese;

and Haitian based upon French (Samarin 1970; Shanks and

Velanti

1990).

Some examples
tail'

'relatively

[fififi];

moving

creature'

of ideophones include: (Ndjuka) 'shaking

movement with

fast

'action

[hilKl..)];

respect to a slow

of grabbing tightly'

[gwaa(a..)l; 'hyperactive' [fafa]; 'quick pinch or twist'

'limping walk' [kata kata]

(Shanks and Velanti

1990); (Bini)

[ez^iza]; 'way

[ogbes^ggno]; 'bush ghost'

'blabbermouth'

[kuwow];

up high'

[gololo]; 'cowering' [kpijkpijkpij]; 'pipe' [epipd]; 'jingle-jangle'

so begins the

Ijojojo]; 'to follow' [lelel;

sounding like the wind

the trees' [titiliti]

in

Amerind. Paleo-indians

North

tale' [s^iss

first

iss ie];

(Wescott 1980a)

entered the North

American continent anywhere from 35,000 years

years

to

125,000

ago depending upon which of numerous disputed

archeological

sites

Cave, Calico and

peoples which

one recognizes.

its

alluvial fan,

became

Among

these include Sandia

and Del Mar. Regardless, the

North American Indian created

the

hundreds of languages, including Iroquois, Lakota, Navaho,

Hopi, Salish, Kwaikiutl, Winnebago, Menomini, Cree, Cherokee,
Tzeltal, Yucatec, Porno, Algonquian,

Sound symbolism
of

ways

Some

linguistically expressed in a variety

is

many languages

include

specific
ablaut.

in

consonantal

and Blackfoot.

in

the North

Amerind phylum.

symbolism, reduplication, feature-

symbolism, and vowel contrast symbolism or vocalic

In

the

Coeur D'Alene language vowel symbolism

highly developed.

When

word stem contains

the

vowels

is

/i

14

115
made

o/, the subject has been

or

or caused to act, or to assume

by an outside agent. In the case that the stem

a condition

contains /a u u or a/, the thing has a quality or

is

given

in a

condition automatically or without an outside force or agent

(Reichard

1945:49).

These examples

same

in

Coeur D'Alene are suggestive of the

other languages. Also, the

in

Tannebaum (1957) and

their

work of Osgood,

"potency"

Suci, and

and "activity"

dimensions of connotative evaluation of words comes

in

view of

'squeeze, press'
[tcil'];

'scare'

[piy];

'cough up' [teal] versus 'be nauseated'

1945).

vowel contrasts

Coeur D'Alene uses a

series of

the diminutive. These are: /tc/>/ts/;

/c/>/s/.
'just

Examples include:

below

[tsis];

level'

'wait' [catc]

[man] versus 'smear

[latk^'] versus 'cause to jerk' [litk^'j

'jerk'

In addition to using

value,

'be

to indicate

consonant

/tc/>/ts/;

shifts

semantic
to create

/gw/>/w/; and

low, below' [gwiiant] versus

[want]; 'be long'

[tsic]

versus 'be slender'

versus 'be firm, solid' [cats] (Reichard

1945).

Siouan language, Dakota, expresses an interesting

correspondence between

back points of articulation and

increasing intensity in activity (Boas and Deloria

Reduplication

Yucatec,

light

[pay] versus

are: 'milk'

[xat] versus 'fear' [xitj; 'rub'

grease' [min];

(Reichard

Some examples

this.

to

is

found

in

Coeur D'Alene,

1941:17).

Tzeltal,

Cocopa,

Mohawk, Oneida, Seneca, Cayuga, and Algonquian

languages. For

all

of these, reduplication functions to indicate

116
augmentation,

intensification,

Crawford

1963:211;

and and continuation (Berlin

Durbin

1978:20-220;

Cowan

1969;

1972).

Reduplication probably will be recognized in other North

Amerind languages, but

date

to

it

has been poorly studied.

With consonantal symbolism, the substitution of one or

several

specific

consonants with others causes a change in

connotative meaning (Haas

roundish objects' [dicackl

1970:86). For Wiyot, 'two small

>

[dicackl; 'he sings' [loliswii]

'two large roundish objects'

>

'he

hums' [roriswocii]; and so on

(Haas 1970:88). Yurok uses suffixes much like the English

suffix

/-is/,

which means

'in

manner akin

a general

(as in

to'

"seven-ish" to describe time of arrival, or "moppish" to

describe a hairdo).
is

found

in

More streamlined consonantal symbolism

Yurok where

/t/

/c/

>

in 'ashes'

[pancsc]; 'heart of salmon' [tek'^sa^r]

where

/!/

>

/r/

in 'hair' [^lep]

>

>

'heart of

'to

scrape off mud' [se^let]

>

'dust'

human' [cek'^s],

'eyebrow' [^rep].

the changeable consonants occur in the

be replaced:

[pontet]

If

both sets of

same word, both

>

'to

will

whittle wood'

[se^recl (Haas 1970:89).

Consonantal symbolism

is

seen in

many North Amerind

languages including Yuman, Iroquois, Yucatec, Hupa, Yurok,

Karok, Wiyot, Yokuts, Nez Perce, Miwok, Lower Chinnok,
Salish,

Wishram, and others (Nichols 1971; Mithun 1982;

Gamble 1975; Haas 1970; Sapir

1911).

Expressives, as defined by Fudge (1970), are also present

in

many North Amerind

languages. For

Mohawk, Oneida,

117
Cayuga and Seneca:
'crow

'buzz' [isi:]; 'hello' [kwe:];

More

[ka;ka^].

call'

Seneca except

[krokrok]. The same

Mohawk;

Seneca, and

interestingly, /r/ does not occur in

expressives and taboo words. For instance,

in

'croak of a frog'

'fat

is

true for /I/ in

(Cayuga). Labials occur

A
Amerind

(Seneca) (Mithun

1982:50-53).

Marshal Durbin's "Sound symbolism

is

the

in

Mayan

He demonstrates Yucatecan Maya

sound symbolic language, and one suspects other

Mayan languages may be

signify

[bo^ks]

far-reaching study of sound symbolism in North

language family" (1969).

richly

[kluklukluk]

in expressives as well: 'pow'

[phlo^ts]

'plop'

Cayuga,

legs slapping together' [blaets]

(Seneca); 'croak' [mbldo] (Cayuga); 'glug glug'

(Mohawk);

'oh dear!' [ake];

as

well.

of physical

plasticity

For Yucatec, palatal features

properties,

alveolars

signify

breaks in direction, glottals refer to completion of events,

labials

indicate

long,

narrow, and round things, high tones

low tones are used for

incorporate states and qualities, and

(Durbin

nick-names

1969:19).

Durbin also compares dozens of roots from the amply

documented Yucatecan Mayan language

or

are

because they question

arbitrary

rife

semantic

The

at

the

Durbin

results

assumptions of

sound-meaning connections, usually found

disclaimers
textbooks.

their

cognates in Proto-Indo-European.

counterparts
striking

to

as

beginning of most introductory linguistic

states

that

these...

118
"examples illustrate the fact that in many cases where the
English lexicon derives from the same PIE root we can also
expect the semantic counterparts in Yucatecan Maya to be
phonological similar to each other. This indicates that the
same historical processes found in Indo-European languages
resemble those in Yucatec Maya, a not very surprising fact.
But it also indicates that the cognitive processes (i.e. the
selection and placement of semantic features for a given
object or event) are comparable for the two languages. For
example, there is no linguistic reason why [drying] should be
associated with [flat open place] in both languages" (Durbin
1969:46).

English and Yucatec could not have been in contact

any time earlier than

span of time

least

at

separating

at

100,000 years ago. With such a

any semantic

these languages,

similarities

should be mystifying, but are instead instructive

one wishes

to

examine sound symbolism as a normal

mechanism through which human senses become coded

sound

into

values.

Some

of the

more

interesting

examples include:

(English) vs. 'shiver with cold' (Yucatec) [kuyl;

abundance' (PIE) [*pell vs.

[*bheld-]
[*sner],

vs. 'to throw'

'to

'fat'

(Yucatec) [poll;

(Yucatec) [pull;

'to

vs.

'to

wind,

twist'

(PIE)

vs. 'to dry' (PIE) [*ters],

vs. 'slimy'

cook'

'toast'

to toast' (Yucatec)

(English); 'put something

(Yucatec) [nos| vs. 'nibble' (English);

[nil]

'to

cook, char' (English); 'crooked'

(Yucatec) [koyl vs. 'crooked' (PIE) [*ger];

(Yucatec)

throw' (PIE)

'to

snare' (English) vs. 'to stretch a rope' (Yucatec)

(Yucatec) [c'aak-|

to the lips'

'cold'

'to fill,

[sin-]; 'sand' (English) vs. 'sand' (Yucatec) [sa'am];

[pookl

if

(PIE)

[*lei];

'to

'to

slip'

slide'

press out liquids'

119
(Yucatec) [pic]

[peh]

vs.

'to

mash' (English);

vs. 'position, place' (English), 'off,

away' (PIE) [*apol; 'dog'

[*kwon];

vs. 'dog' (PIE)

(Yucatec) [peek']

place, put' (Yucatec)

'to

'dust resulting

from

sawing' (Yucatec) [ma'ay] vs. 'saw' (English).

These comparisons cannot be used

to reconstruct genetic

or phonological origins, but they are important in building a

human

semantic base for the reconstruction of

language.

both

In

Indo-European and North Amerind phyla,

proto-words are extended

Yucatec the word

'root'

also

essence,

moving](Durbin
South

For example,

similar manners.

in

is

associated with

closing up, folding, curling] and in English

[abstract,

proto-

contracting,

is

[base,

in

shrink,

associated with

drawing, dragging,

1969:47).

Amerind. The phyla encompassing the South

American continent has not been well

studied.

Hundreds of

languages exist south of Meso-America and few have been

described in any detail. Better
are

known languages

of this phyla

Quechua, Aymara, Tupi, Guarani, Lenca, Huitoto,

Amahuaca,

Jivaro,

Arawak, Mayoruna, and Siriono.

Sound symbolism has been

Estero Quechua. Reduplication
feature refers

to

actions

is

identified in Santiago de

seen in

many words and

this

done intensely or excessively

(DeReuse 1989:61). This language exhibits meaning inside the

phonological feature for various
contradicts

transformations

semantic domains and this

identified

proto-Quechua. The palatal feature

in

as

it

emerged from

Santiago del Estero

120
Quechua

[s],

refers to the diminutive in a variety of suffixes

(DeReuse 1989:58). Whenever

or [-lu],

[-sapa]

the

>

/s/

>

/x/ correspond

word stem appears with

has the more augmentative

[-sapa]

meaning (DeReuse 1989:59).

/s/

to

Finally,

consonantal contrasts of

increasing degrees of stench

(DeReuse 1989:62). De Reuse suggests

parallels

other North

to

Amerind languages

in

contrast

which back point

corresponds to intensity of phenomenon.

of articulation

Nevertheless,

that this

it

is

one instance of negative visceral precursors

and hence negative connotations, taking value

vomiting,

sounds produced
Apalai

is

in

in

the back of the oral cavity.

an Amazonian language that contains

ideophones. These particles of meaning and sound act as a

finite

verb form. They can be used as a direct object, a

separate

sentence,

reduplicated up to

an

infinite

verb substitute, and can be

Koehn 1986:124).

10 times (Koehn and

Finno-Ugric. This language phylum

Northwestern and Northern Asia.

Its

is

best

located in

known languages

are

Hungarian, Finnish, and Estonian. These three languages

are

spoken by about 22 million people.

Samoyedic languages are included

known

of which are Lapp,

in

A number

this

of

group, the best

Saam, Yurak, Ostyak-Samoyed, and

Nenets.

Few

of the Finno-Ugric languages have been studied with

any thoroughness by any but Finnish

the

linguists.

However,

second largest language, Finnish, sound symbolism

in

121
appears

in

for a large verb class involved with affective and

phonesthetic

concepts (Austerlitz

1967:26).

sound symbolism as a mechanism, per

Finnish

rules

of

morpheme

se,

The presence of
violates

normal

the

distribution.

Affective vocabulary was examined in Finnish by Antilla

(1975).

He found

a variety

of phonological instances where

names of people. The

stop could be found in proper

/-ari/

a
is

and infix /-sku-/ appear

variety
also

in

an

increasing

of affective terms, including

noted.

regularities

Antilla remarks

of affective

that

nouns act

verbs.

larger patterns (Antilla

other Urgic

1975:18).

double
suffix

frequency in

Vowel shortening

these phonological
as

phonaesthemes, growing

from minor coincidental identifications between

to

semantic properties of nouns were amplified.

the

He

few words

notes as well that

languages show similar unifying phenomena.

Austro-Tai. The Austro-Tai languages are located

in

South

East Asia and the more well studied languages include Tai,

Black Tai, White Tai, Siamese, Lao, Lue, Phuthai, and Phuan.

Approximately 37 million people speak dialects of Tai.

Thai uses an extensive variety of reduplications to
express words and meanings

sound symbolically. Three major

forms of Tai reduplication include reduplication of the base,

ablauting reduplication,

tone

(Hudak

and reduplication with a change of

1990:767)

In the first, reduplication of a base serves to soften a

quality,

'good'

[dii]

vs.

'rather

good'

[diidii];

intensify meaning.

122
'to

be

'child'

true'

[cir]]

[dek]

vs.

[cigcig];

'really true'

vs. 'children'

and indicate plurality,

[dekdek]

For the second, 'vowel/consonant oppositional pairing,'

reduplication

can form a qualitative

Examples include:
with

its

a.)

or

vowel

the pairing of a back rounded

~

corresponding front unrounded, /u

[mubm'ib];

'mutteringly'

meaning.

quantitative

'wrinkled,

[rjuarjia];

'sleepy'

i/, in

/o~ e/

[ju'^jr^]; 'in tatters, in shreds' [kra^rur|^kra^rir|''];

'leaning to one side or the other' [jo-'ji-^];

mussed'

'scanty'

in

(as in

foliage) [r6r|rer|]; 'limpingly' [p'logp'leg]; /o ~ e/ in 'tottering,

[gon^gEn^]; 'stammeringly'

wobbly' [to^te''];

'not firm, unsteady'

[?5??e

pairing of any vowel with /a/, /i ~ a/ in 'very

far

?];

b.) the

away' [lib lab];

'stridently'

[w'idwad]; /e

a/

in 'radiant,

glowing with health' [plegplagl; 'gangling' [keg^kar)^]; /

~

in 'doubtingly' [nr|nar|]; /i

/a

a/

in 'clumsily' [t's^t'a''];

/o

a/

in 'scatteringly'

'sadly, lamentingly'

The

/u

a/

in 'roaringly' [su'^sa-^];

[mogmar]] (Haas

finally /o ~

1942:

In

speech.

tone in the

first

in

sound symbolic reduplication found

addition,

than man's

a/

2).

Tai involving reduplication with a tonal change

for emphasis.

it

is

Generally,

a/

a/ in 'mumblingly' [p'ip'am];

[proypray]; and

final type of

used more

in

is

in

used often

woman's speech

such words carry the changed

syllable and this tone

longer than the normal tone (Hudak

include 'good' [dii] vs. 'really good' [dT

is

higher in pitch and

1990:767).
idii];

Examples

'forward, bold' (of a

123
woman) [kra^de^kra^de];
'flickering'

[wabwab]

'whisperingly' [krasl

bkra^sab]; and

(Haas 1942:3).

Thai has numerous other, though

rarer,

examples of

sound symbolic reduplication. These are important

notice

to

because in largesse they are semantically identical with

English meanings for the same activities, nouns, or qualities.

Some

of them include: 'beating of a drum' [tumtum] (Tai)

vs.'rum-pa-pum' (English);

'trivially'

[y

Imyrm](Tai)

vs.

'so-so'

(English); 'aimlessly' [k'wer|^k'war|^l(Tai) vs. 'willy-nilly'

(English); 'jokingly'
'hearty laughter'

[p'lump'lam]; 'awkwardly' [rjum^gam^];

[heha];

'bright

'much' [ys^yg^]; and 'rippingly'

In

hodge-podge,

include:

ding-dong,

patter,

slipslop,

dally,

zigzag,

pocus,

to Tai,

[c'

igc'sg] (Haas

Haas remarks

but uses both consonantal

similar,

tonk,

comparison

and smiling' [yem^yim^l;

that English is very

and vowel ablaut. Examples

mamby-pamby,

shilly-shally,

wishy-washy, tick-tock, chit-chat,

sing-song,

criss-cross,

mish-mash,

fuzzy-wuzzy, bigwig, hubbub,

razzle-dazzle,

1942:2).

humpty-dumpty,

fuddy-duddy,

hurly-burly,

and

others

dilly-

riff-raff,

helter-skelter,

humdrum, hobnob,

pitter-

honkyhocus-

(Haas

1942:5).

Sino-Tibetan. More than

Tibetan languages.

Its

billion

people speak Sino-

major languages include Mandarin,

Lahu, Cantonese, Burmese, Classical Newari, Hakka, Fu-Chow,

Tibetan, Miao-Yao, Kachin, Bodo, Min,
the

Wu, and Garo. Despite

numbers of people speaking these languages,

little

work

124
has been done on sound symbolism, again, oustide works by

Chinese linguists.

One paper
Tibetan

describes a comparison between various Sino-

languages and Uto-Aztecan

Surprisingly,
virtually

number of

identical

basic

languages (Shafer

1964).

vocabulary items were

between various languages of each phylum.

This should not happen unless for chance or borrowing

reasons.
items,

the correspondences

Still,

are

and were chance mechanism

at

for

basic

vocabulary

work, any vocabulary

should have the same likelihood of being similar. For these

reasons an

underlying

parallel

sound symbolic system for

each can be suggested

Though Shafer explains

stationary

the similarities as archaic and

words of a macro-phylum (Sino-Aztecoid),

more reasonable explanation would be

words represent semantically
language. Sound symbolism

is

are worth a closer look: 'cloud'

(Old Bodish);

'rain'

(Papago)

pd] (Shoshone)

vs. 'father'

(Nahuatl) vs. 'mouth'

(Dandezongka);

propose that these

concepts for proto-

[namu] (Yaqui)
vs. 'rain'

vs. 'wind' [*li-]

vs.

'cloud'

(Karenic); 'father' ['a'-

[a-p'a] (Garo); 'mouth' [kamatl]

[-kaml

'buttock'

'belly'

[nam]

[yu] (Kukish);

(Rai); 'breast' [pipi]

(Cahita) vs.

(Newari); 'elbow' [tsiku] (Huichol) vs.

[tsOik] (Burmish);

think a

suspected for each. The words

[yuu] (Papago)

'wind' [hwe-li]

'nipple' [pipil]

vital

to

[kup-tca]

[wo'k] (Papago)

vs. 'backside'
vs. 'belly'

'joint'

[kup]

[vok]

(Banpara); 'stomach' [p6-no] (Hopi) vs. 'stomach' [po] (Kukish);

125
[to-

'belly, breast'

(Cahita) vs. 'belly' [Ito-ba] (Old Bodish);

(Cahita) vs. [si-] (Yano); 'hear'

'urine' [sisi]

vs. 'ear'

ma]

[*krna] (Sino-Tibetan), 'deaf [naka-p] (Heve);

[tyo-tum] (Papago)

vs. 'bear'

[puwe-tsi] (South Fork)

spit'

vs.

[luhi-] (North Fork) vs.

(Papago)

vs. 'to call' [k'o]

[ka]

vs. 'sing, song'

'think,

[na-ka-] (Comanche)
'bear'

[ta-hum] (Taying); 'mouse'

'bamboo

[*bwi] (Burmish);

rat'

[t'u]

'spittle'

(Karenic);

(Middle Burmese);

(Mandarin);

'sing'

'call'

'to

[ko]

[ka'a] (Ute)

know' [mati] (Nahuatl)

'see,

vs.

[hmat] (Middle Burmese) (Shafer 1964:104-

consider'

105).
Sino-Tibetan and Uto-Aztecan languages separated more

100,000 years

than

hypothesis

The

ago.

seems absurd here. Chance and borrowing

To make

hypotheses must also be ruled out.

more

forcefully,

encoding of

in

or

'this'

widespread languages. This

['li-]

(Papago),

['r'i]

'this',

(Hopi), [T

and Sino-Tibetan languages show:

(West Bodish),
Altaic.
to

it

[i-]

(Mandarin),

[r

The Altaic languages

i]

[i]

is

discussed more

holds that

concepts be represented with

vowels and the vice versa. For

show:

point even

meaning which was expressed with

length in Chapter IV, but briefly,

'here'

this

Sapir and Jespersen initially pointed out an

'near-far'

vowel ablaut
at

sound-meaning

arbitrary

'near'

or

small front

Uto-Aztecan languages
(Yaqui), [ivi] (Cahuilla);

(West Himalayish),

['i]

(Sgaw) (Shafer 1964:105).

stretch

from Western Turkey

Outer Mongolian and Japan. Well known Altaic languages

include Japanese, Turkish,

Mongolian, Azerbaijani, Kurdish,

126
Korean, Manchu, Ainu, Yakuts and Sibo. As a whole, the Altaic

may

languages

sound symbolism as a diagnostic

take

trait.

Recent work on Japanese shows an extensive spread of sound

symbolic markers which will be discussed further below

(Hamano

Korean

1986).

is

reported to have thousands of

sound symbolic roots (Martin 1962). Finally,

the

have noticed,

at

very least, hundreds of sound symbolic words in both

Kurdish and Manchu dictionaries in the course of acquiring

test

words.

Hamano's study (1986) of Japanese exposes

a wealth of

sound symbolic phontactic, syntactic, and idiomatic markers.

For a native speaker, the language of 'giongo' or 'mimetic'

words and

'gitaigo'

childhood.

However,

did

or 'modal' words
until

not even remotely

sound

Hamano's

is

apparent from

dissertation,

suspect the extensiveness

most scholars
of Japanese

symbolism.
is

It

sound symbolic words and roots

entirely possible that

make up more

than

3%

of the

average Japanese speaker's

vocabulary. The 'giongo' dictionary of Asano (1978)

than

(Hamano

1,450 words

reasons. First,

it

shows

1986:3). This

is

striking

a major language can contain

lists

for

more

two

a massive

sound symbolic system that can be overlooked for decades by

scholars.
act as

Second,

a powerful

it

demonstrates that sound symbolism can

generator of meanings within a given

language rather than serving as an "archaic" and quasistationary

or

static

linguistic

remnant.

127
Japanese contains two main sound symbolic structures.

The

first

The second

actions and sounds.

is

is

concerned with

a class of 'mimetic nominal

and deals with qualities more often than actions

adjectives'

(Hamano 1986:32)
are

and

a class of 'mimetic adverbs,'

is

seems

It

that

more

meanings

'synaesthetic'

present in the second class. These structures have at least

of sound symbolic expressions. These include:

five types

forms,'

which

consonant

CVCV

are

sound symbolic words beginning with the

consonantal doubling;

/p-/;

'p-

CVC

groups;

CVCV-

groups; and finally, irregular forms of sound symbolism

(Hamano 1986:13-31).

Additionally,

vowels express sound

symbolism.

Some

interesting

'crunchy' [pori-pori]

examples
'surprised'

are: 'crisp'

[pari-paril;

[bikkuri (to suru)];

[gikuu (to suru)]; 'enraged' [kaa (to suru)] (Here,

the verb

'to

'tight'

(da/desu)];

'shaky'

[gyuu-gyuu (da/desu)];

The nominal

[da/desu]

are

[to suru] is

which the mimetic adverb modifies.);

do'

'splendid' [riipa

'startled'

[gata-gata (da/desu)];

'spirally'

adjectives here are

[kuru-kuru (da/desu)]

CVCV-CVCV. The

added and act as the verb

'it

copulas

or 'the

is'

way

it

is'

or 'just so'.)

Other types include monosyllabic,

absent verb marker: 'wailing'

contentment' [wii
'very excitedly'

to];

[ween

CVC,

to];

with a small or

'drunken sighs of

'small bell or insect sound' [riin

[ru'n-run]; also

'hi'

[yooj;

'a call'

riin to];

[yaa];

bisyllabic, 'meagerly, exacting' [kookiri]; 'exactly' [kaakiri];

128
'vividly'

[kuukiri];

wooden

clogs'

[kookuri]; and

'tasting rich'

others, 'sound of

[ka'ra-koro]; 'small object knocking about

in a

box" [ka'ta-koto]; 'sound of dry leaves' [ka'sa-kosol; 'sound of

trains'

[ga'ta-goto]; 'being

[de'ko-boko]; 'being
suka];

flustered'

'toiling' [e'etira,

[mu'sya-kusya];

sullen'

[do'gi-magil;

'with

bumps'

[zya'ka-

'noisily'

(Hamano 1986:28-32).

o'otira]

Japanese also exhibits a system of consonantal

symbolism, which

is

paralleled

noted earlier (Indo-European,

e.g.).

Some

'explosion,

Japanese involve:
'hitting

of a surface,

agreement'
puffing

emission

out,

contact,

'opening,

/t/;

movement'

/k/;

friction'

haziness'

coming

/y/;

'softness,
/s/;

and

sound-meaning connections

in

/p,b/;

into close contact, complete

up,

inside,

haziness,

fluid

expanding,

swelling,

surfacing=inward/outward
/w/;

faintness'

many

'sounds from

'rolling,

decisiveness'

breaking,

breaking

from

language phyla

North Amerind, and Dravidian,

nonarbitrary

of the

diverse

in

sources,

movement'

/r/

'soft

childishness,

(Hamano

1986:226).
It

can be readily assumed that such an impressive sound

symbolic vocabulary as possessed by Japanese, also must

contain

some words with which

hypotheses

This

is

the case: 'dog'

to

test

my numerous

[wa'n-wan];

'fox or the act of

coughing' [ko'n-kon]('physiological and culturally specific

homonymity'

e.g.);

'tummy, or the sound made rapping the

stomach' [po'n-pon]
describing

its

('ethnoanatomical' e.g.);

sounds and flashes'

'A-bomb

blast,

[pi'ka-don]('culturally

129
specific' e.g.); 'breast'

'baby

rattle'

pulu];

'dry,

[boi'nlitnune]

('ethnoanatomical'

[gara-garal('baby-talk'

e.g.);

rough place on the

[kasa-kasa]

('ethnoanatomical'

e.g.)

skin'

e.g.);

[putu-

'small dots'

(Hamano 1986:49)

Indo-European. The Indo-European languages are

probably the most well described and analyzed of
phyla.

Most of

language

all

these languages need no introduction:

English,

French, Latin, Pali, Persian, German, Polish, Spanish,

Portuguese, Hindi, Bengali, Greek, Sanskrit, Italian, Rumanian,

Bulgarian, Croatian, Russian, and so on.

Many Indo-European

languages show diverse types of sound symbolism. Not

surprisingly,

symbolism,
clusters,

number of

these

include consonantal

vowel ablaut or alternation, phono-morph

and

reduplication.

Shields provides an interesting theory for the origin of

reduplication
stage,

in

Indo-European languages (1976). At some

Proto-Indo-European carried words which were

with either an amplified or augmented ending.

CV-

Ultimately,

conjecture follows the ubiquitous start of language in a

child.

First a child

sounds.

Vowels and

manipulation are mastered next. Then,

stage.

human

begins with vegetative process and those

physiologically constrained

a 'babbling'

this

Finally, because

CV

their

prosodic

sounds are made in

of the Bernoulli effect and

other constraints upon length of vocalization, a child learns

CVThis

its

utterances can 'go out like a lion' or 'go out like a lamb'.
is

precisely

what Wescott (1980b), and Swadesh (1971)

130
proposed as a feature

in

Proto-Unified-Language: (Biologically

contrainedvocals)>(V+)>(CV)>(CV+)>(CV+(SOFT-UNMODULATED)
or

(HARD-M0DULATED)>(C1VC1), (C1VC2)
For Proto-Indo-European

CV

widespread
shifts

CVC

n/

and

/-s/

as

second consonant, the other the

the

(Shields

/-s/

'strong'

to

mime

carrying'

[epher]

'carry'

[epheren]

Ionic Greek) vs. 'he

[nava]

argues

[estin]

is'

(Sanskrit),

'THE name'
Greek)

Greek);

and

[sniz]

'he

vs.

[esti]

'be'

(Attic-Ionic Greek);

[niesen],

the

(Sanskrit)

[ball]

(Attic-Ionic

(Attic-Ionic

/-

other reduplicative

'name' [nama] (Sanskrit) vs.

[balin];

[naman] (Sanskrit);

'sneeze'

theory

This

For example: CV-N; 'average, norm'

functions.

was

1976:37).

marked forms and when they appear,

meanings they serve appear

vs.

two groups. One prefered dental-

classes split into

fricative

Shields says that the

breakage of a single vowel). Then the entire

alveolar nasal /-n/ as

dental

least,

forms became affected by numerous dipthong

(or prosodic

pool of

at

et cetera.

(Attic-

CV-S;

(English) (Shields

1976:37).

Though

it

is

unclear

may have uncovered

than

distantly
n/

how

true Shield's hypothesis be, he

taking

place

Assuming

the

transformation

Proto-Indo-European.

phones make 'sneeze' doubly sound-symbolic,

interesting

to

see

it

more
/-s/

and

would be

what other language phyla label the process.

Although the following examples are not balanced by

phyletic

numbers of languages sampled, they suggest

frequencies of

/-s/

/-

and

/-n/

in

'sneeze'

and greater use of

131
fricative

and

nasal

than

an

sound-meaning theory

arbitrary

could explain as normal variation. They include: Afro-Asiatic;

Arabic ['atsah], Hausa [atisawa], Burji [hal'is], Amharic

[anattasa], Somali [hindisayyal, Austronesian; Hawaiian

[eki'he],

Tonga [mafatua], Pascuense

[tehi],

Maori [matihel,

Tahitian [tihe], Esicimo-Aleut; Aleut [asukuqig],

Eskimo

[tagiorlorpok], Indo-European; Croatian [kihanje], Danish

[nysen], Gaelic [sraiartec], Hindustani [chi nkl, Icelandic

[hnerva], Lithuanian [ciaudeti],

Pali [khipita],

Portuguese

[espiro], Rumanian [stranut], Italian [starnuto], Niger-

Khordofanian; Mbukushu [ya0imi0al, Swahili [cafya], Shona

hotsira],

Xhosa l-0imlal, Zulu [Gimula], Ndebele [eimula]. North

Amerind; Blackfoot [wa'sl:yl], Crow

Micmac
[hacil,

[-

[a pi

sxi],

Hopi [aasi],

[ejgwitl, South Amerind; Cashibo l?atisanki], Cavine~na

Chama

[ati],

Marinahua

[atisi],

Mayoruna [atisun],

Shipibo-Conibo [hatisainl, Uralic; Finnish [aivastaa], NiloSaharan; Miza [o-si], Ojila [tssl,
[tso], Sino-Tibetan;

Logo [syal, Lugbara

[tso],

Lokai

Cantonese [dahakci], Tibetan [hsptiksp],

and Altaic; Japanese [kusami], Turkish [aksirma], Korean

[caejae], Kurdish [pijme].(/s/=i^/50

/n/=iV50 /fricative/=48/50

/nasal/=23/50)

Languages arising
consonantal

symbolism.

later than

Proto-Indo-European show

The dental/alveolar voiceless stop

/t/

apparently has a long history as the carrier of the meaning

'stubborn resistance'
Latin,

among

Slavic, Russian,

Old French, Old Spanish, Greek,

Old High German,

Castilian,

Old Portuguese,

132
1990c:69-80). Malkiel argues that

Czech, and others (Malkiel

it

would be usual

NOT

obstruction

deliberate

to label

the

of the

occlusion

or

most easily observable

breath

"namely the one effected by pressure of the

against

the

teeth,

meanings about
and

the

Although Latin
language,
its

it

of the tongue

gums, or the hard palate" --with

'resistance',

'stiffness'(Malkiel

tip

passage

'strength',

'firmness',

'toughness',

1990c:71).
is

a generally

commended

reveals a wealth of sound

scholarly

symbolism. Evidently,

people highly prized birds for some 315 terms are known.

Not counting the more ancient Greek

least

107 species with 232 terms.

names

are

not only

More

than

20%

names

at

of these

sound symbolic (Andre 1966:146). Latin people did

name

that could

loans, Latin

a bird for

its

call

because

it

was noisy animal

The same

not be labelled on sight very easily.

species of bird might have a differing song according to

breeding group.
Latin labelled
parts

of

[vano],

its

the

Many Romance

many

of the same birds by picking out differing

tonal vocabulary.

Sanskrit

word

In

name

'litweet'

named

French, a plover was

for 'sneeze'.

description of the plover's song,

English

languages emanating from

'li

Meanwhile

huit',

the

French

somehow became

the

(Andre 1966:148).

Echo-words are found

in Bhojpuri, an

language. The basic type of this word form

Indo-Ayran
is

similar to

conjectures of Shields, Swadesh, and Wescott. In

it,

(CV

-i-

any

133
C)

is

added a duplicated

(C + either

initial

(Tiwary

/o/ or /u/)

1968:32).
For Gujerati and Marathi, the most commonly reported

echo-word begins with

(Tiwary

1968:38).

/b/.

The most favored

in

Hindi

is

/w/

Bhojpuri examples, the echo-words can

In

also act as verbs or nouns.

However, since the root

really

implies the echo, and not the reverse, echo-words are far from

being a unified sound symbolic system with specific and

functions

precise

(Tiwary

Echo words and

sarcastic

1968:35).

their reduplication

can serve as a

speaking register and point out status distinctions

according to role and capital accumulation. Like 'emotional' or

words, echo-words load

'affective'

semantic meaning into

words. Tiwary comments that this sociolinguistic function for

reduplication

is

manipulation

of

highly

young-old

Sound symbolism
negotiating
(hardness,
It

is

softness,

is

social

strategy(Tiwary

1968:36).

best thought of as a process of

between the world's sensory potentials

relations

et

cetera)

and mental and acoustic

reality.

not merely a single group of un-regenerate words which

echo loud phenonmena.

symbolism may be
selecting

Recently,
the

developed as part of the behavioral

If

so,

intrinsically

affective connotations

linked within

and sound

species

complex acoustical exchange of information.

it

has been

'emotional'

suggested that substantial proportions of

vocabulary of ordinary speech contains

symbolic markers as well as intricate prosodic schemes.

sound

134
(Markel 1990). Wescott regarded
anger-context connection

this

as

the reason

labio-velarity

to

an

for

and derogation

in

English (Wescott 1971a). Presumably, the quickening pace of

the physiology in

conflict

social

would produce short

biting

sounds with articulatory sweeps of sound.

English
refered
list

is

to

as

well studied, though generally

is

an 'un-sound-symbolic'

created to

show

that English

language.

it

has been

Below

word

contains a substantial

number of sound symbolic words. These words

contain

either

an etymological primary form of sound symbolism, or show

associated cluster because of semantic cohesion

(Malkiel

1963).
This

list

was compiled using the data reported by

Marchand (1959) Wentworth and Flexner (1960), Malkiel

(1963),

Smithers (1954), Wescott (1980c). Over 1,000

suspected or attested sound symbolic words are given. The

presented in table

3.. a.

is

unanalyzed and, though

far

list

from

presented as a synchronic picture

being comprehensive,

it

is

of American English.

It

suggests that the "un-sound symbolic"

language of English

may be

languages and that English

everyday speaker,

(i.e.

sound symbolic as other

as
is

well equipped to provide

with an average 24,000 word

vocabulary), a vocabulary consisting of

words.

This

its

extraordinary

3%

sound symbolic

amount should warn against

cursory dismissal of sound symbolism in English.

135
Table 3. a.
English sound symbolic words

136
Table
22.

-iggle

3. a.

continued

137
Table
52.

-udge

3. a.

continued

138
Table
71.

p-

3. a.

continued

139
Table
85.

kl-

3. a.

continued

140
Since English
the

is

an important world language, violation of

sound meaning hypothesis

arbitrary

a language can contain such

important because
to

certain registers

it

is

demonstrates

sound symbolism

of semantic intent.

fiction,

and any communicative activity

activity

is

is

is

crucial

The above words

premium. One need only

major event. That

bevy of sound symbolic terms

notoriously evident in poetry, children's

at

are

books, works of
in

which emotional

listen

to the

vocabulary of a sports broadcaster, a preacher, a politician, or

a
is

salesman

to

hear large numbers of these words.

Much work

neccessary to disentangle a disregard for sound symbolic

words from

their

function

as

affective

vocabulary markers.

The next chapter examines dozens of sound symbolism

experiments.

In

unique ways, these experiments were

designed from perplexing questions posed by

symbolism events every speaker

is

sound

aware of from

birth.

CHAPTER IV
OTHER SOUND SYMBOLISM EXPERIMENTS

Types of Experiments and

their

Limitations

This chapter investigates the findings of various sound

symbolism experiments. These

classes.

broadly into three types of

psychological tests can utilize a linguistic

First,

called an artificial
entirely

fall

lexicon.

With

this

technique,

medium

linguists

construct

meaningless words from a phonological range. Linguistic

can

researchers

thereby

present

foreign

and native phonemes

in

nonsense words for evaluation by subjects.

By
to

presenting 'nonsense' phono-morphs or 'unrealized words'

speakers,

scalar

phonetic, acoustic,
the language's

connotative evaluation of specific phonemic,

or prosodic

own phonemic

qualities

may

distribution,

frequency, would bias association

in

be viewed. Otherwise,
its

own word

between sound and meaning. Put

simpler, the a priori assumption for the artificial

lexicon

is that

"nonsense" or "neutral" words can be constructed. In the most

evident

language evolution

were unlikely
behaviors

to

to

be the

scenario,

first

non-affective

vocalizations

of these communicative vocal

emerge.

141

142
Other sound symbolic inquiries use a natural
experiments,

lexicon. In these

words of specific languages are presented

to

speakers.

Again, various measurements can be made concerning a

speaker's

connotative evaluations

of antonymic,

of groups

synonymic, homonymic words. Lists are also made which purport

to

contain

words or words argued

"unrelated"

arbitrarily. connected

phonemes.

with

There are major drawbacks

experiments.

As every

contain meanings

to

to both these types of

knows, numerous perceptual

linguist

parameters exist on which speakers could rate and rank words

connotatively.
is

Deciding which boundaries connect which meaning

complex problem posed

anthropologists.

cultural

neurolinguists, physiologists,

to

Some

borders

perceptual

are

and

shared

widely among mammals, such as the categorical distinction of

(voice-onset-timing,
are not so widely

i.e.).

contrast,

In

many

VOT

physiological routines

shared, such as the ability to

make imploded

grunts.

Simply deciding which perceptual categories a species has a

potential

for

or describing

that

species'

physiological production

range does not easily lead to awareness of which motor routines a

species connects with which meanings. This
difficulty

for ethological

studies

acquistion studies of children.

languages.
is

general

If

a widely realized

is

of alloprimates as well as language

similar bent

languages can be compared

at

is

all,

true for
it

is

human

because there

acceptance that semantic categories are more

overlapped than phonological ones. Even when two languages

143
contain a word which carries the meaning

"up,"

their

ultimate

semantic boundaries for "up" could be culturally dissimilar.

Certainly, the meanings offered to a speaker by his language

can never divorce themselves from those sounds they use

convey information. Longer words are learned

shorter words.
are

involved.

many

layers

longer words more sounds

In

To

functionally

of neuro-motor

later

in

to

than

life

and motor routines

use words, a speaker ascends through

practicing.

enters into the functional use of

Nevertheless,

speaker

the

words on a basic sound-meaning

level.

Language

is,

by definition, a shared associational phonology

The paradox

system.

sound symbolism experiments

in

is

that

they

cannot control the amount of regularity and associative input a

language bestows upon

to

its

speakers and this

is

what they presume

measure. The results they garner are to a considerable extent

influenced by age, intelligence, gender, and dialect group of their

subjects.

So,

perceptual
easily

for

events

any language experiment the greater number of

provided for subjects by their language might

influence the smaller

number of

features

to

be studied. For

example, any speaker might note longer words are rarer in speech
than shorter words.

expressives,

Even words of

interjections,

or

specific

semantic type, such as

exclamations, might easily be

recognized by a speaker as carrying unusual phonological

structures

As

and

elements.

a final type of sound

experiments

proposing

sound

symbolism research, there are

symbolism hypotheses about natural

144
languages modeled upon types of analysis which sometimes
exclude speaking subjects. This type of analysis
to

synthesize

My

research.

symbolism

Of

and

written

sporadically

crucial in order

published

an example of this genre of sound

is

research.
three types, only for the last can sound symbolism's

sound-meaning morphs enable reconstruction of the

contacts

historical

sound

internal

1990).

dissertation

all

identified
distant

differently

is

also can

It

symbolism

as

stages

shift

between world languages or expose

(Malkiel

Hamano

1963;

their

Malkiel

1986;

allow investigation into the treatment of sound

language universal. Finally, with steadily

improving language samples worldwide,

it

can elucidate the

vocabulary use by human culture in a pre-sapiens era.

Sound symbolic experiments

linguistic

features

associated

speaking

subjects.

Viewed with

statistics,

as

associations,

normal distributions of

between languages.
or biased

all

Often,

typically produce a

with

their

semantic

inferential

list

of the

meanings reported by
and nonparametric

occurences violate boundaries of

other sounds
these

and meanings within and

experiments

use

non-equivalent

language samples. One experimenter could choose two

Altaic and one Sino-Tibetan languages to

compare with English, an

Indo-European language. Another might choose

to

Indo-European languages with a Finno-Urgic tossed

study

three

in.

In addition, functional or operational definitions of

sound

symbolism are hazy (witness the preponderance of terms and

overlapping meanings

discussed

in

Chapter

III).

Experimental

145
methodoligies often do not investigate equivalent linguistic
cognitive

Some measure

structures.

phonemes and meanings,

features

associations

and

between entire

and meanings, words and

meanings, and other more involved variants. The data results

issuing
fully

from such mixed definitions and testing elements make

comparable sound symbolic experiments

Needless

to say, the

rare.

conclusions of most sound symbolism

experiments are not easy to compare with one another. Beyond

this,

however, sound symbolism studies have yet

comprehensively reviewed. While Fonagy's work

to
is

be
a

an admirable

melange of documented sound symbolism ideas and conjectures,

presents

eventually

neo-Freudian explanation for

sound

symbolism (Fonagy 1979). Freudian theory when applied

cultural

anthropology and linguistics

More

is

recent and systematic effort

hardly

it

to

bio-

without objection.

forcing sound symbolism

is

again into serious linguistic study (see especially Malkiel (1990)

and

Hamano

(1986)).

Below

review some highlights of the

years of experiments into sound symbolism.

They

last

70

rate a close look

because they demonstrate the difficulties for postulating meanings

for

human words

nonarbitrary

any date 100,000 B.P. without recourse

at

sound-meaning processes.

Sound symbolic words

one of the plausible bridges connecting present language

language.

The experiments

findings presented in
dissertation

stands

also

Chapter

upon

solid

II

to

corroborate

proto-

numerous individual

and bear witness that

theoretical

to

are

ground.

this

146
Sound Symbolism Experiments

"Size"

Otto Jespersen and Edward Sapir are responsible for early

20th century interest in the search for a linguistic gesture

To

representing a size concept.

justify

their

they theorized

search,

on the structure of sound symbolism. Of such, Jespersen says

sound symbolism

of sound

1921/1947:397).

correspondence between

1921/1947:396). There

sense (Jespersen
denial

natural

is

symbolism

in

to

it

no logic

to

an extreme

any language (Jespersen

Since there are words

would argue are instinctvely adequate

says,

is

sound and

a
to

of speakers

majority

meanings, Jespersen

should be important to study what ideas lend themselves

sound

symbolism

1921/1947:378).

(Jespersen

"front-back"

In further study, Jespersen outlines an early

hypothesis centralized about the value of the vowel

/I/.

He

collects

numerous detailed examples from Indo-European languages

including those for

'diminutive endings'. For

high, front
small,

'young animal', 'small thing', and

child',

'little',

these categories, the

all

unround phoneme, would be regarded

slight,

insignificant,

/I/

in

vowel, a lax

names of

and weak meanings (Jespersen

1922/1949:557).
For Indo-European, Jespersen argues that sound symbolism

makes some words more

their

fit

to

survive and gives them strength in

struggle for existence. This

necessarily

one present

in

distant

manner of sound symbolism

is

not

language origins and there might

147
be progressive tendencies

expressions

(Jespersen

towards fuller and

more adequate

1922/1949:559).

Sapir tested the conjecture of Jespersen in his classic series of

experiments

(Sapir

the challenge:

themselves out

1929).

"can
in

His

experimental

be shown, that symbolisms tend to work

it

vocalic and consonantal contrasts

of the arbitrary allocations of these

spite

consonants

in

the

launched

orientation

socialized field

strictly

and scales

in

same vowels and

of reference?"

(Sapir

1929:226).
measuring sound symbolism, Sapir suggests that languages

In

contain

"expressive"

1929:226).

and "referential" classes of vocabulary (Sapir

Preferential

tendencies

"expressive"

for

vocabularies

might include greater use of phonological contrast. than

"referential"

would be

ones.

less

It

presumed the

is

vocabularies

referential

bounded by innate trends than the expressive ones.

Since whatever innate tendencies exist for sound symbolism

might be expressed

in

phoneme, Sapir devised

sound contrasts rather than merely one

a

number of

artifical

lexicons

the range of high-low and front-back vowels for English.

list,

60 pairs of nonsense words were constructed using

vowels
/a,

se,

in
,

a
e,

lineal
i/

were used. In the second

half,

all

English

non-English vowels

chosen meaning, for example, [mil] versus [mal]

to indicate the

words were equal

one

In

inventory. For the first 30, this meant the vowels

were also used. Each pair was presented aside an

was asked

represent

to

to

term for the larger

arbitrarily

'table'.

'table'

omit a mark (Sapir 1929:227).

The subject

or if both

148
Another

list

ages varying from

had 100 pairs and were tested on 500 subjects of

and including English and a smaller

11-21,

number of Chinese speakers (Sapir 1929:229).

In

two runs

these

and for two other experimental variations, Sapir found that

"symbolic discriminations run encouragingly parallel
ones based on phonetic considerations"

objective

to

the

(Sapir

1929:238).

Put more simply, the subjects weighted their responses on a

continuum of

size

lowest vowel

/a/

in
to

response to scalar frequency levels from the

the

highest vowel

/i/.

When vowel

contrast

included vowels in the middle levels, the objective responses to

size

differences

were minimal (Sapir 1929:230). Additionally, the

Chinese subjects' evidence was

speaking ones (Sapir

English

In these

in

the

same direction

1929:231).

few experiments, Sapir sought

connotative evaluations,

as

as that of the

socially

to

constructed

measure
and bordered by

age and speaking group, and to distinguish from that innate

tendencies to symbolism of a presumed and increased use of sound

Many

contrasts.

student,

Stanley

other works build

Newman

restudied

upon
the

this

theme.

size-sound

Sapir's

symbolism

problem (Newman 1933). He asked 606 students of varying ages

to

rank nonsense words according to size value. Youngest subjects

produced the most widely varying rankings, then the older, and
the

oldest

showed

the

most consistent rankings (Newman

1933:59).

Even though age

affect choices, all age groups consistently

rank front to back vowels on a scale of small to large.

He

argues

149

experimental

repeated

with

large

consonants

/n/,

/p/,

Newman

tested

/br/,

and

/d/,

/gr/,

and

/s/,

and dark-bright

small-large

/gl/,

and

consonants

bright

/gr/,

/br/,

small

/r/,

/k/,

/s/,

1933:63).

also did a

word search

in English for

evidence

While Newman's

were not surprising,

results

reasonings were abstruse and unclear.

among

causal explanations.

his

was inconclusive.

of judges according to fitness. His evidence

separate

be

to

size-sound symbolism and had each word rated by

indicating
series

consonants

(Newman

/I/

series

phonemes. Dark consonants appeared

other

and /m/,

and

and visual signals influence the associations.

kinesthetic,

acoustic,

statistical

Additionally,

he

Which causes an

to

fails

idea of

"small", small oral size or high frequency or tongue height or

constriction

kinesthetic

As soon

as

upon the tongue?

Newman's study was published, another restudy

was made. Bentley and Varon (1933) found

that high

pitch

was

associated with "small", low pitch with "large". Front vowels tended
to

be thought small and back vowels large (Bentley and Varon

1933:86).
Since their
they

effectively

neccessarily

upon

this

words were pronounced behind a cloth screen,

deduced

Newman

duplicating

(Bentley

test

that

the

(1933) and

connotative rankings,

Sapir (1929),

largely

were not

based upon visual stimuli from mouth configuration

and Varon

1933:85).

Later researchers

tested

further

note in the manner described below for Brown, Black,

and Horowitz (1955). Siegel, Silverman, and Markel (1967) found

150
the

found for auditory presentations significantly increased

results

with

combined visual and auditory presentation.

Thorndike and Orr measured languages for antiphony and

front-back vowels in certain semantic fields also sought by Sapir

and Jespersen (Orr 1944; Thorndike

1945).

showing opposition of vowels relating

semantic field (tip-top,

to

"roughly" the same

strap-strop

slit-slot,

Orr gathered examples

e.g.)

Thorndike,

meanwhile, conducted an impressive collection of 1,970 English

words and ranked them according

smallness/largeness

to

ratios.

Thorndike's results are worth a closer look simply because

and Orr. The systematic

they corroborate Sapir, Jespersen,

collection

many words

of data entailed collecting as

as possible for

each of the following 17 categories of English vowel sounds:

in

bit',

or [el],

machine',

as in

[i]

as in

[se]

or

[1]

[e] as in 'bet', [el]

[i],

in 'beauty', [u],

[u:]

The

final

and

[i]

as

probable

it

that

'small'

words

[o]

'bait', [e]

smallness,

as in

'bite',

1945:11).

number of words suggesting

for

largeness,

and probable largeness.

smallness/largeness ratio was twelve times larger for

chance

and

[a] as in 'but', [al]

or [yu:] (Thorndike

Each sound was then coded

smallness,

and

He

was

for [o]

the

difference of -1-0.046

in [I] or

[i]

[o].

percentage of

states succinctly,

is

about

in

'small'

words

is

about

in

"the theoretical

words

1700 (.0005

chance for the difference of -0.051

'large'

between the percentage of

and the percentage of

could occur by chance

the corresponding

as

as in 'box', [ou] as in 'bone', [o] or [ou], [u]

'bat', [o]

as in 'bush', [u:] as in 'fool', [u] or [u:],

[yu] as

as in

[I]

in

16,000 (.00006

i.e.)

in [o]

and

the
i.e.)

The

151
chance for joint occurence

(Thorndike

than

less

is

in

250,000,000"

1945:10).

Thorndike also notes

that at least for English,

German, Russian,

Greek, Finnish, and Hungarian certain phoneme clusters are found

more frequently
for

(Thorndike

unpleasantness

phoneme word
showed

there

small-large

words connotating pleasantness than

in

lists

is

[i]

In

similar

size-

some association between front-back vowels and

meanings. The evidences were much weaker than for

or

[I]

is

it

is

a special case for English. (Thorndike

Thorndike and others.

she remarks that

symbolism

possible that the size-sound

Maxime Chastaing expounded upon

/i/,

tallying

from Greek, Hungarian, and Finnish, Thorndike

English, however, and

utilizing

1945:12).

ones

in

In

these speculations of

a review of the significiation

when considering

1945:13).

sound's

of the vowel

motivation,

not only determined by circumstance, text, context, but also

phonetic uniqueness and what

(Chastaing

height,

vowel

quickness,

sharpness,

narrowness,

obtains

meaning largely from

is

a function of

its

represent

to

quieter forms of talking,

meaning

she notes

/i/,

Indo-European languages

its

is

its

used to name

1958:413).

In considering the

dozen

actually

is

it

it

birds,

relation

use,
to

is

used

in at least a

smallness,

clarity,

stages of breathing,

lightness,
its

it

and rapidity.

It

but importantly,

its

The

/i/

other sounds.

vowel

has value also as a structural or gestural event in the articulatory

musculature.

152
Her
led

I'll

interest in the qualities of

seven experiments (1962). The

to

speaking students
darkest vowel.
it

meaning assigned

list

vowel

The more

series

to the

vowel

had 30 French

first

order of lightest vowel to

in

and higher the vowel, the brighter

frontal

was reported. Next, 20 students rated 4 words, Kig, Kag, Kog, and

Kug, according

whether they were bright, neutral, or dark. The

to

Kug was

nonsense word Kig was brightest, Kag was neutral, and

darkest

(Chastaing

1962:2).

Chastaing asked 168 elementary students to replace the vowel

in

the word, Pab, with one befitting the

Most common

twilight.

and twilight

/u/

(Chastaing

asked to replace the vowel

clarity.

66%

dawn was

for

replaced

the

1962:3).

/e/,

with

/u/

day

for

Thirty-five

the word,

in

dawn, the day, dusk, and

/i/.

Grum,

for dusk /o/

/i/,

were

students

to best indicate

Another experiment asked

students to modify words so as to darken or lighten their

41

meanings. The vowels

darkened.

Word

were asked
lightest

as

to

pairs

/i/

and

/e/

were presented next

choose the

compared with

light
/u/

to

or dark pair.

for

vowels

lightened,

91%

/o/

students

The vowel

and

/u/

and they
/i/

was

of the subjects. Finally,

elementary students were shown pictures of night and day, and

large objects and

and
large

/u/

small objects. Over

connected

/i/

with day

with night. The same percentages held true for small and

(Chastaing

The

1962:5).

results of these experiments

subjects are far from

is

75%

surprising.

whether sound symbolism

is

What
innate,

upon French speaking

they do not show, however,
learned,

or a combination

of

153
Chastaing remarks on

both.

and suggests certain sounds carry

this

more non-arbitrary meaning than others because they

during essential behaviors. Her

list

some of

contains

discussed and examined in this dissertation:

and

words

the

cough,

suck,

drink

Most
in

vomit,

created

are

sound symbolism experiments lacking

linguists believed

hard cross-cultural results.

One attempt

vowel-size controversy was Ultan (1978).

from 137 languages.

It

answer the

to

His study gathered data

coded for the existence of various phonemic

and morphic structures carrying meanings involving

force,

quantity,

/i/

intensity,

pejorativeness,

time,

age,

size,

distance,

sweet-

gender,

good, sensation, and so on.

Ultan finds size symbolism represented by a number of sonal

forms

in

27%

of his languages. This result

reservation though.

is

obtained with

Of 137 languages sampled, over 40

are

from

North Amerind. For the other 16 language phyla, only the following
have an
(15),

N=5

their

for

sample: Austronesian

(9),

Indo-European

Afro-Asiatic (5), Altaic (13), Niger-Khordofan (14), South

Amerind

and Dravidian (11). This sample exemplifies the

(9),

difficulties

of obtaining data on language phyla distant to Indo-

European.

He

reports

symbolism.
nearness

is

of the 137 languages contain distance

The overwhelming
front or high

corroborates
suggests

33%

that

many

earlier

distance

favorite

feature

representing

vowels (Ultan 1978:546). This

sound symbolism experiments.

symbolism

is

It

also

proto-language conceptual

154
fragment because
within

exists

it

between disparate languages and

both

mental frameworks of the language speakers.

the

Another sound symbolic tendency was the presence of a

short-long,

one-many, part-whole conceptual overlap.

represented

shorter

events

or

singular

Long sounds represented many

whole events (Ultan

instances,

few instances.

or

more than one

or

instance,

set

longer,

1978:547).

In regard to Sapir's "affective" words, Ultan finds

languages

sounds

Short

many

emotions of a verbal praising or hyporcoristic

the

and pejorative nature into words with the use of ablauting devices
(Ultan

1978:547). In

"emotional"

speech,

this

means speakers can

change distinctive features, vowels, articulations, and consonants

within a word or phrase to enhance connotative intent. For

example,
in

in

hypocoristic speech,

the pejorative

flip-flop over the

italicized

mish-mash

call

my

friend

it

is

Bobby-Lee,or

"it

can do a

about the hub-bub". In each

phrase a front-back ablaut

hypocoristic,

is

present.

not odd to find that the pejorative,

and affective speech registers share the same formal

(Ultan

1978:547). Extending this thought,

sound symbolism

is

symbolism definitions of Chapter

events, this could not be true.

recognize a brief

varied conceptual

list

styles

it

appears that

mechanical system possessed by speakers

which transcends various speaking

to

can

can trivialize an event by saying

Ultan remarks that

features

But

styles.
III

it

are

If

the

numerous sound

labelling

entirely

separate

might be easier for speakers

of sound symbolic axioms to apply to

of speech

than

to

construct similar

155
rules

linguistic

when changing speaking

we do know

least

As

the

a fact,

speakers do recognize a series of sound

that

is

registers.

symbolism axioms and apply them appropriately when within

where connotative

settings

Lexicons

Artificial

Artifical lexicons

sounds.

individual

intent

in

is

be amplified.

to

Sound Symbolism Experiments

were designed

to find connotative value for

nonsense words

of Sapir's

variants

Interesting

sometimes included the use of the semantic differential

(Osgood, Suci, and Tannebaum 1957).
three

which provided

tests

Oyama and Haga

test

(1963) ran

semantic profile of nonsense words

and visual figures. Their students rated

14 line drawings and

16

phonetic items of 35 semantic scales.

surprisingly, they found certain sounds,

Not
were likely
/r/,

/o/,

to

and

be named sharp and angular. Others as /m/,

and

/b/,

/I/,

/k/,

/t/,

/n/

were pinned

to

/z/

/u/,

round figures (Oyama

and Haga 1963:141). Their results duplicated Miron (1961) and

Newman
farther,

(1933) in that /u/ and /o/ vowels were

fuller,

softer,

and unreal than

high vowel
/i,a,e,u,o/

and

/r/

stronger

/i/

heavier,

and

/i/

/e/

hotter,

was considered brighter than

than

happier than

felt

/m/,
/i/

to

size.

happier than

/t/

or

and vowels

/u/

and

and

/o/

to

be deeper,

more smooth,

stable,

(Oyama and Haga 1963:138). The

vowels varied according

were

wetter,

felt

/m/,
/a/

/o/,

front

and on a scale of

Finally,

consonants /k/

and consonant

/k/

was

were more sober yet

(Oyama and Haga 1963:139).

156
Lack of information upon how subjects rank nonsense words
compelled Weiss (1964)
its

efficacy

to

sound. Here,

to

measure stimulus meaningfulness and

88 female undergraduates ranked

contrasted "high" and "low" meaning nonsense words on scales of

magnitude, brightness,

and angularity. Weiss argued that a variety

of schemes are used to "decide" connotative regard of nonsense

words.

Some

related

to

(Weiss

subjects reportedly

their

would think of

Latin,

understanding of "baby language" to

1964:261).

He

is

make

a choice

negative experimental results

believed early

with nonsense words and figures

others

due

to

the fact that a direction

of evaluation must be introduced for the subjects before they will

achieve significant agreement as a group.

His results showed the brightness dimension increased in

response to requesting
1964:262).

this

He comments

particular

that

judgement (Weiss

Brown's learning theory appeared

closer to the truth than the Gestalt theory of physiognomic

perception.
for
in

In

other

words,

speaker agreement increased

with

age

sounds and meanings merely because the linguistic associations

a particular language

have had more time

to

become known

to

speaker.
Important data fueling a disconfirmation for universal sound

symbolism
to

present

is

seen in Taylor and Taylor (1962). Their method was

144 nonsense

unrelated languages.

and vowels
languages.

common

CVC

words

to

speakers from four

These words were composed of

to

all

the English, Japanese, Korean, and

consonants

Tamil

Speakers rated the nonsense syllables according to

157
warmth,

movement, and pleasantness. They found

size,

the

"meanings associated with any particular sound were different

from language

to

language and there was essentially zero

correlation

between the symbolism scores found

languages"

(Taylor and Taylor

They

to

1962:356).

Korean and Japanese responses were the

also note that

most similar due

any pair of

in

being the most similar cultures (Taylor and

Taylor 1962:356). This

is

have

not surprising considering both

sound symbolism systems and are from the Altaic language

phylum (Martin 1962)(Hamano 1986). The major problem with

Taylor and Taylors' study

is

that

prosodic-meaning

possibility

of

symbolism.

A phoneme

is

fails

it

feature-meaning

or

large

mental event and

by the speaking group which uses

many
This

to account for the

it.

Further,

is

to

features

say,

when

phoneme and

sonogram

is

utilized to distinguish the

contained in a phoneme for a speaking group, the

often

physically do not exist

experimental design

may have

(Snowdon 1986:496). Their

missed

entirely

front consonants

Further,

test

four languages

all

represent

used

features

phoneme

merely because

because they intersect

the

or

might each be regarded on a similar semantic

but with the experiments, the

of sound.

common

Voicing, vowel length, or

universal regard for sound and meaning.

unit

constructed

is

prosodic markers do not exist in a non-mentalist reality.

structures

pole,

the

sound

in

sound

is

is

the

least divisble

phonemes

are

chosen

no guarrantee they

symbolism.

158
In a critical review of then recent

sound symbolism

experiments, Taylor (1963) entirely dismisses

Some experiments may have been

foreign equivalents which

mimed

translation"

well as

as

English words. At least

common

mal" nonsense pair

"vig-vag"

may have

such,

it

is

mode

of the

controlled

Insup

Kim

entirely

learned,

patterns

in

in

(Taylor

the

argued, Sapir's "mil-

English,

"if

if

were

it

stimulus presentation,

1963:205).

Taylor remarks that sound symbolism

and that

clear

is

it

may have been

kinds of verbal subjects, and subject tasks

incorrectly

lost

is

returned different results

Similarly the

instead.

universal.

Other experiments did not

people.

As

control for consonantal effects.

case as

by researchers choosing

biased

also use the adage that "something

linguists

its

we

obtain

is

probably

phonetic symbolism

German, Russian, and one non-Indo-European

language, there should be a heirarchy in the degree of relatedness

among

those

1963:209).

languages as reflected in phonetic symbolism" (Taylor

She restates

must be found

tenet

of Saussure;

"a

new hypothesis

not only for the fact that people

that accounts

associate certain
that

sounds with certain meanings, but also the fact

people speaking different languages associate the

sounds

with

different

meanings"

(Taylor

same

1963:206).

Contrarily, researchers continued to find data disputing a

more

relative

language specific sound symbolism. In

found British children tended

angular pictures and

much

like

to place

ULOOMU

African children

the

word

1961,

TAKETE

Davis

with

with rounded ones. They did this

who had never been exposed

to

English.

159
Johnson, Suzuki, and Olds (1964) tested deaf and hearing high
school and college students. In rating

14 pairs of

artificial

words

with polar adjectives, the two hearing groups established

significant

correlation.

The deaf group matched neither hearing

group nor matched each other within

and Olds

group (Johnson, Suzuki,

their

1964:236).

Mixed

results

(Atzet and Gerard

were found

in

Navaho speakers

a test of 60

1965). Using the

of antonyms

list

Brown, Black, and Horowitz (1955), subjects were

meanings of Hindi and Chinese
only Chinese many-one,
one,

From

guess the

For twenty antonym pairs,

pairs.

smooth-rough,

examples were

hard-soft

to

known from

and Hindi many-

thin-fat,

significant.

these results, the authors state "the

amount of overlap

between a given pair of non-cognate languages of such images

called forth by
it

does occur

it

similar-sounding words

probably minimal and

is

will often be coincidental"

if

(Atzet and Gerard

1965:528). The numbers of these words must be very small, they

aver,

and are localized

"sizzle"

to imitative

(Atzet and Gerard

words such

as

"cricket"

and

1965:527).

Their comments pay homage to the arbitrary principle, again,

without

much comparative language

data.

How,

for example,

they explain the fact that in the 50 language sample for

seen in Chapter

Or

that a

II,

presumably quiet noun

features? Worse, for a

over

80%

98%

obstruents are present in

much

like

NECK

of a balanced language sample

COUGH

of the examples?

should show

quieter noun like

show

70%

FOOD, why
it

would

velar

should

carrying a nasal

160
and Atzet and Gerard (1965), the

feature? For both Taylor (1963)

of feature-meaning

denial

argued

with

little

supporting

a retest

in

were presented with 28 word

subjects

words and one Japanese equivalent

mean percent agreement was

is

research.

Antonyms were eliminated

level

between distant languages

association

with two English

one of the two words. The

to

60.8,

sets,

by Weiss (1966). 318

beyond the .001

significant

(Weiss 1966:99). The correct English choices were; frosty,

twitch, stun, gnaw, cut, sting, ache, grope, rub, lick, kiss, wince,

bleed, whip, itch, tickle, sweat,

whisper,

moist, thud, vibrating, tick,

scald,

harmonious, howling, hoarse, wheezy,

1966:100-101). This

list

words chosen are of

is

antonym of

imperative

sensory nature and most do not carry easily

"tickle"

second

(Weiss

given to demsonstrate that most of the

accepted antonyms. The semantic difficulty

the

splendid

for

person

example?

Is

it

is

easily

"torture"

seen.

What

is

or "pain" in the

case?

While Weiss provides evidence

that high

agreement may be

obtained in guessing choices, the results are not easy to analyze.

Many

of his translations of English words into Japanese directly

evoked the Japanese vast sound symbolism system. "Wheezy",

example,

is

1966:101).

translated into Japanese as

As Hamano noted

"Zei-zei

(Suru)" (Weiss

(1986), the lack of reference to the

existence of such a Japanese sound symbolic system at

undermines

the

for

all,

results.

Taylor did not appreciate the Weiss word

list

She noted many of the words were sound symbolic

either (1967).
in

Japanese

161
and that

this

consideration along with the factor of word length

influenced choice more than sound (Taylor

denies universal phonetic symbolism (then

1967:237). Again, she

known

as

UPS) (Taylor

1967:238).
Taylor's criticism brought

more experiments, but her

counterpoint mostly failed on one point.

"imitative"

added

in

words

are onomatopoetic,

are

only need to read the

A
carried
to

rate

It

failed

to explain

of

less

than duplicates of one another.

COUGH
It

agree

all

terms

seems clear

the

in

that

Appendix

One

to see

"onomatopoeia" or

have been used as a garbage bin for words

labels

to

analyze.

antonym problem was

creative investigation into the

out by

Gebels (1969). Fifty Australian students were asked

22 pairs of antonyms according

from -12
set

list

have refused

linguists

still

statement.

of this

"imitative"

bonus.

a paralinguistic

as

guessing and

manufactures words, which even when

culture

the truth

merely substituted

causal to "better than chance"

as

word length

how each

It

to

semantic poles ranging

They were given runs consisting of an antonym

to +12.

translated into 5

languages: Old Hebrew, South Malaita, Kiwai,

Tongan, and Finnish. Subjects were not told

were antonyms nor what were

a given sequence of 10

He found

that

their

that

any of the words

meanings when asked

to

sort

words of 5 languages.

matching occured beyond the .05 chance level

and argued that a positive relation existed between the structure

of the language and cognitive processes.
the

antonym

class

was

that

it

carried

The only requirement

for

words of a sensory nature

162
because,

predicted,

as

antonyms would arrange

"sense-expressing

themselves around opposite semantic poles on a phonetic scale"

(Gebels

1969:311).

Gebels also notes that each sound symbolism experiment

incomplete

in

itself

because the data

it

is

purports to explain has not

been fully described. He says, "only with a large number of cross-

which are supposed

cultures

linguistic

perceive the world

to

(Whorf), can the hypothesis of the existence of phonetic

differently

symbolism across two or more contrasting

be

supported."

(Gebels

One such
Russian

subjects

nonsense forms

linguistc

1969:312).

study was done by Crockett (1970).

were asked
to

communities

to

Two

hundred

rank phonotactically Russian

semantic scales involving brightness, size, and

mood. Diffuse vowels were associated above chance with

compact vowels with voluminosity. Acute vowels

smallness,
carried

bright and

gay meanings, grave vowels possessed dark and

unhappy connotations.

Finally,

was considered universally

the

voiced velar stop phoneme /g/

be large and dark (Crockett

to

1970:112).
Crockett remarks that

words

as

it

would be

better to regard imitative

primary sound symbolism and the values deriving from

word associations within

specific

language as secondary sound

symbolism. In some cases, the value of the sounds from the

domain mute those from the second. Crockett

these

secondary associations

(Crockett

1970:113).

may become

also

first

argues that

linguistically

diffused

163
Where

the secondary sound

symbolism might

pool of features perceived by speakers

experiment touched upon

this

is

unclear.

lie

within the

An

interesting

and retested college students with

23 pair of Brown, Black, and Horowitzs' (1955) antonym

list

(Kunhira 1971). Japanese antonyms were presented visually,

monotone

voice,

significant in

and with an expressive voice. Correct pairing was

all

cases,

but also significantly greater for the

expressive voice

mode

of presentation.

expressive voice

the

happy-sad,

the

Interestingly

enough, for

up-down, and good-bad

79%

antonyms were guessed above 79%. For the monotone, above

was

fast-slow,

(Kunhira

in

and

was

this

true

for the printed

form as well

1971:428).

This study indicated that prosodic elements play a parallel

role

to

structural

phoneme elements

directing

in

cognition

toward

sound-meaning associations. While suggesting such, the

experiment again uses a word

list

full

of sound symbolic words

from Japanese. The English words are never considered for

sound symbolic value, as

suggest should be done for English as a

whole, by sheer word volume in Chapter

compare

a false

set

to

further

it

Further,

it

does not

phonemes

as

does obtain. Again, the results are difficult

analyze.

Contrasting antonym word

studies

III.

of antonyms with contrasting

control for the results

their

list

experiments are seen in other

examining specific contrasting antonymic concepts

to

sounds. Tarte and Barritt (1971) ran three forced choice

experiments consisting of a number of

CVC

English trigrams and

164

most often preferred for

/u/

They found

or angular drawings.

elliptical

preferred.

least

triangles,

The vowel

/a/

vowel

the

that

next most preferred,

was also most preferred

/u/

round figures. Not surprisingly, the vowel

was preferred

/i/

was

/i/

and
for
for

smaller items and the vowel /a/ for larger ones. Consonants were
not consistently tagged to either type of figure.

They remark
undetermined,

that there

factor

figures

to

to be

"some, as

yet,

which permits monolingual adult native

American speakers of English

sounds

seems

to

agree on the assignment of vowel

(Tarte and

of different size"

1971:168).

Barritt

Tarte and Barritt chose the continua of vowels from /a/ to /u/ to
/i/

represent large-to-small

to

oral

cavity

size,

low back-to-high

terms of tongue position, and compact/grave to

front in

diffuse/acute

in

terms

of distinctive

experiment (Tarte and Barritt

features

1971:168).

theory

for

Their results

their

suggest that

with this method, the vowel continua was shown associated with
the

dimension. Therefore, they argue, the phonetic continuum

size

are also related for native

/a-u-i/

astutely remark,

factors

are

"what

is

not clear

They

speakers of English
is

whether any or

causative in producing these results"

of these

all

(Tarte

and Barritt

1971:168).

follow up study was undertaken by Tarte (1974).

Monolingual English and Czech speakers were asked

trials

of

CVC

nonense words and geometrical figures. In

both English and Czech subjects chose the vowel

figures,

/i/

to rate

with

small,

/u/

with ellipses, and

/i/

/a/

this

252
case,

for large

with triangles

165
(Tarte

1974:92). In a quick retest, Tarte replaced one

By

within his 9 word nonsense trigram.

same

were obtained except

results

muted and

that

replacing
the

/s/

experiments suggest

it

with

/g/,

the

dimension was

size

shape dimensions amplified (Tarte

the

phoneme

1974:94).

His

should not be unusual to obtain agreements

of these sorts from related languages. Both English and Czech are

Indo-European.

Symbolism Experiments

Natural Lexicons in Sound

The works of Sapir and Jespersen sparked

experiments

phonemic

early

the

in

stimuli.

In

1930's.

Few

a flurry of

included natural lexicons as

Tsuru and Fries study (1933), 25 pairs of

English words were paired with their corresponding words in

Japanese. Only a few of these words were antonyms, however, and
the entire
4.

sweet-bitter,

good-evil,

5.

1.

bird-worm,

fast-slow,

high-low,

13.

are-are not,

clear-muddy,

22.

18.

14.

10.

blue-yellow,

19.

23.

a tremor in

old-young,

15.

the arbitrary

20.

16.

8.

12.

hot-cold,

big-small, 21.
24.

right-

1933:283).

and 2 Japanese words

Tsuru and Fries found that up

translated.

peace-war,

soft-hard,

crooked-straight,

In their presentation of 2 English

meanings could be correctly

3.

square-circle,

11.

thick-thin,

wrong, 25. sharp-dull (Tsuru and Fries

their subjects,

7.

far-near,

kite-boat,

enemy-friend,

red-green,

2.

white-black,

6.

praise-deprecate,

9.

smart-dull,
17.

included:

list

Of

to

75%

itself,

this

to

of the
result caused

hypothesis for linguists of the day. Later

166
showed

criticism

that

"marked" terms

Japanese and English corresponded

in

the

vocabulary of

and that perhaps subjects were

only guessing according to overall word length. In binomial choice

tests,

merely using word length could easily allow guessing

50%

higher than

at

rates

or chance levels.

Probably the most cited sound symbolism study using natural

lexicons
study,

was done

21

pairs

in

1955 by Brown, Black, and Horowitz. In their

of words were matched

with corresponding words

from Hindi, Chinese, and Czech languages. The

under two conditions: the words had

to

list

was formed

be of a sensory nature, and

two, their frequency had to be at least 100 in one million words

used (Brown, Black, and Horowitz 1955:389).

Eighty-six subjects correctly guessed meanings of three
foreign language groups of antonyms twice as often

as

they were

wrong. The highest rate was for English-Hindi, followed by EnglishChinese and English-Czech. In further analysis,
significant

difference

it

was shown no

between male and female responses.

existed

However, introducing expressive quality of voice when

pronouncing

test

items did

make

difference.

Somehow,

pronouncers could iconically introduce haste into the "fast" word,

or sharpness into the "sharp"

word (Brown, Black, and Horowitz

1955:391). For the monotone condition, choices correct above a

chance level of guessing

(blunt-sharp),

at

.01

include: Eng-Czech, Eng-Hindi

Eng-Chinese, Eng-Hindi

(bright-dark),

Eng-Hindi

(down-up), Eng-Chinese, Eng-Czech (fast-slow), Eng-Chinese, Eng-

Czech

(hard-soft),

Eng-Chinese (light-heavy), Eng-Hindi (one-

167
many), Eng-Czech (thunder-lightning), and Eng-Chinese (widenarrow) (Brown, Black, and Horowitz 1955:391).

These experiments raised more questions than they answered.

For one, the idea that sound symbolism effects should be apparent
in

antonyms

is

supposed, but

not proven.

is

Word

length,

while

apparent earlier to Tsuru and Fries (1933), cannot be controlled

when using
categories

lexicon data

a natural

more

liable

set.

Exactly what makes various

correct guessing

to

is

unclear,

though

expressive voice in a vocal presentation does enact augmenting

influence.

Finally,

their

list

considered non-auditory and non-

is

onomatopoeic, but many of the pairs are indicated

sound symbolism outlined

argue that sound symbolism

languages. They argue
increasing in

its

it

Chapter

in

significant

The

results

In

all

can either be decreasing or actually

series of

experiments were run by Maltzman,

scheme was

to

English-Croatian, English-Japanese,

Croatian words. The

(1933).

Regardless, the authors

an important sub-segment of

is

Morrisett, and Brooks (1956). Their

Fries

types of

scope in present languages.

Another similar

guessing of 25

III.

in

list

was

the

same noted above

the English-Japanese

and Japanesein

at

Tsuru and

.001

indicated the English equivalents

words can be selected with

the correct

and English-Croatian

choices above chance were obtained

"clearly

test

trials,

probability.

of Croatian

frequency significantly beyond

chance expectancy, and quite as effectively as the selection of

English and Japanese equivalents"

Brooks

1956:250).

(Maltzman, Morrisett, and

However, Croatian-Japanese and Japanese-

168
Croatian presented pairs were far below any

statistical

significance

level.

Their results called them to doubt the "gestalt organization of

trace

systems,

(Maltzman,

and the physiognomic language"

Morrisett,

and Brooks 1956:251). Once again, the experimental

method

flawed because the word

is

the

reflect

auditory

or

Tsuru and

chosen does not clearly

list

subject of study. Are the concurrence due to gestural

synaesthetic
Fries'

list,

associations?

kinesthetic

or

extrapolation

is

or

With the

impossible.

Retests of experiments by Brown, Black, and Horowitz (1955)

and Maltzman, Morrisett, and Brooks (1956), were done by

researchers

Brackbill

Little

Essentially

(1957).

meanings of word
English-Hebrew,

Hebrew

for

pairs

of English-Japanese, English-Chinese,

Chinese-Japanese,

Chinese-Hebrew,

and Japanese-

items.

The

subjects were able to guess at better than chance rates of

English-Hebrew (53%), Japanese-Chinese (54.8%), and

Japanese-Hebrew (52.3%) (Brackbill and

the

duplicating

experimental designs, they asked subjects to guess the

earlier

.01

and

other studies,

1957:318).

Little

subjects guessed chance rates

for

Unlike

English-

Japanese (50.3%), English-Chinese (49.9%), and below chance rates

for

Chinese-Hebrew (48.1%). Visual presentation of word

aided English-Foreign guessing,

pairs

made no

difference.

Though

their study

modes of data

the

same

for

pairs

Foreign-Foreign word

used auditory, visual, and audio-visual

presentation,

and

this

is

commendable, the

169
experiments return weak inference for strong data.

Exactly what

allows better than chance guessing on word pair testing

They note
of

that

word

lengths,

unclear.

is

vowel and consonantal types, spacing

compound words, and connotation markedly influenced

agreement of subjects as
and Little

(Brackbill

to

the sameness of

the

meaning of word

pairs

This remark implicates prosody,

1957:318).

consonantal and vowel distinctive features,

and even

graphically

iconic factors as causal to the results of their guessing behaviors.

Other researchers did not view these results with equal

Brown and

disfavor.

Morrisett,

Nuttal

(1959) regarded the Maltzman,

and Brooks 1956 word

non-antonym items, such

as

with suspicion.

list

"when",

Consequently, a 36 item antonymic

"first",
list

"this"

above chance for

all

included

and others.

was created

Chinese, and Hindi. Their subjects matched correctly

significantly

It

for English,
at

levels

groups. They also achieved

extreme significance when the native language English was paired

with

words versus foreign-foreign pairs (Brown and Nuttal

foreign

1959:445).

(1958) to get

at

word-meaning

fitness.

1.

break, 2. clean, 3. cool, 4. cut, 5. rush,

9.

teach, and

differential
fit

its

was used by Wertheimer

different nonparametric approach

10.

scale

carried

connotative

write.

Ten words were presented:

belong, 7. knee,

6.

Each was ranked on two

according to whether

meaning and on

meaning was

it

scales:

8.

sun,

a semantic

contained sounds which

a bipolar scale as to whether

angular-rounded,

weak-strong,

its

rough-

170
smooth,

active-passive,

relaxed,

wet-dry,

Not

small-large,

fresh-stale

good-bad,

cold-hot,

(Wertheimer

1958:413).

.01

or binomial expansion,

Apparently

level.

than

qualities

words

fitting

non-fitting

were also asked

meaning

words

surprisingly, "on each of the ten scales, the fitting

were rated more extremely. Whether measured by the

test,

tense-

to

the difference

li-

beyond the

words have more clear-cut emergent

(Wertheimer

words"

indicate

significant

is

^-test,

when

1958:413).

word

particular

lost

Subjects

its

being continuously shown on a screen. The fitting

after

took longer for their implicit cultural meaning to

all

disappear in

condition

of saturation

This study suffers in that

it

fails

(Wertheimer 1958:414).
to

measure precisely what

allows a word to be rated more fitting to a group of assigned

sounds than another.

Sensory words apparently carry more

information than other kinds of words, but

sorts

all

of qualities

and meanings might be assigned objects and concepts according

the activities or contexts in

would

the

object

it

to

which they are found. For example,

word "key" be considered

often causes a clicking

"fitting"

because as a metal

sound when used, and therefore

should contain a stop? Or should the sounds be fitting because a

"key"

is

a small object and, following regard in other sound

symbolism

studies,

Even
stimulus

with

is

PET

so, the

it

should carry a small front vowel

delay of loss of meaning evaluation of a word

interesting.

It

should be possible,

if

it

is

so,

scans whether different or more long term

of the brain

are

/i/?

utilized

for various

to

measure

memory

areas

sound symbolic vocabularies.

171
Further,
the

such

PET

brain channels

vocabularies

to

may

scans
the

enable researchers to note whether

memory

different

and "referential"

of "affective"

areas.

Another interesting experiment, similar

(1958),

is

Wertheimer

to

Miron (1961). Here, 76 American monolinguals and 41

Japanese speaking bilinguals were asked to rank nonsense

words created from

a consonant

differential

7 point scales.

are actual

words, and for

and vowel matrix on 16 semantic

great

this

CVC

many

reason,

of these

CVC

combinations

consider this experiment

involving a natural lexicon for both groups.

Using rank order correlations

to

determine similarity of

groups, he found that Japanese and English speakers agreed most

heavily for the evaluative dimension (.89),

(.74),

and the activity dimension

significant at the

the

.01

level.

the

potency dimension

All correlations are

(.64).

Miron remarks

that

this

reliable

means

"two language groups use similar semantic dimensions"(Miron

1961:626).

To note Miron's

findings: A.) the highest ranked evaluative

vowel and consonant were; English,

/p/:

the

least

/a/

highly ranked evaluative

English /u/ and

/g/;

Japanese

/e/

and

and

/p/;

Japanese

least highly

and

vowel and consonant were;

/g/;

B.)

the

potency vowel and consonant were; English /a/ and

/q/ and /g/: the

/i/

highest ranked

/g/;

Japanese

ranked potency vowel and consonant

were; English /i/ and /p/; Japanese /e/ and /s/ and; C.) the highest

ranked activity vowel and consonant were; English /u/ and /n/;

Japanese /a/ and /n/: the least highly ranked activity vowel and

172
consonant were; English /i/ and /c/; Japanese /a/ and /p/ (Miron

1961:628-629).
Miron's study can be criticized for the choices
its

the vowels

which

e,

a,

o, u],

made

in

consonant and vowel matrix. He chose

selecting constituents for

[i,

it

is

fine because Japanese only

He

contains 5 vowels, but the consonant choices are problematic.

and

includes two affricates [c

nasal

What

[n].

evaluated

exactly

More

two stops

makes any of

unclear through

is

s],

clearly, both

this

the

[p

and

phonemes

/p/?

Is

Japanese and English subjects rank the

it

the voicing distinction?

the /g/ being

suffers

due

to

unit

of sound

the

phonemes

two

in

is

/p/

it

is

not systematically chosen.

phonemes

phoneme

list.

in

stop

being front? Miron's study

symbolism mechanics. This

are

it

the position in the oral

assumption that the phoneme

an

consonantal
his ten

Or

back and the

frequency than occurence

results

What makes

than both their highest picks, the voiceless bilabial

cavity,

rarest

significantly

design.

voiced velar stop /g/ as "bad" or "undesirable".

"uglier"

and one dental

g],

universally,

is

the

significant

problematic because
Affricates

are

the

and here, he includes

probably 5 times larger a

This

is

real

world languages. The means the

cannot be easily viewed through a universal perspective

the face of vast

numbers of

affricateless

in

tongues.

Nevertheless, he states the front vowels and consonants seem

to refer to

"pleasant" and

"weak" things, the back vowels and

consonants to "unpleasant" and "strong" things for Japanese and

English speakers (Miron

1961:630).

In

itself,

these

remarks concur

173
with trends seen in a large number other languages. His remarks
lead

toward a hypothesis that a vocabulary of purgatives and

emetics compiled from any language in any culture should contain

back features with

and

"unpleasant"

sound-meaning

"strong"

connotations.
In an ingenious experiment, 48 English speaking

Hawaiian

four-year olds were tested for sound symbolism (Roper, Dixon,

Ahern, and Gibson 1978). Researchers compiled words for loud,

soft,

large,

and small categories from Hawaiian, French, Spanish,

and English. These words were pronounced

or she

was allowed

or large

to take

to

each subject and he

a small black, large black, small white,

white token to represent what was heard.

Their results indicate a relationship between token choice and

word category. All subjects associated

denoting words.
large

Interestingly,

where the reverse was

a large

token with loudness

males preferred black as

true

for

females.

soft

and

Although white

tokens were associated with small and soft words, the results were
opposite for Hawaiian words (Roper, Dixon, Ahern, and Gibson

1978:95).
This study

is

significant in that

differences of sound and color/size

categories.

CVC
is

is

It

words

to

weakest

in

loudness.

girls

demonstrates gender

associations

measure token choice against. As

choose

in

for

specified

word

not having created a series of nonsense

the unit of perception and

boys and

it

it

is

it

stands, the

word

unclear which linguistic features

labelling a certain color or size or

174
Symbolism Experiments

"Goodness-of-Fit Sound"

made choices

Subjects in sound symbolism experiments often

between a diverse array of phonetic and semantic examples. Even

when

their

choices

indicated

must be taking place,

selective

their reasons

experiments were designed with

wished

selection

procedures of sound

An

were unclear. Many

for success

problem

this

explicate those cognitive

to

meaning-sound associations

mind. They

in

arenas encasing the linguistic

symbolism.

early study into the perceptual processes underlying

choice was labelled "The fitness of signs to words" (Hall 1952).

Thirty-four males and female subjects were to rank sets of 5
to
art,

50 words. The words included examples such

as:

fear,

signs

madness,

energy, help, tradition, visionary, and so on. Each set of 5 signs

was composed of two

or three conventional

signs

with meanings

associated with the test word and two or three which were

purposely vague. For example, the

test

word, energy, had one sign

choice that was a line drawing of the sun.

Results showed that while there was great agreement upon a
particular sign for a word,

there

grounds for popularity" (Hall

any figure that

is

was "no absolute and consistent

1952:23).

He remarked

that

"almost

not purely arbitrarily connected with a word

may, by some subjects, be likened

was considered, from

the evidence,

to

some associated

that

it

object, but

was not always

the

capacity of the figure to suggest an object that was primary in

influencing choice"

(Hall

1952:23).

it

175
Hall's study

not because

important with respect to sound symbolism, but

is

connects any sounds to any clear-cut types of

it

concepts. Instead, the results lead to a suggestion that subjects use

a

variety of

schemes

fitting

in

qualities for concepts,

signs,

and

words. There appears also to be quicker responses from subjects

where

less

apprehension

"for several words, the

appropriate

is

agreement

in

states

that

choice of a sign as the most

but the type of fittingness varies. While

high,

of obvious conventional

signs

He

present about choices.

is

significance are chosen,

it

some

seems

probable that those signs which combine the formal qualities of

simplicity

and regularity of design with a familiar structural

appropriateness

to

verbal

the

are

setting

both

quickest choices

the

and the "best symbols" (Hall 1952:31).

Sound symbolic words may be

excellence.

If

this

is

true,

the "best symbols" par

quicker retrieval times for sound

symbolic words would be predicted. Unfortunately,

this

has not yet

been done and Hall's study reaches only general conclusions. The
sign

drawings are not applicable to any other sound symbolic

visual

experiments.

figure-sound
In a slightly

more streamlined

(1958) tested 37 college students on a sign-word

semantic differential.
justice,

serene,

ten

word

McMurray

retest of Hall,

list

was

test

created;

using the

rhythm, wrong,

storm, philosophy, visionary, crisis, peace, and

enthusiasm. Each of these words was paired with two sign drawing

modified after Hall (1952). Subjects were then asked to pick one or
the

other

signs

to

best represent the

word.

176
The 10 words and

the

20 drawings were then rated with 15

polar adjectives on 7 point intervals to obtain the semantic

differential

Subjects described a word or sign as angular-

scores.

rounded,

weak-strong,

cold-hot,

good-bad,

slow,

hard-soft,

tense-relaxed,

ugly-beautiful,
"the

mean

be closer to the

mean

McMurray concluded
found

to

rough-smooth,

of the non-chosen

signs"

active-passive,

kind-cruel,

heavy-light,
or

green-red,
ratings

fast-

sick-healthy.

of the chosen

ratings of the

high-low,

signs

were

word than were those

(McMurray 1958:312).

This brief experiment demonstrated that connotative

meanings for words and signs can overlap when there

The kinds of

similarity

Part of the reason

of words.

It

is

cannot be deduced from

this

is

experiment.

the choice of signs and the other part

would be no surprise

to

find

similarity.

is

choice

worldwide consensus

representing a basic term with a distinct sign

in

and parallel semantic

connotations. For instance, given the test word, snake,

believe a

high probability exists that most subjects would not choose a circle

over an S to represent

it.

Further,

predict that in using the

semantic differential adjective scales, no one would be surprised

if

snakes assumed connotations including ugly, cruel, cold, green,

angular, fast, active, and sick.

Another
15

women

test

use of the semantic differential had 15

men and

rank 360 words upon 20 bipolar adjectival scales

(Jenkins, Russell, and Suci

1958).

The same

test

was regiven

to

540

subjects later as a measure of coding reliability. These tests were

largely designed to test the use

of the semantic differential for

177
solving

word association problems. As such,

sound symbolism

is

its

importance to

Results indicate that

unstated.

all

words do not

have equal connotative value and that meanings and sound cluster
on occasion. The manner of these occasions awaits further
investigation.

For example, one of the bipolar adjectival pairs of the

semantic

differential

was cruel-kind. When over 500 subjects

ranked 360 words on a scale where l=:cruel to 7=kind, interesting

evaluative

similarity

was seen

for

certain

words but not

The following words had mean rankings below

2,

(or a

others.

"very cruel"

connotative judgement): abortion (1.6), anger (1.8), bad (1.6), cold

criminal (1.6), deformed (1.9), devil (1.3), discomfort (1.9),

(1.8),

mad

fraud (1.8), grief (1.9), hate (1.2), heartless (1.2), hurt (1.4),
(1.8),

nasty (1.6), pain (1.7), putrid (1.9), rage (1.9), scalding (1.5),

severe (1.8), sin (1.9), starving (1.4), sword (1.7), thief (1.8),

tornado (1.3), trouble (1.9), war (1.2) (Jenkins, Russell, and Suci

1958:695-699).

By

itself,

words according
But

if

that

their

"cruel"

the

list

to

the

only leads one to note that subjects use

meanings they perceive them

to

contain.

words are analyzed, do they carry certain features

opposites the "kind" words do not?

The opposite pole words ranked above

6,

(or very kind) on the

cruel-kind scale include: baby (6.3), beautiful (6.1), calm (6.3),

clean (6.1), comfort (6.3), doctor (6.3), faith (6.3), farm (6.0), flower

happy

(6.3),

god

(6.2),

lenient (6.2),

(6.1),

(6.6),

heal (6.1), holy (6.5),

home

(6.3),

joy

loveable (6.7), mild (6.0), minister (6.1), mother

178
music

(6.5),

(6.2),

nice (6.5), nurse (6.4), peace (6.7), puppies (6.4),

relaxed (6.0), sister (6.0), sky (6.2), sleep (6.1), sunlight (6.3),

sweet (6.1), and trees (6.2) (Jenkins, Russell, and Suci 1958:695-

699).

very simple hypothesis about these two groups

the fashion

is

of this dissertation.

can be

immediately desirable

in

suggested the "cruel"

words should contain more back consonants

It

and stops than the "kind" words. Other experiments have found
this

tendency, so the hypothesis

Conversely,

conservative.

is

be supposed front consonants should be found more

can

it

"kind"

in

words than "cruel" ones.

tally of the

26 "cruel" words find

that 6/26

(23%) words

back consonantal velars and 3/26 (11%) contain

contain

glottals,

22/26 (85%) contain stops, and finally 8/26 (31%) contained

frontal

bilabial

consonants and 25/26 (96%) contained dentals.

Contrarily, of the 32 "kind"

velars,

words; 7/32 (22%) words contain

4/32 (13%) contain glottals, 21/32 (66%) contain stops,

15/32 (47%) contain bilabials, and 26/32 (81%) contain dentals.

For

this

cursory further analysis of Jenkins, Russell, and Suci data,

the emergent
the

"cruel"

word category

"kind" words. Also,

"kind"

carries

twice as

many

stops

as

words contain more frontally

produced bilabials than "cruel" words. Back consonant appear

even frequency for "cruel" and "kind" word

in

an

lists.

This type of quick analysis of semantic differential testing

results

is

flawed because the phoneme groups composing the

words are not mutually exclusive.

follow-up

test

to

correct this

179
was performed on 342 male

1000 English words on a number of

Subjects ranked

(1966).

The

bipolar scales.

Navy by Heise

enlistees in the U.S.

test

items included frequently used and short

words of English. Each word contained one of the 45 phonemes of

was opposed

English. Each

in

comparison with a

As

did not contain that one phoneme.

connotative potency,

evaluation,

of words which

list

a result, ratings according to

and activity were derived for

every phoneme in English.

Unlike previous experiments done by Miron (1961), Heise

found

Igl

was considered both "good" and

much

Nevertheless,

symbolism

studies.

being potent:

[a, a,

of the data corroborated

k, r,

s].

sound

earlier

The "un-potent" ones were:

[v,

r,

9].

Phonemes most

['9,

'sr,

g,

as

1,

y].

"Un-active" phonemes were:

The phonemes connotatively responded

[g,p,v].

(Heise 1966:23).

The following phonemes were agreed upon

Highly active phonemes were

[1,0].

"soft"

to as

"good" were:
dl (Heise

indicated as "bad" were: [au,

1966:18-19).
This study

fails

associate with which

is

the printed

controlled.
"hard",

is

which distinctive sound features

meaning evaluations. Since the

unit of testing

word, word length and rhyming effects cannot be

For example, because

Igl

is

considered

the deciding factor the voicing feature?

words makes
easily

to identify

it

impossible to answer this question.

"soft"

and

Using natural
It

could just as

be a connotation created by the angles found in the

orthographic /k/ and lacking in

the

Igl.

/k/

180
Finally, Heise's subjects

male and

their

situation

as

were a

They were

select group.

may have

inductees

all

influenced their

decisions on which sounds were to be most important in their

impending indoctrination

into the U.S.

Naval Corps. As such,

this

study can be used as illustrative of a speaking group's connotative

assignments of meaning to sound.

in

language examples versus

real

though they differ

in

the results

Synaesthetic

Studies

is

It

symbolism

in

act

language examples,

artificial

they produce.

into

Sound Symbolism

Scholars since antiquity have noted sound

with various sensory perceptions.

how phonemes

not clear

Synaesthesia

which words, phonemes, and

easily associable

is

is

their

type of sound

structural

elements attach to identities involving colors, smells, shapes,

tastes,

and even temporal perceptions. Like other sound

symbolism experiments,

debate rages

capacities to consistently

categorize

over whether the

sounds according to widely

disparate

senses

There

evidence pointing in both directions.

is

An

is

universal or culturally

early study

was done by Odbert, Karwoski, and Eckerson

(1942). For this study, 243

various classical

Wagner's Fafnir,

and language specific.

students listened to

10 selections of

works including Stravinsky's Sacre du Printemps,

Sibelius'

Second Symphony, and so

students were to rate each piece according to

on.

The

10 sets of adjectives.

These categories included, for example, category A;

spiritual,

lofty.

181
awe-inspiring,
F;

or category

serious

merry, joyous, gay, happy, cheerful, bright.

subjects had rated these 10 selections they were asked

Once
to

solemn, sober,

sacred,

dignifies,

imagine that each selection was a color. Responses were divided

according to whether the subjects reported

seeing

thinking

colors,

feeling colors, or forced their color-sound choice (Odbert,

colors,

Karwoski, and Eckerson

1942:157).

Color was rated on three continua. Subjects could report

spectral characteristics,

They could

purple.

white,

gray,

such as red, orange, yellow, green, blue, or

also

indicate

intensity

medium

mixture, or dark

responses.
Their most striking result was that the peaks on

measures of vision
the

mood

follows:

orange,

that

all

three

were reported varied systematically with

of the selection (Odbert, Karwoski, and Eckerson

1942:161-163).
as

through

gray and black, and black. Finally, hue saturation

could be described with light mixture,

mixture

of brightness

The

classical

tender-blue,

music selections matched sound-color

leisurely-green,

vigorous and exciting-red,

Karwoski, and Eckerson

They

exciting-

solemn and sad-purple (Odbert,

1942:163).

indicate their study

musical selection number.

gay-yellow,

One

was limited because of the small

selection of "sad"

music did not

cover the range of somber tone, nor did a couple of lively pieces

cover
tonal

all

"happy" musical possibilities. Also,

features

lead toward

it

seems unclear what

which color association. What makes a

182
certain piece of

music "leisurely"?

Why

upon the mood of the piece, but also

Even

so, their data

its

can people not only agree

color?

corroborate Berlin and Kay's system of

color term universals (Berlin and

Kay

1969).

Odbert, Karwoski,

In

and Eckerson's study, subjects use basic color terms

levels

at

higher

than color terms which are peripheral to languages

worldwide. For example, red, blue, and yellow use as descriptors

were much commoner than purple, orange, and pink.

In

the

schemes outlined by Berlin and Kay and modified by Witkowski

and Brown (1977), the pink, brown, purple, and orange color terms
are the last

words for basic colors added

a language's lexicon.

to

Another synaesthetic study measured

association differently.
the

start

that

some

subjects

experiment exploited

Osgood (1942)
visualizers

Since

this

it

this

was evident

to

sensory-sound

researchers from

were vivid "photistic" visualizers, one

phenomenon. Karowski, Odbert, and

collected drawings from

subjects

who were good

during the presentation of music and from a control

group.

The

first

group was given colored pencils and the control

group merely one gray pencil. Essentially, both groups drew the

same types of

figures (Karwoski, Odbert, and

This strikingly apparent event led them to

notion
relating

that

culture

contains

and transmits

Osgood 1942:212).

test

further

common

upon the

analogies

both to sight and sound.

Their testing created a Group Polarity Test which compared

musical moods with visual adjectives. The music moods of bad.

183
depressing, heavy, happy, bass, alive, relaxed, loud, fast, and

harmony were
down,

angular,

thick,

moving

on semantic differential scales for large,

rated

dark,

blurred,

They remark,

adjectives.

"in

far,

crooked, background, and

almost every case the

way

majority of subjects related the words in the same

had related the qualities" (Karwoski, Odbert,

visualizers

photisitc

and Osgood

that

1942:213).

In their discussion, these researchers point out that

standpoint

of

remains

causality

experimental

particular

this

Individuals

unclear.

may

design,

from the

synaesthetic

acquire

mood-music-

color associations from their cultural experiences, as a result of

natural

associations

from large animals,

tones

(bass

small, and so on), or from

some

sort of "unity of the

treble

from

senses" neural

network (Karwoski, Odbert, and Osgood 1942:213).

What
describing

they do present

the

because many
Principle

synaesthetic
later

tests

is

a set of 5 principles helpful in

phenomenon. These are

reinspect their claims. First,

instructive
there

is

the

of Polarity. They state that "in color-hearing every

quality of sound or sight implies

its

opposite" (Karwoski, Odbert,

and Osgood 1942:216). The "Janus-like" words so noted by

linguists,

color and

"those that look at once in both directions", are part of

hearing perceptions.

are used in

order to

When

color or auditory

adjectives

any language, the existence of one implies the other. In

know

light,

dark must be understood.

Next, the Principle

pair of opposites

of Gradients

may come

to

argues that "in color-hearing a

represent extremes of a continuum,

184
in

which intermediate steps are represented" (Karwoski, Odbert,

and Osgood 1942:217). This

is

especially true, they note,

mood

fuzzy polar adjectives, such as those describing

imagery,

are

or inner

used.

Third, the Principle

"in

when

of Parallel Polarities and Gradients

states

color-hearing a linkage of an auditory pole with a visual pole

implies a linkage of their opposites. Gradations along an auditory

continuum may be paralleled by gradations along a visual

continuum" (Karwoski, Odbert, and Osgood 1942:217). Music which
off in high pitch

starts

may be

represented as bright, but as

slows down, the photism becomes darker. Effects such as

it

can

this

be confounded by overlapping polar associations. For example, a

connection between thick and heavy might be linked with one

involving

Odbert, and Osgood

homonymic polysemy
word
its

heavy-deep,

thick-heavy,

light

and

deep-dark

(Karwoski,

1942:218). In comparison, English shows

for

some of

these

synaesthetic

means happy, lightweight, and

bright

events.

The

depending upon

context.
Fourth, the Principle

Gradients

states

that "in color-hearing not all

need be represented

in

the

Osgood 1942:219). Some

selection

within

of Alternate Auditory Polarities and

visual

subjects,

response"

aspects of the music

(Karwoski, Odbert, and

they note, respond to the entire

of music where others refer only to individual items

it.

Finally, the Principle

Gradients

of Alternate Visual Polarities and

avers that in color-hearing, any of a

number

of visual

185

may

polarities

be paralled with a given auditory polarity"

(Karwoski, Odbert, and Osgood 1942:219).

be referred to as thin-thick, bright-dark,

soft

These
demonstrate
within

and

have been

complexity

on

real

cultural

attempted

synaesthesia

cross-modality

informants, and data from Russian

backgrounds.

further

these lines. In a series of interviews with

Jakobson,

because they

listed

disentangling

of

human psychology from

experiments

When

far.

five principles

the

fast-slow.

or

subjects can also refer to loud

reporting about a soft-loud gradient,

as near,

soft-loud gradient can

analysis

events

Further

along

German, Czech, Serbian

and Dakota speakers, Reichard,

and Werth (1949) showed that color-audition was very

uneven between cultures. Jakobson noted

that

the

sound-color

equation might be expected to be particularly vivid and regular in

languages with a high degree of sound symbolism (Reichard,

Jakobson, and Werth 1949:230). To date, no study of
done. This did nothing to stop speculation.
there exists in the brain a

map

similar topographically

(Masson 1952:41). To

to

date,

has been

this

Masson suggested

that

of color contours part of which

map
there

of acoustic frequencies
is

little

evidence for

is

there

this.

review and retest of cross-cultural visual-verbal

synaesthetic tendencies

was carried out by Osgood (1960). Testing

took place upon 40 Navaho, 10 Mexican-American, 27 English, and

20 Japanese speakers. Subjects were presented a word such as

"heavy"

and asked to choose whether the term meant up-down,

vertical-horizontal,

and so on.

186
Results were an interesting mix. English speakers felt "heavy"

was down,

speakers saw

"heavy"

as

down, horizontal, hetereogeneous,

as

saw

crooked, blunt, and near; Japanese speakers

dark,

thick,

Navaho speakers saw "heavy"

dark, crooked, hazy, and large;

thick,

and near; Spanish

dark, concentrated,

thick,

colorless,

"heavy" as down, colorless, thick, dark, crooked, hazy,

concentrated,

large,

near,

and blunt (Osgood

1960:149).

Such an example does not display what Osgood actually

discovered. For the data,
against

13

all

are judged

"when 28 verbal concepts

different visual

alternatives

in

possible

all

combinations (364 items), approximately half of the items yield

evidence for consistent intra-cultural
1960:152).

His cross-cultural significance reached .05 for

language pairings. Anglo

vs.

(Osgood

synesthesia"

Spanish 72%, Anglo

and Navaho

vs.

vs.

vs.

Japanese

all

Navaho speakers agreed 65%, Anglo

Japanese 78%, Navaho

69% (Osgood

vs.

Spanish 61%,

1960:152).

These data are striking because they highlight both cultural

differences

and similarities. For example,

all

three

speaking groups

agreed "calm" was bright, but differed as to whether

or

straight

was

was down,

(English) or small (Navaho). All agreed 'heavy"

dark, and near, but differed as to whether

it

large

thick,

was crooked (English)

it

(Navaho and Japanese) (Osgood 1960:153).

Osgood's conclusions are merely the form of more hypotheses

about synaesthesia.

framework
denotative,

for

He

argues that there

humans, and

structural

this

is

is

common

connotative

buried under the weight of the

requirements of symbolic

language as

we

187

know

One

it.

type of synaesthesia

may be

reference to the red spectrum and

cold, for example. Another

dimension with

may

which

large,

warm and

its

increases in

increases in

The semantic

blue spectrum as

the

simply a characteristic of the

is

visual

loudness (Osgood

common

be learned, such as the loud

physical world that as any noise-producing

approached,

the

innate,

approaches or

object

is

angle are correlated with

1960:168).

differential

was used

Osgood's remarks

to test

by Pecjak (1970). Subjects included Americans, Hungarians, Turks,

Italians,

He

Belgians, Dutch, Germans, and Japanese.

for gray, red,

yellow, blue, green, and white with emotions and

days of the week.

He found

results

which

some general variables extend beyond

led

specific

him

to

a),

environment determines the nature of the world,

e.g.

large

is

warm,

red,

and

b).

ethno-graphic ranges (Pecjak

second
the day

is

cultural

conclude that

culture influence

verbal synaesthesiae. These can be of two kinds:

cold or blood

tested ratings

the

common

night

is

dark,

conventions crossing

1970:625).

Evidence for the

noted in that for Belgians, Dutch, Germans, and Italians

Sunday

also

meant white. This was not the case

Americans, Japanese, Turks, and Yugoslavs (Pecjak

for

1970:625).

Pecjak's study hardly disentangled innate from learned

synaesthesia.

He even remarked

that

the

denotative

meaning of

words may influence synaesthetic effects more than had been

thought because the different methods of measuring
(semantic
techniques

differential,
i.e.)

similarity

judgements,

and

to

meaning

association

do not correlate very highly (Pecjak 1970:626).

18^
Still,

study showed that an astonishing amount of association

his

between sound and other senses regularly occurs without much

among

conscious decision

speakers.

Further research by Marks (1975) has led researchers to

10%

believe that only

of the population has vivid color-sound

His retests led him to argue that the natural world

perceptions.

does bend cognition along parallel metaphoric paths among

pitch

categorized

are

1982).

and senses. Loudness

subjects

differeing

He argues

similarity

in

modalities

the

as

that

brighter

this

than

in

their

amplitude and higher

polar contrasts

(Marks

emanates from phenomenological

make-up of sensory experiences of different

(Marks

1982:177).

Williams took a different approach

to synaesthesia. In

undertaking an analytic study of English over the past 1200 years,

he

uncovered semantic laws which regulated sensory modal

(Williams 1976).

He argued

languages, regularly
to

another.

become
received

that English,

as well as

shifts

other

moves metaphors from one domain of sensing

For example, what was once touch (warmth) can

taste

(hot/spicy)

scant

attention

Summary

of

later

or

on.

To

is

his

impressive work has

corroboration.

Sound Symbolic Experiments

Sound symbolism experiments

symbolism. Each

date,

are never strictly about sound

an attempt to view the cognitive nature of

language use. As a whole, they have presented a loose behavioral

189
and social scheme

to

sound-meaning

arbitrary

which

explain results

This

hypothesis.

meaning assumption cloaked the

beliefs

the

strain

well-worn

sound-

"arbitrary"

about insufficiently

languages for decades.

studied

real synthesis has not

been done for the maze of sound

symbolism experiments or hypotheses known

Nevertheless,

speakers

studying

was

intent

their

of foreign

clear.

century.

this

These experiments, whether

languages and their

or native

responses to real or imaginary words, were designed to show

similarity

"structural

in

historically

unrelated

meaning" (Weiss 1964:456). The reason for

words of the same

this

that

is

"the theory

of phonetic symbolism does not specify that a single meaning

becomes associated with

single

sound, but rather that

many

meanings may become heirarchically associated with a sound, and

the

vise

versa,

may

differ"

Such
century to
principle

heirarchies

(Weiss

1964:456).

testing intents led

come

basic

of the arbitrariness

sensing

sound symbolism experiments

these conclusions:

to

absolute nor inviolable"

a

by different cultural groups

established

of language

(Ultan

(light-dark,

a.)

and inescapable

"the basic

symbols

this

is

neither

1978:551), b.) semantic concepts of

small-large

sweet-sour,

windy-calm,

and so on) and orientating nature (up-down, fast-slow, in-out,

near-far,

and so on) are regularly identified with phonetic

contrasts

universally,

c.)

some amount of sound symbolism

is

learned in early childhood universally, at least as part of

"motherese" and

its

affective bonding regime,

d.)

the

degree to

190
which sound symbolism expresses innate or genetically inherited
perceptions

is

unknown,

e.)

its

presence

numbers of words

in

some languages,

symbolism events

is

large

and

is

f.)

is

expressed as large

the scope of sound

pervasive in

many

languages.

These observations disallow independent invention or diffusion

explanations to account for

its

one disputes sound symbolism

cross-culturally

presence globally and,

is

and bio-culturally

a
it

g.)

finally,

no

meta-language process, and

allows

various

phono-

semantic decisions to be reached, allowing a speaker to find a best

mental

"fit"

for

sounds and communicative intents.

CHAPTER V
CONCLUDING REMARKS

Summary
This dissertation began by raising questions about the

Saussurean

arbitrary

sound-meaning

hypothesis.

Its

general

acceptance has been based upon anecdotal rather than rigorous,

systematically

produced evidence. Sound symbolism,

argued, should be examined carefully because of

an

importance to

its

language origins.

understanding of proto-languages and

violated the bounds of Saussure's linguistics because

human communication system should

close association of

meaning with

find

signs.

was

it

adaptive

The more

it

It

holds that a

value in
signifiers

a
that

can be placed in a referent symbol, such as a word, the more easily

any member

is

able to recognize that which

is

signified.

This simple argument has been revived recently in the works

of

Hewes

(1983),

LeCron Foster (1978), Wescott (1980c), and

Malkiel (1990a). Sound symbolism, they argue,

is

the logical bridge

between what must have been a rudimentary and highly gestural

language of

Homo

language of

Homo

63

erectus

and a more arbitrary sound-meaning

sapiens. For these reasons,

a conservative

hypotheses about sound symbolism was proposed

dissertation.

They were

tested

191

upon

in

this

geographically and

set

of

192
sample of languages

genetically distanced

measured with three nonparametric

These

statistical

and

II

tests.

were supplemented by a search for examples of

tests

sound symbolism

Chapter

in

world languages,

in

Sound symbolism was present

in

this

virtually

Chapter

detailed in

language phyla.

all

III.

Its

absence in some phyla was due to lack of research data and also
imprecise

sound symbolism definitions.

were scrutinized next

in

Sound symbolism

tests

Chapter IV. Their findings were found

more supportive of sound symbolism

as

a cognitive

universal than

were negative. As a whole, sound symbolism experiments

they

have not been incorporated into a unified sound symbolic scheme.

The inclusion of sound symbolism

adaptation
these

in

as part of the cognitive

human evolutionary

experiments

use

In contrast,

my

history

disparate

is

incomplete because

methodology.

sound symbolism experiment held

sub-phonemic unit carried meaning, not the utterance

that the

as

with

prosody nor the word or phoneme as with most other sound

symbolism experiments. Using the Chi-square

hypotheses about 16 glosses were significant

test,

at

23 of the 63

p<.05. The results

broken into gloss category showed ethnoanatomical

BREAST, TOOTH, NOSE, NECK,

significant

association

18 predictions
significant.

MOUTH,

the easiest to predict a

between feature and meaning. There were

for this group of

words and 9 or

The physiological words were next

They included:
and

made

and

DRINK.

words,

COUGH, VOMIT,

SPIT,

easiest

50% were
to

predict.

SUCK, EAT, SWALLOW, CHEW,

Thirty-three hypotheses were

made

for

this

group and

193
1 1

or

33% were

were the most

DOG,

significant.

final

group, semantically ancient,

WATER, FOOD,

For the words

dificult to predict.

12 hypotheses were

These

The

made and

results cause reflection

or

25% were

significant.

upon the use of ethnoanatomical

words when reconstructing distant language families.

number of

glottochronological

the

and

words commonly used

comparative linguistics are words for body

large
in

The high degree

parts.

between feature and meaning found here for

of association

ethnoanatomical terms should be a warning to linguists that more

than one feature and one gloss

between two languages, especially when that word

relationship

involves

necessary to indicate a

is

ethnoanatomy.

the

In regarding features, each feature tested

one time. The easiest

least

was

significant at

feature-meaning association

predict

followed by velar. The incidence significance for features

stop,

tested with Chi-square goes:

(1) palatal,

(2)

bilabial,

(1)

labio-velar,

alveolar.

to

was

(1)

(1)

resonant,

(6)

stop, (3) velar,

back vowel,
(1)

(1)

(2)

glide, (2) nasal,

fricative,

front vowel,

(1)

affricate,

and (1) dental-

These results indicate sound symbolic effects are not

limited to one

or

few features

as

some researchers have suggested,

although some features are more liable to use as sound symbolic

features. This tally also

shows

that

the

speaks well of the testing design because

it

hypotheses offered sound symbolism as a broad

phenomenon, not one

restricted

to

few sub-phonemic

features.

These 63 hypotheses were also combined into 240 hypotheses

so to run

the

Kruskal-Wallis and Jonckheere-Terpstra

tests.

For the

194
Kruskal-Wallis

were predicted

features

important in

velar,

bilabial,

test,

suggesting

at

affricate,

significant

Homo

of a unified proto-language spoken by

are place of articulation,

significant features
logical

owing

that,

These

levels.

primary

features

the

and glide

stop,

in

reconstruction

the

Two

erectus.

and

bilabial

the largely binary nature of

to

were

results

of the

velar.

It

human language,

proto-language was binary as well. Sounds most easily contrast

front,

articulations

bilabial

manners of

as

articulation,

with back, velar articulations.

stops,

glides,

predicted at significant levels. All

countless

mammalian vocal

indeed

the

at

in

present

language, more research

instance,

be better

are

the

rarest

known

as

stop feature joined

repertoires.

is

in

Further,

and affricates were

are recognized

However,

of the elaborate mental

basis

made

distinctions

three

is

if

features

they

in

are

and phonemic

languages and emanating from proto-

necessary to show

this.

Affricates, for

type of consonantal phoneme.

They might

"complex phonemes" because they contain a

with a fricative feature.

Affricates

could either

be arising or diminishing in regard to a proto-language era

according to this data, but their rarity remains unexplained.

Finally, the use of the Jonckheere-Terpstra test

my
a

63

showed

that

hypotheses predicted an astonishing amount of order upon

data set

that

should

have only shown random sound-meaning

associations.

Two

strengths are evident about the dissertation.

that the design

other scholars.

is

simple, though

Second, the design

it

is

The

first is

has never been done before by

easily replicable.

Skeptics only

195
need

put up their hypotheses, find

to

they

features

the dictionaries,

and

tally

the

predict.

Weaknesses

Theoretical

Research designs are limited because they explore the

unknown
this

to

century,

degree corresponding to existing knowledge. Until

few languages were known adequately,

enable a study of this type to be done.

that

the data sample

is

fault in

all

to

this research

The universe of human languages

small.

vast and at least 5,000 languages

extent

One

of

least

is

is

have been described. Besides, the

and intricacy of individual languages virtually defies

complete

description.

In the past thirty years linguistics has demonstrated the scope

of language and communication

specialties

is

no

have arisen including child language, language

development,

language

sociolinguistics,

origins,

ethnolinguistics,

psycholinguistics,

languages

of phonetic

is

variation

unknown, though
in

individually

as

paralinguistics,

and zoosemiotics. The

semiotics,

range of information transfer among individual

specific

Splintering

small.

less

vast.

speakers

of

Likewise, the range

described languages

is

incompletely known. This study draws from a pool of 229

languages, only

semantically

10 of the

basic,

The extrapolation of

17-24 debated language phyla, and

though culturally intended

results

proto-language phenomena

to
is

nonidentical

the 5,000 present

manner of

16

words.

languages and

statistical

faith.

to

196
The periods of human development

are large. Presumptory

conclusions about what was of importance

anytime

at

in

distant

eco-zones and within differing neurological templates are easy to

Nevertheless, this study contradicts a cherished belief of

reach.

modern
word
its

That

linguistics.

is,

major Saussurean tenet

an amotivational construct and

is

largely

is

is

that the

unconnected

in

connotation and denotation. The assumption that sounds and

meanings are entirely disconnected

doubtful.

is

The extent

sounds or features of sounds carry meaning in themselves

that
is

now

an open discourse. Larger samples than the one presented here

should corroborate the conclusions

reach.

Another weakness of the dissertation

sound symbolic scheme upon which

of hypotheses in

metaphor

is

As

pointed out in the

sometimes

I,

formed because of visual similarity

bilabiality, i.e.), gestural

acoustic mimicry

metaphor

Chapter

the lack of a unified

rank languages, or

to

individual instances of sound symbolism.

discussion

is

iconism

(COUGH

(WATER

(TOOTH

and stopping,

and labio-velarity,

sound-meaning

(NOSE

and dentality,

i.e.),

i.e.).

and
i.e.),

or kinesthetic

Which metaphor

is

used

by which cultural group probably varies as does which events

become

labelled in

such fashion. The similarities exposed in

my

conclusions point to complex naming schemes in each individual

language. Yet, no researcher has compiled a complete listing of

sound symbolic words, events, or guidelines for their recognition.

Indeed, the

norm

in

most dictionaries

such words according to principles

is

that

known only

the editor identifies

to

his

own

design.

197
Research

Future

human

Children enter
of

human

society probably sensitive to a variety

vocalization. There

specific

evidence that as they

is

acquire mastery of their language, they use a

number of sound

symbolic words, features, and techniques to enhance the

connection of meaning and sound inside their memory. Though
not entirely clear

why

language competence

at

children reach about the

same

same ages worldwide,

the

levels

part

it

is

of

of the

reason must be that they are exposed to language in structurally

ways

similar

may

It

in

all

societies.

be, as Bickerton (1990) suggests, that a type of proto-

language mechanism exists within

obvious

in

this

is

more

children than in adults because of the incomplete neural

development.

If

so,

sound symbolism

is

part of the birthright of all

and they recapitulate their phylogeny through verbal

children

play with
a

humans and

all

much
It

sounds and meanings

larger

predominantly

might also be

that

in

order to achieve mastery within

arbitrary

sound-meaning language.

humans contain

a tacit level of language

awareness which enables choices to be made on a less than

conscious level. Such a level, indicated

in

numerous language

experiments, suggests the cognitive use of language engages an

inner Gestalt

to

achieve parsimony.

Sound symbolic use and

a sound

symbolism vocabulary

deserve increased focus by scholars because they signal a doorway

into

an

understanding

of rudimentary

language,

language

198
development, and human cognition. Given
interesting

to

impetus,

this

it

would be

determine a sound symbolic vocabulary, along lines

similar to this dissertation, and apply

it

to a setting

number of mentally handicapped individuals need

working vocabulary.

sound symbolic vocabulary

develop

to
is

one

in

which

semantic intent of the words.

the sounds are rich in cues about the

Such a vocabulary might be more successful

language for such

which a

in

in

providing access to

handicapped individuals than other traditional

methods.
There

way

elicit

the possibility that the brain

times

provide different retrieval

to

items.

is

is

for

organized in such a

vocabulary

different

Research with positron emission tomography (PET) might

evidence that sound symbolic words cluster

in

various

specialized regions of the brain, or that they are a bilateral

phenomenon. Again, research

in

this

area

may

also

be usefully

applied to the mentally handicapped in order to develop a working

vocabulary.
There
systems.

It

is

increased interest in alloprimate communication

would not be surprising

elements in any of these systems.

calls

often

It

to
is

find

sound symbolism

already

known

that contact

involve high frontally produced sounds and threat or

danger warnings the reverse, back harshly produced sounds. Given

an more unified
systems,

scope of the vast alloprimate communication

sound symbolism may be seen vindicated as a form

intermediate

between

call

and

phoneme

structures.

199
Finally, as stated previously, studies into

unlock

proto-language

and indicate

concepts of early humans. Until

humans used
symbolism
traits

their

relates

of an object

identifying

vital

semantically

important

modern instrumentation,

sensory abilities as "scientific" probes.

closely

imbued

to

the

early

Sound

"doctrine of signatures", where the

that object with

its

powers. In

concepts in the distant past for humans, and

identifying them with the help of sound

better recognition

emerge. This

sound symbolism can

may

symbolic vocabularies, a

of their ability to cognitively parse events can

provide insight into the nature of mental

evolution which has taken place

among our

ancestors.

APPENDIX A
16 CONCEPTS

WORD LIST FOR

1

Language gloss

"breast"

of female, n.
1.

lAmharic

Afro-Asiatic

[tut]

lArabic [Gadi]

[ununa]

IBurji

IHausa

[mama]

ISomali [naaske:du]

4.Austronesian
4Maori

[rei]

4Palauan [tut]
4Tahitian [u]
4Tolai [au]

4Tonga [fatafata]
7.Dravidian

7Gondi [bo mi]

7Kolami [pom:e]

7Manda [may]
7Pengo [may]
7Tamil [mey]

8Tndo-Pacific
8Hiri

Motu

[rata]

SSiane [ami]

SFore [no no]

SGadsup [naami]
STairora

[maama]
9.Niger-Khordofanian

9Bini [ewuere]

200

201
9Bini [ewuere]
9Ife

[pmu]

9Igbo [ara]

9Mbukushu [dyere]
9Yula [hyr

le]

10. North

Amerind

11. South

Amerind

lOApache [ipe]
lOBlackfoot [mon:IkIs]

lOCakchiquel [^'um]

lOMicmac [pesgun]
lOHopi [pi:hu]

llQuechua [coco]

llAmahuaca

[xoci]

llCashibo [soma]

llChama [soma]
llChaninahua [pa^oti]
13.Nilo-Saharan
ISKanuri [teg am]

13Kaure [yfre]
ISErenga [juud]
13Mararit [slid]

13Tama

[ojut]

16. Sino-Tibetan

16Cantonese [nin]

16Lahu [cuni^]
16Mandarin [rufag]
16Tibetan

[nums]

16Newari [dudu]
17. Altaic

17Japanese

[mune]

17Korean [cut]

202

17Mongolian [oeboer]
17Turkish [gogus]

17Manchu [hunhun]
2 Language gloss "chew"

v.t.a.,
1.

v.t.

Afro-Asiatic

lAmharic [ahfiaka]
1

Arabic

[yam dag' u]

IBurji [lek'ans]

IHausa [tauna]
ISomali [alalinaya]

4.Austronesian
4Hawaiian

[mama]

4Palauan [meriget]
4Tahitian

[mama]

4Tanga [ug]

4Tonga [lamu]
6.

Indo-European

6Croatian [zvakati]
6Icelandic [lygyu]
6Pali [cabati]

6Rumanian [rumega]
6Spanish [maskar]

7.Dravidian

TKannada [avudu]
7Kui [muht]

7Kuwi [hok-]
7Pengo [muh-]
7Telegu [nemarueu]
S.Indo-Pacific
SHiri

Motu [ania]
9.Niger-Khordofanian

9Mbukushu [tahuna]

203
9Nyanga [kutafuna]
9Shilluk

[nyam]

9Swahili [tafuna]

9Xhosa [hlafuna]
10. North

Amerind

11. South

Amerind

lOChontal [di^ij'ma]

lOMicmac [alisgopgl
lONavaho

['as]

lOOjibwa [sa:sa:kotn]

lOZoque [wyajsu]

llAymara [turula]
llCavinena [nako]

llChaman [naka]
llJaqaru [cakca]

llTacanan [hobol
12.Uralic
12Finnish [pureskela]

15.Austro-Tai

15Khmer

[d.lar]

15Pearic [ke:t]

15Thai [k'io]
16. Sino-Tibetan

16Cantonese [jiuh]

16Gurung [geba]
16Lahu [be]

16Mandarin

[jiao]

16Newari [tap]
17. Altaic

ITJapanese [sosaku]

ITKorean [si^ipda]

17Manchu [nlyar]]
17Mongolian [zaglah]

204
nXurkish [eigne]
3

Language gloss "cough"

v.t.,v.i.
1.

lAmharic
1

Afro-Asiatic

[sal]

Arabic [su'aal]

IBurji [k'ufay]

IHausa

[tari]

ISomali [qufa'aya]

4.Austronesian
4Fijiian [vu]

4Indonesian [batuk]

4Nukuoro [kobe]
4Tahitian [mare]

4Tikopia [tare]
6.

Indo-European

6Czech [kalati]
6Hindustani [khansi

6Norwegian [hoste]
6Polish [kaszlek]

6Portuguese [tose]

7.Dravidian

7Pengo [kroki]
7Malto [inqe]

7Kurukh [iukhna]

7Manda [kruk-]
7Konda [kok-]
S.Indo-Pacific

SBagupi [doro-]
SBikol [abo]

SGarus [dalu^-]
SHiri

Motu [huahua]

8Kare [dagAl-]

9.Niger-Khordofanian
9Fula [d'oya]

9Igbo

[i

kwa]

9Mbukushu [dikohwera]
9Shona [kosora]
9Swahili [kohoa]
10. North

Amerind

11. South

Amerind

lOBlackfoot [sals:klna:]

lOChoctaw [hotilhko]
lOHopi [oho]

lOMicmac [nog eg]

lONavaho [dikos]

llAymara [k'ajaha]
llCashibo [?oko]

llChacobo [?oko]

llChama [oho]
llGuarani [hu'u]
16. Sino-Tibetan

16Cantonese [kat]

16Lahu

[ci]

16Lisu [tssctt]

16Mandarin [kesou]
16Tibetan [16k 9 p]
17. Altaic

njapanese [seki]
IVKorean [kic'im]

ITMongolian [xanad]
17Turkish [oksuruk]

ITKurdish [qoz]

205

4 Language gloss "dog, "jaguar", "fox", "animal", "deer"

I.Afro-Asiatic
1

Arabic [kalb]

[woccoo]

IBurji

IHausa [kare]

IHebrew [kalab'j
1

Somali [el]
3.Austroasiatic

3Alak [coo]

3Lawa

3Mon

[so'^]

[kl9]

SSouei [^acool
3 Vietnamese

[k'uyen]

4.Austronesian
4Fijian [koli]

4Hawaiian

[ilio]

4Indonesian [anjig]

4Kemak

[asu]

4Tolai [pap]

T.Dravidian

7Gondi [nai]

7Konda

[nukuri]

7Mayalam [nay]
TPengo [neku.r]
7Tamil [nay]
S.Indo-Pacific

SAmele

[pa]

SKare [kui]

SMawan [kwA:r]
SSihan [pAy]
SSilopi

[wAy]
9.Niger-Khordofanian

9Basa [gbe]

207
9Ife

[ads a]

9Igala [abla]

9Mbukushu [mbwa]
9Yoruba [ads a]
10. North

Amerind

11. South

Amerind

lOBiloxi [cuhki]

lOCrow [biegye]
lOHopi [puko]

lOMenomini [tihseh]
lOTzotzil [^'i?]

llAmahuaca [paihega]
llJaqaru [hai q'u]

llLenca [aguingge]
llTarascan [axuni] "deer" "animal"

llTotonac [kuri]

13.Nilo-Saharan
13Erenga [wui]

13Fongoro

[bisi]

ISSinyan [bisi]

13Tama

[wi]

13Yulu

[bisi]

16. Sino-Tibetan

16Atsi [khiii]

16Burmese

[khiji]

16Cantonese [gau]

16Mandarin [dou]
16Tibetan

[c'l]

Language gloss "drink"

v.t.i.,v.i.,

n.i.

I.Afro- Asiatic

lAmharic [tatta]
lArabic [yasrabu]

208
IBurji [d'uw]

IHausa
1

[sa]

Somali [aba]

3.Austro-Asiatic

3Cambodian [phsk]

3Lawa [hu^]

3Muong

[^og]

3Thin Pook]

3Vietnamese ['uer)]

4.Austronesian
4Fijiian

[gunuva]

4Indonesian [minuinani]

4Nukuoro [unu]
4Pascuense [unu]

4Tonga [inu]
6. Indo-European

6Albanian

[pi]

6Bengali [panio]

6French [bwar]
6Gaelic [dyoc]

6Lithuanian [gerti]

9.Niger-Khordofanian

9Mbukushu [kunwa]
9Ndebele [-naGa]

9Shona [cekunwa]
9Swahili

[nywa]

9Zulu [p'uza]
10. North

lOBIackfoot

[si

ml]

lOCakchiquel [kum]

lOHopi [hiiko]

lOOjibwa [minikwe:]

Amerind

209
lOSquamish

[taq']
11. South

Amerind

llHuitoto [yirode]
[u plan a]

llQuechua

llReseigaro [-i^du]

llTotonac [k'ota]
llTupi [uu]

13.Nilo-Saharan
13Erenga

[lifo]

13Fongoro [auw]
13Kara [ay a]
IBMerarit

[fa]

13Mileri [liyo]
16. Sino-Tibetan

16Cantonese [yam]

16Gurung [Guba]
16Mandarin [he]
16Newari [Iwone]
16Tibetan [tun]
17. Altaic

17Japanese

[nomu]

IVKorean [ma si]

17Manchu [omimbi]
17Mongolian

HTurkish
6

[o:'r]

[ic]

Language gloss

"eat" v.t.a., v.i.

I.Afro-Asiatic

lAmharic [balla]
1

Arabic [ya'kulu]

IBurji [it-]

IHausa

[ci]

ISomali [naya]

S.Austro-Asiatic

3Cambodian

[sii]

SChaobon [caa^]

3Lawa [som]

3Mon

[cea'']

3Vietnamese [ag]
4.Austronesian
4Fijiian [kai]

[makan]

4Indonesian

4Kemak

[a]

4Maori [haupa]

4Tonga

[kai]

S.Indo-Pacific

8Awa [nono]
8Bena Bena [na-]
8Fore [na-]

SKamano-Yagiria [no

8Rao [mi]
9.Niger-Khordofanian

9Mbukushu [kudya]
9Ndbele [-dla]

9Shona [-dya]

9Xhosa [-tya]
9Zulu [dla]
10. North

Amerind

11. South

Amerind

lOBlackfoot [o:wat]
lOChorti [we']

lOHopi [noosa]
lOKwakiutl [hemx^i d]

lOOjibwa [miicimaw]

llAmahuaca [cocoquin]
llGuarani

[u]

210

21
llJaquaru [palu]

llTotonac [hua]
llTupi [umbau]

13.Nilo-Saharan

BErenga

[rjgAn]

13Fongoro [us a]
13Merarit [sin]
ISMileri [QAn]

13Tama [gan]
15.Austro-Tai
15Chrau

[sa]

15Katu [ca]

15Mon

[g']

ISPearic [ca]

ISSedang [ka]
16. Sino-Tibetan

16Cantonese [sihk]

16Gurung [cab
16Mandarin

a]

[cf]

16Newari [khan]
16Tibetan [see]

Language gloss"food"
I.Afro-Asiatic:

lAmharic [mabsl]
1

Arabic [ta'a:m]

IHausa [abinsi]
ISomali [unlo]
IBurji

[it

ay]
2.

Australian

2Aranda [amirna](vegetable only

e.g.)

2Diyari [puka](vegetable only e.g.)

2Gumbaynggir [yul'a]

212
2Dhuwal

[n'aGa]

2Wailbri [magari]

3.Austro-Asiatic

3Mon [ksna^]
4.Austronesian
4Hawaiian

[hiai]

4Indonesian

[makanan]

4Tagalog [pagkain]
4Tolai [nian]

4Tonga

[kail
6.

Indo-European

6Czech [potrava]
6French [alima]
6Hindi

[k'

ana]

6Lithuanian [maistas]

6Russian [eda]
7.Dravidian

7Telegu [era]

7Tamil [unti]

7Toda [un]
7Tulu [uta]
7Brahui

[irag']

9.Niger-Kordofanian

9Ewe [nudlidu]
9Ndebele [ukudla]

9Bobangi

[boli]

9Swahili [cakula]

9Xhosa [ukutya]
10. Amerind-North

lOBlackfoot [ao:wahsIn]

lOChontal [galg'ejuaw]

lOCrow

[ba:ru;k]

lOHopi [nuva]

213

lOMicmac [ma man]

1

1.

Amerind-South

llAymara [mankka]
llGuarani [tembi'u]
llTupi [miu]
llHuitoto [ecagoi]

llTotonac [tahua]
16. Sino-Tibetan

16Mandarin

[si

rwu]

16Tibetan [saja]

16Newari [ann]
16Cantonese [caan]
17Altaic

nJapanese [sokumotsu]
17Turkish [y

IVKorean

z]

umsik]

nUzbek [owkat]
ITAzerbaijaini [xuraek]

Language gloss "mouth"

3.Austro-Asiatic

3Jehai [tansd]

3Kensui [hah]

3Mon

[paig]

SSemaq

Beri [ksnut]

3 Vietnamese

[mi eg]

4.Austronesian
4Fijiian [gusu:na]

4Kemak

[i:borro]

4Malayan [mulut]
4Ponapean [ahu]
4Tagalog [bibig]

T.Dravidian

VKannada [kaltu]

VKonda

[gadli]

7Malayalam [karuttu]
7Tamil [karuttu]
7Tulu [kantelu]
S.Indo-Pacific
8Hiri

Motu [udunal

8Kare [kase-]

SManit [egere-]

8Rao [dototno]
SSilopi [owe-]

9.Niger-Khordofanian

9Bobangi [munye]
9Igbo [onu:]

9Mbukushu [kanwal]
9Shona [muromo]
9Sango [yanga]
lOCakchiquel

10. North

Amerind

11. South

Amerind

[ci']

lOHopi [mo'a]
lOKwakiutl [sems]

lOMenomini

[to:n]

lOMixtec [yuhu]

1 1

Aymara

[laka]

IJaqaru [simi]

UOuyana

[yipota]

llBotocudo [himpma]
1

llnga [sim]

13.Nilo-Saharan
13Erenga [kul]

13Fongoro [tara]

214

215
13 Kara [ta]

13Merarit [^awl]

IBNubian

[agil]

16. Sino-Tibetan

16Tibetan
16Lisu

[k'a]

[manA]

16Lahu [mags]

16Akha [mobef]

16Burmese [meisei]
17. Altaic

nJapanese [kuci]
17 Korean [ip]

17Kurdish [detn]
ITTurkish [agiz]

nUzbek

[orrjz]

9 Language gloss "neck"

I.Afro- Asiatic

lAmharic [angal]
1

Arabic ['unuq]

IBurji

[marmari]

IHausa [wuyu]
ISomali [lukunta]

3.Austro-Asiatic

SKhmu?
3Kuy

3Mon

[kak]

[tskoor)]

[ka?]

3Souei [takoog]
3 Vietnamese

[ko^]

4.Austronesian
4Fijiian [do

mo]

4Hawaiian

[a:i]

4Indonesian [leher]

216
4Maori [hakii]

4Ponapean [kasag]
7.Dravidian

7Konda

[gadli]

VKota [kartl]
7Kuru>k' [k'es]

7Mayalam [karutu]
7Tamil [karutu]
S.Indo-Pacific

SGirawa [pstu]

SMunit [ha]
SMurupi [gutnara]

8Nake

[fA:-]

8Rao [bagrs]
9.Niger-Khordofanian

9Mbukushu [eiggo]

9Mvumbo

[tsiug]

9Shona [tnutsipa]
9Swahili [ku]

9Zulu [Iggila]
10. North

Amerind

11. South

Amerind

lOJacaltec [nuk]

lOKwakiutl [k'uk'un'a]

lOMicmac

[j

hag an]

lONavaho [akos]
lOZoque [kAkA]

UAymara [kunka]
llCavinena

llChama

[e:piti]

[e:piki]

llGuarani [aju]
llHuitoto [kimaigo]

15.Austro-Tai
ISBriou [takog]

15Chrau

[r|ko]

15Katu [luar]
ISPearic [k3:k]

ISSedang [krbk]
16. Sino-Tibetan

16Burmese

[le]

16Cantonese [geg]
16Lisu [kstsi]

16Mandarin [bwodz]
16Tibetan [smgul]
10 Language gloss "nose"

n.,

n.i.

I.Afro-Asiatic

lAmharic [afsnca]
lArabic [manahir]
IBurji [suna]

IHausa [hansi]
ISomali [san]

3.Austro-Asiatic

3Alak [muh]

3Cambodian [cramoh]
3Vietnamese [muy]

3Muong [muy-]
3Lawa [maah]
4.Austronesian
4Fijiian

[uku:na]

4Hawaiian [ihu]
4Indonesian [hidug]

4Maori [ihu]
4Tagalog

[ilog]

217

S.Indo-Pacific

8Gal [no-]

SGumalu

[nit-]

SKare [neme-]

8Rao

[ra:t9]

SSihan [msde-]

9.Niger-Khordofanian

9Ewe
9Igbo

[got!
[i

mi]

9Shona [mhuno]
9Swahili [pua]

9Xhosa [impumlo]
10. North

Amerind

11. South

Amerind

lOSquamish [ma'qsn]
lOQuiche [txa'm]
lOHopi [yaqa]

lOMicmac [sigon]
lONavaho

['aci

h]

llHuitito [dofo]

llQuechua [singa]
llReseigaro [-hitako]

llTotonac [quincan]
llTupi [tin]

13.Nilo-Saharan
13Erenga [mi si]

13Tama

[a

mil]

13Runga [mondu]
13Bora Mabang [boji]
13Mileri [misi]

15.Austro-Tai
15Chrau [muh]
15Katu [moh]

218

219
15Pearic [mstot]

15Sedang [mbh]
15Thai [ya:lmuk]
16. Sino-Tibetan

16Newari ['nas]
16Tibetan [nskuli]

16Mandarin

[bi

dz]

16Cantonese [beih]

16Burmese [hna]
11

Language gloss

"spit"

or

v.t.

1.
1

v.i.

etc.

Afro-Asiatic

Arabic [busaaq]

IBurji [tuf]

IHausa [tofa]
1

Somali [anduuf]

lAmharic [laffa]
S.Austro-Asiatic
3 Vietnamese

[fun]

SKensiu [bej]
3Kintaq [bej]

3Bateg

[tsf]

3Temoq

[6 oh]

4.Austronesian
4Hawaiian [kuha]
4Indonesian [ludah]

4Manam [mwar|o]
4Tahitian [tub a]

4Tonga

[a'a'nu]
6.

6Hindustani [0uk]
6Icelandic [spyta]

6Lithuanian [yiesmas]

Indo-European

6Pali

220

[bhuhesike]

6Rumanian [pa mint]

9.Niger-Khordofanian

9Mbukushu

[61 pa]

9Ndebele [k'afula]

9Shona [-pfira]
9Swahili [tema]

9Zulu [p'umisa]
10. North

Amerind

11. South

Amerind

lOCrow [^U9]
lOHopi [toha]
lOKwakiutl [kwis^id]

lOMicmac [lusgwalign]
lOSquamish [pa'xn]
llGuarani [udqvu]
llHuitoto [tuanole]

llQuechua [tucana]
llReseigaro [choo]

llTotonac [cujmak'an]
12.Uralic
12Finnish [sylkea]

12Hungarian [pokni]

IS.Nilo-Saharan

13Twampa

[t'ak']

15.Austro-Tai
15Briou [kucoh]

15Chrau [choh]
15Katu [kalwiq]
ISPearic [chu:s]

15Sedang [ka'cow]
16. Sino-Tibetan

16Cantonese [tou]

221
16Mandarin [tut an]
16Tibetan [lupa]
17. Altaic

nJapanese [^ubaki]
IVKorean [c'impaet']
17Turkish [tukur]

[tupur-]

ITAzerbaijaini

12 Language gloss "suck"

v.t.,

v.i.

etc.

I.Afro-Asiatic

lAmharic [tnattata]
1

Arabic

[yamussu]

IBurji [t'unt']

IHausa [cotsa]
ISomali [nuugayya]

3.Austro-Asiatic

3Kensiu [jshud]

3Temiar [jod]
3Semai [no:^]

3Semaq

Beri [sok]

3Bateq Nong

[jot]

4.Austronesian

4Kemak

[tnus]

4Manam

[sir]]

4Maori [mo mi

4Tahitian [ote]

4Tonga [huhu]
6. Indo-European

6Icelandic [syuga]

6Lithuanian [ciulpti]

601d English [sucan]

6Pali [cusati]

6Rumanian [suge]

7.Dravidian

7Toda

[ixc-]

7Tamil [un]

7Kota [un]
7Telegu [kuducn]

7Kuwi [undali]
9.Niger-Khordofanian
9Igbo [ira]

9Mbukushu [yamwa]
9Shona [svetu]
9Swahili [fyondu]

9Xhosa [ncanca]
lOBlackfoot

10. North

Amerind

11. South

Amerind

[s:ta:]

lOCrow [daci]
lOIxil [^'ub']

lONavaho

['eesto't]

lOWinnebago [wi:kom'
llCashibo [cucuka]

llMarinahua [coco]
llShipibo-Conibo [^oyo]

llTacanan [coco]

llChacobo [coco]
13.Nilo-Saharan

13Miza [o-ndro]
130jila [ndro]

13Logo [ndro]
ISLugbara [ndru']
13Lokai [ndro]
17. Altaic

nJapanese [suu]
17Korean [bal]

222

223
17Manchu [jembe]
IVTurkish [em]

17Uzbek [simip]
13

Language gloss "swallow"

v.t.,v.i.

1.
1

Afro-Asiatic

Arabic [yabtaliu]

IHausa [ha'diya]
ISomali [liquaya]

[d'eem-]

IBurji

lAmharic [wala]

4.Austronesian
4Fijiian [liloma]

4Hawaiian

[iale]

4Indonesian [teguk]

4Nukuoro [holo]
4Tahitian [horotnii]
6.

Indo-European

6 Albanian [kaptoy]

6Bengali [khoao]
6Croatian [gutati]

6French [avaye]
6Lithuanian [ryti]

7.Dravidian
7Tamil [virukku]

TKodagu [mugg-]
7Telegu [mringu]

7Konda [erg-]
7Toda

[irk-]

S.Indo-Pacific
8Hiri

Motu [hadonoa]

8Kare [arjgAn-]

SGirawa [ni^'ane-]

224
SMunit [kurtiys-]

SKamba [unub-]
9.Niger-Khordofanian

9Ewe [minu]
9Igbo

[i

16]

9Mbukushu [mina]
9Shilluk

[mwoni]

9Zulu [gwlga]
10. North

Amerind

11. South

Amerind

lOChoctaw [balakaci]

lOCrow [apahik(y)]
lOTzotzil [bik']

lOMohawk [atskahu]
lOYokuts [meeki]
llGuarani [moko]

llQuechua [miypuna]
llHuitoto [cicode]

llTupi [umocone]

llTotonac [hua]
12.Uralic
12Finnish [niela]
16. Sino-Tibetan

16Newari [gras]
16Cantonese [tan]
16Tibetan

[mTkeuu laan]

16Gurung [k'lxyoba]

16Burmese [tnyou]
17. Altaic

17Japanese [notnikomu]

17Turkish [yutma]

17Korean [samk'i]

ITManchu

[nur|]

225
14 Language gloss: "tooth"

I.Afro- Asiatic

lAmharic [tars]
1

Arabic [asnaan]

IBurji [irk 'a]

IHausa [haic'ora]
1

Somali

[iligl

3.Austro-Asiatic

3Alak [canah]

3Cambodian [tmih]

3Kuy [kanssy]

3Mon

[ns]

[nan]

3 Vietnamese

4.Austronesian
4Hawaiian [naniho]
4Indonesian [gigi]

4Tonga

[nifo]

4Palauan

[ui

ngel]

4Tahitian [niho]

S.Indo-Pacific

SAngoram
8Hiri

[sisig]

Motu [isena]

8Kare [ogo-]

SMunit

[ai-]

8Rao [traga]
9.Niger-Khordofanian
9Bobangi

[lino]

9Dogon [tonu]

9Mbukushu [dyegho]
9Zulu [izinyo]
9Swahili [jino]

lOZoque

10. North

Amerind

11. South

Amerind

[tAJo]

lONavaho ['awo']

lOMenomini [pet]
lOChontal [lahay]
lOBlackfoot [mohI:kIn]

llAymara

[k'aci]

llHuitoto [izido]

llQuechua [quiru]
llTotonac [tatzan]
llTupi [ainha]

13.Nilo-Saharan
13Bora Mabang [sat^ik]
ISMasalit [k a cine]

13Merarit [r|or|od]

13Runga [sAdi]

13Tama

[rjiit]

15.Austro-Tai
15Briou [kaneig]

ISChrau

[se'c]

15Katu [kaniag]
15Pearic [kho:y]

ISSedang [haneq]
16. Sino-Tibetan

16Burmese [swe]
16Cantonese [gah]

16Mandarin [ya]

16Maru

[tsoi]

16Tibetan [so]

15

Language gloss "vomit"

227
v.t.,

v.i.

I.Afro-Asiatic

lAmharic [asm alas a]

1

Arabic [qay'l

IBurji

[huusad']

IHausa [amai]
1

Somali [yux(w)9t]

3.Austro-Asiatic

3Vietnamese

[6i]

3Kensiu [ka^]

3Semai [ke:?]

3Temoq [ku^]
3Sre [ha?]

4.Austronesian
4Indonesian

m un

ah]

4Manam [kulena]
4Nukuoro [hagaku]
4Tagalog [sumuka]
4Tahitian [tu:tu:]
6.

Indo-European

6Czech [zvraseti]
6Dutch [braken]
6French [vomir]
6Nepali [okeunu]

6Norwegian [kaste]
S.Indo-Pacific

SBagupi [pa-]

SGirawa [?es-]
SKare [pas a-]

SPanim [buhade-]
8Hiri

Motu [tnumuta]
9.Niger-Khordofanian

9Bobangi [lua]

228
9Ndebele [hlanza]
9Swahili

[kokomoa]

9Mbukusu [ru0al

9Ewe [dexe]
10. North

Amerind

11. South

Amerind

OBiloxi [kna]

OCrow [kara]
OKwakiutl [gokwaia]

OOjibwa [sikakowe:]
OTzotzil [k'sb]

IReseigaro [i^kapu]

ITotonac [irp'atlanan]
IGuarani [gue'e]
IHuitoto [ctcuede]

IJaqaru [ahri]

15.Austro-Tai
5Briou [kuta]

5Chrau [hoq]
5Katu [kata]
5Pearic [chsQul]

5Sedang [he a]
16. Sino-Tibetan

6Atsi [phat]

6Cantonese [gau]
6Lisu [pe^]

6Mandarin

[tii]

6Tibetan [clikps]

Language gloss "water"

n.i.,

n.

I.Afro-Asiatic

Amharic [wsha]
Arabic [maa']

229
IBurji

[waa]

IHausa [ruwa]
1

Somali [biyyo]

3.Austro-Asiatic

3Brao [daak]

3Bru [daa?]

3Mon

[dac]

3Muong [dak]
3 Vietnamese

[niak]

4.Austronesian
4Fijiian [wai]

4Kemak

4Manam

[bi:a]

[dag]

4Tahitian [vai]
4Tolai [tava]

S.Indo-Pacific

8Bemal

[ze]

8Bena Bena [nag a mi]

SFore [wani]

SGende [nogoi]
SSihan [va]

9.Niger-Khordofanian
9Igbo

[mmi

ri

9Shona [mvura]
9Swahili [maji]

9Ndebeie [amanzi]

9Xhosa [amanzi]
10. North

lOChoctaw [ficak]
lOChorti [ha]

lOHopi [paaliu]

lOMicmac [samgwan]

Amerind

230

lOKwakiutl ['wa:p]
11. South

Amerind

llHuitoto [jainoi]

llJaqaru

[uma]

llQuechua [yacu]
llTotonac [c'ucut]
llTupi

[i]

13.Nilo-Saharan

13Bongo Bagirmi [mane^

13Erenga [kaal]

13Fongoro [aiAn]
ISKara [mana]

13Logo

[yi]

16. Sino-Tibetan

16Burmese [yei]
16Cantonese [seui]

16Gurung [kyu']
16Mandarin [swei]
16Newari [na]
17. Altaic

17 Japanese [mizu]

ITKorean [mul]

17Manchu [muke]
17Mongolian
17Turkish

[us]

[su]

APPENDIX B
SUPPORTING DICTIONARY REFERENCES FOR
1.

I-Amharic-Semitic::Leslau,

W.

16

GLOSSES

Afro-Asiatic
1976. Concise Amharic Dictionary:

Amharic-English: English- Amharic.

Weisbaden: O. Harrassowitz.

S. 1983. Handbook of English-Arabic for

Bombay: Oxford University Press.

l-Arabic-Semitic::Shaikh,
Professionals.

l-Burji::Sasse, H.J.

1982.

An Etymological Dictionary of

Burji.

Hamburg: H. Baske.
l-Hausa-Chadic::Bargery, G.P.

1934.

A Hausa-English Dictionary

and English-Hausa Vocabulary. London: Oxford University

Press.

l-Hebrew-Semitic::Ben-Yehuda, E.&Weinstein, D. 1964. Ben

Yehuda's Pocket English-Hebrew, Hebrew-English Dictionary.
York: Washington Square Press.
1-Somali-Cushitic:: Abraham, R.C.

Somali. London: University of

2.

1966. Somali-English,

London

New

English-

Press.

Australian

2-Aranda::Yallop, C. 1977. Alywarra, An Aborigine Language of

Central Australia. Canberra: Australian Institute of Aboriginal
Studies.

2-Dhuwal::Holmer, N.M. 1983. Linguistic Survey of Southeastern

Queensland. Sydney: Australian National University.
2-Diyari:: Austin, P.

1981.

A Grammar of Diyari South

Cambridge: Cambridge University Press.

231

Australia.

232
2-Gumbaynggir::Dixon, R.M. 1979. Handbook of Australian
Languages. Canberra: Australian National University Press.
1979. Dictionary of Wilbri Language. Sydney:

2-Wailbri::Reece, L.

University

of Sydney.
3.

Austro-Asiatic

3-Alak::Huffman, F.E. 1977. An examination of lexical

correspondences between Vietnamese and some other
Asiatic languages. 43 :171-198.
3-Bateg::Benjamin,

G.

subgroupings and

Austroasiatic

1976.

Malay Peninsula.

Austro-

In Austroasiatic

Studies. Edited
by P. N. Jenner Thompson,L.C., and Starpsta,S. 37-128. Honolulu:
University of Hawaii Press.

prehistory in the

subgroupings and
Studies. Edited
by P. N. Jenner Thompson,L.C., and Starpsta,S. 37-128. Honolulu:
University of Hawaii Press.

3-BateqNong::Benjamin, G.

prehistory in the

1976.

Malay Peninsula.

Austroasiatic

In Austroasiatic

3-Brao::Huffman, F.E. 1977. An examination of lexical

correspondences between Vietnamese and some other
Asiatic languages. 43 :171-198.

Austro-

3 -Bru:: Huffman, F.E. 1977. An examination of lexical

correspondences between Vietnamese and some other AustroAsiatic languages. 43 :171-198.

1974. A Concise Cambodian-English

Oxford University Press.

3 -Cambodian:: Jacob, J.M.

Dictionary. London:

3-Jehai::Benjamin,

G.

1976.

Austroasiatic

Malay Peninsula.

subgroupings and

Studies. Edited
by P. N. Jenner Thompson,L.C., and Starpsta,S. 37-128. Honolulu:
University of Hawaii Press.
prehistory in the

3-Kensui::Benjamin, G.
prehistory in the

1976.

In Austroasiatic

Austroasiatic

Malay Peninsula.

subgroupings and

In Austroasiatic

Studies.

Edited

233
by P. N. Jenner Thompson,L.C., and Starpsta,S. 37-128. Honolulu:
University of Hawaii Press.

3-Khmer::Huffman, F.E.&Proum, I. 1978. English-Khmer Dictionary.

New Haven: Yale University Press.

3-Khmu?::Huffman, F.E. 1977. An examination of lexical

correspondences between Vietnamese and some other AustroAsiatic

languages.

43

3-Kintaq::Benjamin, G.

:171-198.

1976.

Austroasiatic

Malay Peninsula.

subgroupings and

Studies. Edited
N. Jenner Thompson,L.C., and Starpsta,S. 37-128. Honolulu:
University of Hawaii Press.

prehistory in the

by

In Austroasiatic

P.

3-Kuy::Huffman, F.E. 1977. An examination of lexical

correspondences between Vietnamese and some other AustroAsiatic languages. 43 :171-198.
3 -Lawa:: Huffman, F.E. 1977. An examination of lexical
correspondences between Vietnamese and some other AustroAsiatic languages. 43 :171-198.

3-Mon::Shorto, H.L. 1962.

Oxford University Press.

Dictionary

Of Spoken Mon. London:

3-Muong::Huffman, F.E. 1977. An examination of lexical

correspondences between Vietnamese and some other AustroAsiatic languages. 43 :171-198.
3-Semai::Benjamin, G.

1976.

Austroasiatic

Malay Peninsula.

subgroupings and

Studies. Edited
N. Jenner Thompson,L.C., and Starpsta,S. 37-128. Honolulu:
University of Hawaii Press.
prehistory in the

by

In Austroasiatic

P.

3-SemaqBeri::Benjamin, G.

1976.

Malay Peninsula.

Austroasiatic

subgroupings and

Studies. Edited
by P. N. Jenner Thompson,L.C., and Starpsta,S. 37-128. Honolulu:
University of Hawaii Press.
prehistory in the

In Austroasiatic

234
3-Souei::Huffman, F.E. 1977.

correspondences
Asiatic

An

examination of lexical

between Vietnamese and

languages.

43

3-Sre::Benjamin, G.

some other Austro-

:171-198.

1976.

Austroasiatic

Malay Peninsula.

subgroupings and

Studies. Edited
N. Jenner Thompson,L.C., and Starpsta,S. 37-128. Honolulu:
University of Hawaii Press.

prehistory in the

by

In Austroasiatic

P.

subgroupings and
Studies. Edited
by P. N. Jenner Thompson,L.C., and Starpsta,S. 37-128. Honolulu:
University of Hawaii Press.
3-Temiar::Benjamin, G.

prehistory in the

Austroasiatic

1976.

Malay Peninsula.

In Austroasiatic

3-Temoq::Benjamin, G. 1976. Austroasiatic subgroupings and

Malay Peninsula. In Austroasiatic Studies. Edited
by P. N. Jenner Thompson,L.C., and Starpsta,S. 37-128. Honolulu:
University of Hawaii Press.
prehistory in the

3 -Thin:: Huffman, F.E. 1977. An examination of lexical

correspondences between Vietnamese and some other AustroAsiatic languages. 43 :171-198.

3-Vietnamese-Vietmuong::Dinh-Hoa, N.
Dictionary. Tokyo: Charles E. Tuttle
4.

1966.

Company

Vietnamese-English
Publishers.

Austronesian

1979. A Fijian and English and an English

and Fijian Dictionary. London: Sampson, Low, Marston, and

4-Fijian::Hazelwood, D.

Company.
4-Hawaiian::Pukui, M.K.&Ebert, S.H. 1957. Hawaiian-English
Dictionary. Honolulu: University of Hawaii Press.

4-Indonesian::Eschols, J.M.&Shadily, H. 1975. An EnglishIndonesian Dictionary. Ithaca: Cornell University.

4-Kemak::Stevens, A.M.
:32-38.

1967.

Kemak: An Austronesian Language.

235
4-Maori::Biggs, B.&Reed, A.H.&Reed,A.W. 1966. English-Maori

Sydney: Wellington.

Dictionary.

4-Nukuoro:: Carroll, V.&Soulik, T. 1973. Nukuoro Lexicon. Honolulu:

University of Hawaii Press.

4-Palauan::McManus, E.G. 1977. Palauan-English Dictionary.

Honolulu: University of Hawaii Press.
1960. Diccionario y Gramatica de la lengua
J.
Pascua, Pascuense-Castellano Castellano-Pascuense.
Haven: Yale University Press.

4-Pascuense::Fuentes,

de

la

New

Isla

De

4-Ponapean::Rehg, K.&Sohl, D.G. 1979. Ponapean-English Dictionary.

Honolulu: University of Hawaii Press.

4-Tagalog::deGuzman, M.O. 1966. An English-Tagalog, TagologEnglish Dictionary. Manilla: G.O.T. Publishers.

4-Tahitian::Davies,

J.

1978.

Tahitian and English Dictionary.

Chicago: University of Chicago Press.

4-Tanga::Bell, F.L.S. 1977. Tanga-English English-Tanga.

Sydney:

University of Sydney.
4-Tikopia::Firth, R.

University

1985. Tikopia-English.

4-ToIai::Franklin, K.J. 1962. Tolai

Territory of

Auckland:

Auckland

Press.

Papua

New

Guinea:

Language Course. Ukarumpa,

Summer

Institute

of Linguistics.

4-Tonga::Churchward, CM. 1959. Tongan Dictionary. London:

Oxford University Press.
6.

Indo-European

6-Albanian::Mann, S. 1957. English-Albanian Dictionary. London:

Cambridge University Press.

236
6-Bengali::Dabbs,

J. A.

1962.

Short Bengali-English, English-

A&M

Bengali Dictionary. Austin:

College of Texas, Department of

Modern Languages.

New English-Croatian,
York: MacMillian Publishing

6-Croatian::Bogadek, F.A. 1944. Cassell's

Croatian-English Dictionary.

New

Company.
6-Czech::Cermak, A. 1963. English-Czech, Czech-English.

Saphrograph

New

York:

Company.

6-Dutch::PrickvanWely, F.P.H. 1971. Cassell's English-Dutch, DutchEnglish Dictionary. London: Cassell and
6-French::Girard, D. et

Dictionary.

New

al.

1973. The

Company

New

Cassell's

Ltd.

French

York: Funk and Wagnall's.

6-Gaelic::Macalpine, N. 1955.

A Pronouncing Gaelic-English

Dictionary. Glasgow: Alexander

MacLaren and Sons.

6-Hindi-Urdu-Iranian::Craven, T.&Chitambar, J.R. 1932. The New

Royal Dictionary: English into Hindustani and Hindustani into
English. Lucknow: Methodist Publishing House.
6-Icelandic::Bogason, S.O.

1966. Icelandic-English

Icelandic Dictionary. Reykjavik:

6-Lithuanian::Lalis, A.

1915.

and English-

Isafoldarprentsmaija H.

F.

Dictionary of English and Lithuanian

Languages. Chicago: Leituva.

6-Nepali::Burrow, Y.&Emeneau, M.D. 1961. A Dravidian
Etymological Dictionary. Oxford: At The Clarendon Press.

6-Norwegian-Germanic::Scavenius, H.&Berulfsen, B. 1979. McKay's

Dictionary. New York: McKay Company.

Modern English-Norwegian

6-01dEnglish::Jember, G.K. 1975. English-Old English, Old-EnglishEnglish Dictionary. Boulder: Westview Press.
1955. English-Pali Dictionary. Colombo,
Ceylon: The Pali Text Society.
6-Pali::Mahathera, A.P.B.

237
J.

1988.

New

York:

6-Polish::Stanislawski,

English Dictionary.

McKay's English-Polish Polish-

Random House.

6-Portuguese::Avery, C.B.&Houaiss, A. 1964. The New Appleton

Dictionary of the English and Portuguese Languages. New York:
Appleton -Century -Crofts.

6-Rumanian::Schonkrok, M. 1961. Rumanian-English and EnglishRumanian Dictionary. New York: Frederick Ungar Publishing.
6-Russian::Katzner, K.

1984. English-Russian, Russian-English.

New

York: John Wiley and Sons.

6-Spanish::Williams, E.B.

Espanol Diccionario.

New

1962. Spanish-English Dictionary, Ingles y

York: Holt, Rhinehart, and Winston.
7.

Dravidian

7-Brahui::Burrow, T.&Emeneau, M.D. 1961. A Dravidian

Oxford: At The Clarendon Press.
Etymological Dictionary.

7-Gondi::Burrow, T.&Emeneau, M.D. 1961.

A Dravidian Etymological

Dictionary. Oxford: At The Clarendon Press.

7-Kannada::Burrow, T.&Emeneau, M.D. 1961. A Dravidian

Etymological Dictionary. Oxford: At The Clarendon Press.

7-Kodagu::Burrow, T.&Emeneau, M.D. 1961. A Dravidian

Etymological Dictionary. Oxford: At The Clarendon Press.
7-Kolami::Burrow, T.&Emeneau, M.D. 1961. A Dravidian
Etymological Dictionary. Oxford: At The Clarendon Press.

7-Konda::Burrow, T.&Emeneau, M.D. 1961. A Dravidian

Etymological Dictionary. Oxford: At The Clarendon Press.
7-Kota::Burrow, T.&Emeneau, M.D. 1961.

A Dravidian Etymological

Dictionary. Oxford: At The Clarendon Press.

238
7-Kui "Burrow, T.&Emeneau, M.D. 1961. A Dravidian Etymological
Dictionary. Oxford: At The Clarendon Press.

7-Kurukh::Burrow, T.feEmeneau, M.D. 1961. A Dravidian

Etymological Dictionary. Oxford: At The Clarendon Press.

7-Kuwi::Burrow, T.&Emeneau, M.D. 1961. A Dravidian Etymological

Dictionary. Oxford: At The Clarendon Press.

7-Malto::Burrow, Y.&Emeneau, M.D. 1961.

A Dravidian Etymological

Dictionary. Oxford: At The Clarendon Press.

7-Manda::Burrow, Y.&Emeneau, M.D. 1961. A Dravidian

Etymological Dictionary. Oxford: At The Clarendon Press.

7-Mayalam::Burrow, Y.&Emeneau, M.D. 1961. A Dravidian

Etymological Dictionary. Oxford: At The Clarendon Press.
7-Pengo::Burrow, T.&Bhattacharya, S. 1970. The Pengo Language,
Grammar, Text, and Vocabulary. London: Oxford at the Clarendon
Press.

7-Tamil::Burrow, Y.&Emeneau, M.D. 1961.

A Dravidian Etymological

Dictionary. Oxford: At The Clarendon Press.

7-Telegu::Burrow, Y.&Emeneau, M.D. 1961. A Dravidian

Etymological Dictionary. Oxford: At The Clarendon Press.

7-Toda::Burrow, Y.&Emeneau, M.D. 1961.

A Dravidian Etymological

Dictionary. Oxford: At The Clarendon Press.

7-Tulu::Burrow, Y.&Emeneau, M.D. 1961.

A Dravidian Etymological

Dictionary. Oxford: At The Clarendon Press.

8.

Indo-Pacific

8-Amele::Roberts, J.R. 1987. Amele. London:

Croom Helm.

8-Angoram::Foley, W.A. 1986. The Papuan Languages of

Guinea. London: Cambridge University Press.

New

239
8-Awa::Foley, W.A. 1986. The Papuan Languages of
London: Cambridge University Press.

New

Guinea.

8-Bagupi::Z'graggen, J. A. 1980. A Comparative Word List of the

Mabuso Languages Madang Province Papua New Guinea. Canberra:
The Australian National University.

8-Bemal::Z'graggen, J. A. 1980. A Comparative Word List of the

Mabuso Languages Madang Province Papua New Guinea. Canberra:
The Australian National University.

8-BenaBena:Foley, W.A. 1986. The Papuan Languages of

Guinea. London: Cambridge University Press.
8-Bikol::Mintz,

New

M. 1971. Bikol Dictionary. Honolulu: University of

Hawaii Press.
8-Fore::Foley, W.A. 1986. The Papuan Languages of
London: Cambridge University Press.

New

8-Gadsup::Foley, W.A. 1986. The Papuan Languages of

London: Cambridge University Press.

Guinea.

New

Guinea.

8-Gal::Z'graggen, J. A. 1980. A Comparative Word List of the Mabuso

Languages Madang Province Papua New Guinea. Canberra: The

Australian

National

University.

8-Garus::Z'graggen, J. A. 1980. A Comparative Word List of the

Mabuso Languages Madang Province Papua New Guinea. Canberra:
The Australian National University.

8-Gende::Foley, W.A. 1986. The Papuan Languages of

London: Cambridge University Press.

New

Guinea.

J. A.
1980. A Comparative Word List of the
Province Papua New Guinea. Canberra:
Languages
Madang
Mabuso
The Australian National University.

8-Girawa::Z'graggen,

8-Gumalu::Z'graggen, J. A. 1980. A Comparative Word List of the

Mabuso Languages Madang Province Papua New Guinea. Canberra:
The Australian National University.

240
8-HiriMotu::Dutton, T.E.&Voorhoeve, C.L. 1974. Beginning Hiri
Australian National University.

Motu. Canberra:

8-Kamano-Yagiria::Foley, W.A. 1986. The Papuan Languages of

Guinea. London: Cambridge University Press.

New

8-Kamba::Z'graggen, J. A. 1980. A Comparative Word List of the

Mabuso Languages Madang Province Papua New Guinea. Canberra:
The Australian National University.
8-Kare::Z'graggen, J. A. 1980. A Comparative Word List of the
Mabuto Languages Madang Province Papua New Guinea. Sydney:

Australian

National

University.

8-Manam::Gregersen, E.A. 1976.

Papua New Guinea. 18 :95-lll.

Note on the

Manam Language

of

8-Manit::Z'graggen, J. A. 1980. A Comparative Word List of the

Mabuso Languages Madang Province Papua New Guinea. Canberra:
The Australian National University.

8-Mawan::Z'graggen, J. A. 1980. A Comparative Word List of the

Mabuso Languages Madang Province Papua New Guinea. Canberra:
The Australian National University.
8-Munit::Z'graggen, J. A. 1980. A Comparative Word List of the
Mabuto Languages Madang Province Papua New Guinea. Sydney:
Australian

National

University.

8-Murupi::Z'graggen, J. A. 1980. A Comparative Word List of the

Mabuso Languages Madang Province Papua New Guinea. Canberra:
The Australian National University.
8-Nake::Z'graggen, J. A. 1980. A Comparative Word List of the
Mabuso Languages Madang Province Papua New Guinea. Canberra:
The Australian National University.

8-Panim::Z'graggen, J. A. 1980. A Comparative Word List of the

Mabuso Languages Madang Province Papua New Guinea. Canberra:
The Australian National University.

241
8-Rao::Stamhope, J.M. 1980. The Language of the Rao People,
Grengabu, Madang Province, New Guinea. Canberra: Australian
National

University.

8-Siane::Foley, W.A. 1986. The Papuan Languages of

London: Cambridge University Press.

8-Sihan::Z'graggen,

J. A.

A Comparative Word

1980.

New

Guinea.

List of the

Mabuto Languages Madang Province Papua New Guinea. Sydney:

Australian

National

8-Silopi::Z'graggen,

University.
J. A.

A Comparative Word

1980.

List of the

Mabuto Languages Madang Province Papua New Guinea. Sydney:

Australian

National

University.

8-Tairora::Foley, W.A. 1986. The Papuan Languages of

London: Cambridge University Press.
8 -Tolai:: Franklin, K.J.

Guinea:

Summer

9.

Niger-Khordofanian

Akademische

Verlagsanstalt.

9-Bini::Koelle, S.W.

Druck-u

New

of Linguistics.

9-Basa::Koelle, S.W. 1963. Polyglotta Africana. Graz:

Druck-u

Guinea.

Ukarumpa,

1962. Tolai Language Course.

Institute

New

1963. Polyglotta Africana. Graz:

Akademische

Verlagsanstalt.

9-Bobangi::Whitehead,

J.

1964.

Bobangi Language. Ridgewood,

Grammar and

New

Jersey:

Dictionary of the

Gregg

Press.

9-Dogon::Caiame-Giraule, G. 1968. Dictionaire Dogon, Dialecte,

Langue, et Civilization. Paris: Libraire C. Klincksieck.

9-Ewe::Westermann, D. 1973. Evefiala or Ewe-English Dictionary,

Gbsela Yey or English-Ewe Dictionary. Berlin: Dietrich Reimer.
9-Fula::Swift,

L.B.&Tambada, K.&Imhoff,P.G. 1965. Fula Basic

Course. Washington: Foreign

Service Institute.

242
9-Ife::Koelle,

Druck-u

S.W. 1963. Polyglotta Africana. Graz:

1963. Polyglotta Africana. Graz:

9-Igala::Koelle, S.W.

Druck-u

Akademische

Verlagsanstalt.

Akademische

Verlagsanstalt.

1972. Igbo-English Dictionary based on the

Onitsha Dialect. Benin City, Ethiopia: Ethiopia Publishing Company.

9-Igbo::Williamson, K.

9-Mbukushu::Wynne, R.C. 1980. English-Mbukushu Dictionary.

London: Averbury.
9-Mvumbo::Koelle, S.W. 1963. Polyglotta Africana. Graz:
Akademische Druck-u Verlagsanstalt.
9-Ndebele::Pelling, J.N.

Longman, Rhodesia:

1971.

Practical Ndebele Dictionary.

Bureau.

A Short English-Nyanja Vocabulary.

1960.

9-Nyanja::Price, T.

Literature

Lusaka: The Publications Bureau.

9-Sango::Taber, C.R. 1965.

Seminary

9-Shilluk::Heasty,

Dictionary.

Dictionary of Sango. Hartford:

Hartford

Foundation.

Doleib

J. A.

Hill,

1984. English-Shilluk,

Shilluk-English

the Anglo-Egyptian Sudan:

The American

Mission.

9-Shona::Hannan, J. 1959. Standard Shona Dictionary. London: The

Bureau.

Literature

9-Swahili-Bantoid::Rechenbach, C.W. 1967. Swahili-English

Dictionary. Washington: Catholic University of America Press.
9-Xhosa::Fischer, A. 1985. English-Xhosa Dictionary.

CapeTown:

Oxford University Press.

9- Yoruba:: Abraham, R.C. 1958. Dictionary of Modern Yoruba.
London: University of London Press Ltd.

243
9-Yulu::Koelle, S.W. 1963. Polyglotta Africana. Graz:

Druck-u

Akademische

Verlagsanstalt.

9-Zulu::Doke,

C.M.&MckMalcomb, D.&Sikanana,J.M.A.

Zulu Dictionary. Johannesburg:

10.

10-Apache::Porter, J.C.

Witwatersrand

1958. EnglishUniversity Press.

North Amerind

1980. Vocabulary of the

Apache

or Inde

Mexico. In Museum of Anthropology

Linguistics Series. Edited
Anthropological
in
Publications
Occasional
by C. J. Condie. Greeley, Colorado: University of Northern Colorado.

Language of Arizona and

New

10-Biloxi::Swanton, J.R.&Dorsey, J.O. 1912. The Biloxi and Ofo

Languages. Washington: Washington Government Printing Office.
10-Blackfoot::Frantz, D.G.&Russell, N.J. 1989. Blackfoot Dictionary of
Stems, Roots, and Affixes. Toronto: University of Toronto Press.

10-Cakchiquel::Carmelo, S.S.M. 1940. Diccionario CakchiquelEspanol, Recopilado par Carmelo de Santa de Santa Maria.
Managua: Tipografia Nacional.

10-Choctaw::Byngton, C. 1915. A Dictionary of the Choctaw

Language. Washington: Washington Government Printing Office.
1971. Chontal to Spanish-English; Spanish to
of Arizona Press.
University
Chontal. Tuscon:

10-Chontal::Turner, S.

10-Chorti::Mayers, M.K.

1966. Languages of Guatemala.

The Hague:

Mouton.
10-Crow::Lowie, R. 1960. Crow

Crow Word

Word

Lists:

Crow-English; English-

Vocabularies. Berkley: University of California Press.

10-Hopi-Uto-Aztecan:: Albert, R. 1985. A Concise Hopi and English

John Benjamins.

Lexicon. Philadelphia:
10-Ixil::Mayers,

Mouton.

M.K. 1966. Languages of Guatemala. The Hague:

244
10-Jacaltec::Mayers,

M.K. 1966. Languages of Guatemala. The

Hague: Mouton.
10-Kwakiutl::Boas, F. 1906. Kwaikiutl Texts.

and

New

York: E. Stechert

Company.

10-Menomini::Bloomfield, L. 1975. Menomini Lexicon. Milwaukee:

Milwaukee Public Museum Publications in Anthropology and
History.

10-Micmac-Macro-Algonkian::DeBlois, A.D. 1984. Micmac Lexicon.

Ottawa: Canadian Ethnology Service.
10-Mixtec::deGarcia, C.L.&Alavez, D.F.G. 1986. Dictionario Mixteco
de San Juan Colorado. Verano: Instituto Linguistico de Verano.

10-Mohawk::Michelson, G. 1883. One Thousand Words of Mohawk.

Museum of Man.

Ottawa: National

10-Navaho-Athapascan::Young, R.W.&Morgan, W. 1980. The

Navaho Language: A Grammar and Colloquial Dictionary.
Albuquerque: University of New Mexico Press.
10-Ojibwa::Piggott, G.L.&Graf stein, A. 1983.
Museums of Canada.

An Ojibwa

Lexicon.

Ottawa: National

10-Quiche::Edmonson, M.S. 1965. Quiche-English Dictionary.

Tulane University.

New

Orleans:

10-Squamish::Kuipers, A.H.

Grammar,

Texts, Dictionary.

The Squamish Language,

The Hague: Mouton.

1967.

10-Tzotzil::Laughlin, R.M. 1975. The Great Tzotzil Dictionary of San

Lorenzo Zinacantan. Washington: Smithsonian Contributions to
Anthropology.

10-Winnebago::Radin,
Themselves.

P.

Baltimore:

10-Yokuts::Newman,

S.S.

1950. Winnebago Culture as Described by

Waverly Press.
1944.

York: Viking Fund Publications.

Yokuts Language of California.

New

245
10-Zoque::Hamson, R.&Harrison, M.B. 1984. Vocabulario Zoque de
Rayon. Verano: Institute Linguistico de Verano.

South Amerind

11.

ll-Amahuaca-Pano-Tanoan::Key, M.R.

Tacanan

1968.

Comparative

1965.

Grammatico y

Phonology. The Hague: Mouton.

ll-Aymara-Quechumaran::Ebbing,
Diccionario Aimara. La Paz:

ll-Botocudo::Monteiro, C.

Sao Paulo: Museu

J.E.

Don Bosco.
1948.

Vocabulario

Portugues-Botocudo.

Paulista.

ll-Cashibo::Key, M.R. 1968. Comparative Tacanan Phonology.

The

Hague: Mouton.
ll-Cavinena::Key, M.R. 1968. Comparative Tacanan Phonology.

The

Hague: Mouton.
ll-Chacobo-Pano-Tanoan::Loukota, C. 1968. Classification of South
American Languages. Los Angeles: UCLA.
ll-Chama::Key, M.R. 1968. Comparative Tacanan Phonology. The
Hague: Mouton.
ll-Chaninahua::Key, M.R. 1968. Comparative Tacanan Phonology.

The Hague: Mouton.

ll-Gurani-Tupi::Guasch, A. 1961. Dictionario Castellano-Guarani y
Guarani-Caste llano: Sintactico, Fraseologico, Ideologico. Seville:
Editiones Loyola.
ll-Huitoto::Minor, E.E. 1987. Vocabulario Bilingue: Dialect Minica,
Huitoto-Espanol Espanol-Huitoto. Lomalinda, Columbia: Editorial

Townsend.
ll-Inga::Levinsohn, S.

Mouton.

1976.

The Inga Language. The Hague:

246
ll-Jaqaru::Hardman, M. 1966. Jacaru: Outline of Phonological and
Morphological Structure. The Hague: Mouton and Company.
ll-Lenca::Rivard,
Dictionary.

New

J.J.

1988.

Maya-Spanish

Spanish-Maya

York: Occasional Publications on Mesoamerican

Anthropology.
-Guyana: :Loukota, C. 1968. Classification of South American
Languages. Los Angeles: UCLA.
11

ll-Quechua-Quechumaran::Parker, G.J. 1969. Ayacucho

Dictionary. The Hague: Mouton.

Quechua

Grammar and

ll-Resigaro::Allin, T.R.
Institute

Linguistico de

1979. Vocabulario Resigaro. Verano:

Verano.

ll-Shipibo-Conibo::Key, M.R. 1968. Comparative Tacanan

Phonology. The Hague: Mouton.
11 -Tacanan: :Key,

M.R. 1968. Comparative Tacanan Phonology. The

Hague: Mouton.
ll-Tarascan::Friedrich, P. 1971. The Tarascan Suffixes of Locative
Space. Bloomington: University of Indiana Press.

ll-Totonac::Aschmann, H.P. 1973. Diccionario Totonaco de

Papantla, Veracruz, Totonaco-Espanol, Espanol-Totonaco. Verano:
Instituto Linguistico de Verano.
ll-Tupi::Mello, G. 1967. Dicionario
Sao Paulo: Folco Masucci.
12.

Tupi-Portugues, Portugues-Tupi.

Uralic

12-Finnish::Tuomikoski, A.&Sloor, A. 1957. English-Finnish

Dictionary Suomalaisen Kirjallisuuden. Helsinki: Seura.

n. Nilo-Saharan
13-BongoBagirmi::Doornbos, P.&Bender, M.L. 1983. Languages of
Wadai-Darfur. In Nilo-Saharan Language Studies. Edited by M. L.

247
Bender. 43-79. East Lansing: African Studies Center Michigan State
University.

13-BoraMang::Doornbos, P.&Bender, M.L. 1983. Languages of

Wadai-Darfur. In Nilo-Saharan Language Studies. Edited by M. L.
Bender. 43-79. East Lansing: African Studies Center Michigan State
University.

13-Erenga::Doornbos, P.&Bender, M.L. 1983. Languages of WaidaiDarfur. In Nilo-Saharan Language Studies. Edited by M. L. Bender.
Lansing: African Studies Center University of Michigan.

13-Fongoro::Doornbos, P.&Bender, M.L. 1983. Languages of Wadai

Language Studies. Edited by M. L. Bender.
43-79. East Lansing: African Studies Center Michigan State
Darfur. In Nilo-Saharan

University.

13-Kanuri-Saharan::Lukas,

J.

1967.

Study of the Kanuri

Language, Grammar and Vocabulary. London:

Dawson's.

13-Kara::Doornbos, P.&Bender, M.L. 1983. Languages of Wadai

Darfur. In Nilo-Saharan Language Studies. Edited by M. L. Bender.
43-79. East Lansing: African Studies Center Michigan State

University.
aspects of Logo phonology. In
Edited
by M. L. Bender. 272-279.
Nilo-Saharan Language Studies.
State University.
Michigan
Center
Studies
African
Lansing:
East

13-Logo::Goyaerts, D.L.. 1983.

Some

13-Lokai::Tucker, A.N. 1940. The Eastern Sudanic Languages.

London: Oxford University Press.

13-Lugbara::Tucker, A.N. 1940. The Eastern Sudanic Languages.

London: Oxford University Press.
13-Masalit::Doornbos, P.&Bender, M.L.
Darfur. In Nilo-Saharan

Language

1983. Languages of WadaiM. L. Bender.

Studies. Edited by

43-79. East Lansing: African Studies Center Michigan State

University.

248
13-Merarit::Doornbos, P.&Bender, M.L.
Darfur. In Nilo-Saharan

Language

1983. Languages of WaidaiM. L. Bender.

Studies. Edited by

Lansing: African Studies Center University of Michigan.

13-Mileri::Doornbos, P.&Bender, M.L.
Darfur. In Nilo-Saharan

Language

1983. Languages of WaidaiM. L. Bender.

Studies. Edited by

Lansing: African Studies Center University of Michigan.

13-Miza::Tucker, A.N. 1940. The Eastern Sudanic Languages.

London: Oxford University Press.

13-Nubian::Armbruster, C.H. 1965. Dongolese Nubian, a Lexicon,

Nubian-English, English-Nubian. Cambridge: At The University
Press.
13-0jila::Tucker, A.N. 1940. The Eastern Sudanic Languages.
London: Oxford University Press.

13-Runga::Doornbos, P.&Bender, M.L. 1983. Languages of WadaiLanguage Studies. Edited by M. L. Bender.

43-79. East Lansing: African Studies Center Michigan State
Darfur. In Nilo-Saharan

University.
13-Sinyan::Doornbos, P.&Bender, M.L.
Darfur. In Nilo-Saharan

Language

1983. Languages of WadaiM. L. Bender.

Studies. Edited by

43-79. East Lansing: African Studies Center Michigan State

University.

13-Tama::Doornbos, P.&Bender, M.L. 1983. Languages of WaidaiLanguage Studies. Edited by M. L. Bender.

Lansing: African Studies Center University of Michigan.

Darfur. In Nilo-Saharan

13-Twampa::Thelwall, R. 1983.

Twampa

Phonology. In Nilo-

Saharan Language Studies. Edited by M. L. Bender. 323-335. East

Lansing: African Studies Center Michigan State University.
13-Yulu::Doornbos, P.&Bender, M.L. 1983. Languages of WadaiDarfur. In Nilo-Saharan Language Studies. Edited by M. L. Bender.
East Lansing: African Studies Center Michigan State University.

249
15.

Austro-Thai

15-Brou. 1966. Mon-Khmer subgrouping in Vietnam. In

Austroasiatic Studies. Edited by P. N. Jenner Thompson,L.C., and
Starosta,S. 195-213. Honolulu: University of Hawaii Press.
15-Chrau. 1966. Mon-Khmer subgrouping in Vietnam. In
Austroasiatic Studies. Edited by P. N. Jenner Thompson,L.C., and
Starosta,S. 195-213. Honolulu: University of Hawaii Press.

15-Katu::Thomas, D.D. 1966.

Mon-Khmer subgrouping

in

Vietnam.

Studies. Edited by P. N. Jenner Thompson,L.C., and

195-213. Honolulu: University of Hawaii Press.

In Austroasiatic

Starosta,S.

15-Pearic::Headley, R.K. 1978. English-Pearic Vocabulary. In MonKhmer Studies VIZ. Edited by P. N. Jenner. Honolulu: University of

Hawaii Press.
15-Sedang::Thomas, D.D. 1966. Mon-Khmer subgrouping in
Vietnam. In Austroasiatic Studies. Edited by P. N. Jenner
Thompson,L.C., and Starosta,S. 195-213. Honolulu: University of

Hawaii Press.
15-Thai::Robertson, R.G. 1980. Robertson's Practical English-Thai
Dictionary. Tokyo: Charles E. Tuttle Company.
16.

16-Akha::Burling, R.

Sino-Tibetan

1967. Proto-Lolo-Burmese.

The Hague:

Mouton.
16-Atsi::Burling, R.

1967. Proto-Lolo-Burmese.

The Hague: Mouton.

16-Burmese::Cornyn, W.S. 1958. Burmese Glossary.

American Council of Learned Societies.

New

16-Cantonese-Sinitic::Huang, P.Po-fei.

1970. Cantonese

Cantonese-English English-Cantonese.

New

Press.

York:

Dictionary,

Haven: Yale University

250
16-Gurung::Glover, J.R.&Gurung, D.B. 1979. Conversational Gurung.
Series D no. 13. Sydney: Australian National

Pacific Linguistics

University.
J. A.

16-Lahu-Lolo-Burmese::Matisoff,

1988.

The Dictionary of Lahu.

Berkley: University of California Press.

16-Lisu::Burling, R.

1967. Proto-Lolo-Burmese.

The Hague: Mouton.

16-Mandarin-Sinitic::Cowie, A.P.&Evison, A. 1980. Concise EnglishChinese Chinese-English Dictionary. Hong Kong: Oxford University
Press.

16-Maru::Burling, R.

1967. Proto-Lolo-Burmese.

The Hague:

Mouton.
16-Newari::Jorgensen, H.

1936.

A Dictionary

of Classical Newari.

Copenhagen: Ejnar Munksgaard.

16-Tibetan-Himalayish::Jaschke, H.A. 1965. A Tibetan-English
New York: Frederick Unger Publishing Company.

Dictionary.

17.

Altaic

17-Azerbaijaini::Householder, F.W. 1965. Basic Course in

Azerbaijani. Bloomington: Indiana University Press.

17-Japanese::Kondo, K.C. 1947. Romanized English-Japanese

Dictionary with Chinese Characters. Tokyo: Japan Publishing

Trading Company.
17 -Korean ::Underwood, J.V. 1954. Concise English-Korean
Dictionary Romanized. Tokyo: Charles E. Tuttle Company.

17-Kurdish::Wahby, T. 1966.
Oxford University Press.

Kurdish-English Dictionary. London:

1978. A Concise Manchu-English Lexicon.

J.
University of Washington Press.

17-Manchu::Norman,
Seattle:

251
17-Mongolian::Hangin, J.G. 1970. A Concise English-Mongolian
Bloomington: Indiana University Press.

Dictionary.

17-Turkish::Iz, F. 1978. The Oxford English-Turkish Dictionary.

London: Oxford University Press.

17-Uzbek::Waterson, N. 1980. Uzbek-English Dictionary.

Oxford University Press.

New

York:

APPENDIX C
CODING PARAMETERS FOR ALL GLOSSES
Table C
Ethnoanatomical Glosses anc

253
Table C-2.
Ethnoanatomical Glosses and Consonantal

254

Physiolo

255
Physiological

256
Table C-5.
Culturally Primary Glosses and Vowel Codin g Tallies
Glosses:

257
Table C-6
Culturally Primary Glosses and Consonatal Coding Tallies
Glosses:

APPENDIX D
RANKINGS OF FEATURES AND GLOSSES

INITIAL

Table D-1

Consonantal

Raw

Score Ranking for 16 Glosses

Bilabial

Dental-

Palatal

Nose (27)

Alve
Food

Suck (17)

Swallow
(22)
Breast

(43)

Tooth (42) Swallow

(13)
Spit

(39)

Eat (11)

Velar

LabioVelar
Water

(8)

Glottal

Neck

(35)

Drink (8)

Cough

Dog

Swallow

(7)

(28)

Cough

(13)

Vomit

(13:

(23)

(21)

Swallow

Drink (11) Spit (5)

Vomit

Food (19)

(38)
Suck (37)

Chew

Dog

Mouth

Chew

Dog

Chew

(20)

(36)

(11)

(10)

Food

(4)

(23: Spit

(22)

(11)

Tooth (10)

Swallow

Vomit

(3)

Chew

Mouth

(3)

Mouth

Drink (8)

Eat (3)

(22)
Tooth (20)

Chew

Swallow

Food (19)

Dog

Eat (7)

(22)

(9)

(19)

Water

(18^

Mouth

Tooth

(9)

(35)
Spit

Nose (15)

(18)

Nose (34)

Breast (8)

Drink (17) Cough (33) Spit (8)

(7)

(7)

(3)

Vomit

10

Food

(16^ Breast

(7)

Neck

(2)

Spit

Suck

(2)

Eat (14)

(16)

(33)

Dog

(14)

Vomit

(32)

Water

Nose

(7)

Suck (12)

Neck

Neck

Drink (32) Mouth

(11)

(32)

Mouth

(5)

Tooth

(2)

Drink (12) Neck (5)

(5)

Cough

(2)

Water (11) Water

(6)

(4)

Eat (10)

Eat (30)

Vomit

(4)

Breast (1)

Nose (11)

Breast (4)

Cough

(6)

Water (27) Cough

(3)

Chew

Suck

Suck

(4)

Tooth

(3)

Dog

Food

(4)

(26)

Neck

(2)

258

Nose

(1)
(0)

Breast

(8)

(7)

259
Table D-2

Manner

of Articulation

Raw

Score Ranking for 16 Glosses

260

APPENDIX E

ACTUAL RANKINGS OF FEATURES AND GLOSSES

Table E-1
Actual Rankings of 16 Glosses on 15 Tested Features (1-8)

262

Table E-2
Actual Rankings of 16 Glosses on 15 Tested Features (8-15 )

APPENDIX F
PHONETIC CHARACTERS
Table F-1

Vowel Coding Phonetic C h arac ter

Front High

Round

264
Table F-2
Consonant Phonetic Coding Characters

REFERENCES
Andre,

J.

1966

Onomatopees

et

noms d'oiseaux en

Societe de Linguistique du Paris

Latin.

61

146-156.
Anttila, R.

1975

Affective vocabulary
invitation.

47

in

Another
Jahrbucher

Finnish:

Ural-Altaische

10-19.

Argyle, M.

The syntaxes of bodily communication.

1973

Linguistics 112

71-91.

Asano, T.
Giongo-gitaigo ziten (A Dictionary of Mimetic

1978

Words)
Atzet,

J.,

Tokyo: Kadokawa.

and H. Gerard.

study of phonetic symbolism among native

Navajo speakers. Journal of Personality and
524-528.
Social Psychology 5

1965

Austerlitz,

R.

1967

The distributional
morphophonemes.

identification

Language 43

of Finnish
:

20-33.

Barry, W.J.

1981

Prosodic functions revisited again. Phonetica

320-340.
38
:

265

266
Battistella, E.L.

Markedness: The Evaluative Superstructure

of Language. Albany State: University of New

1990

York
Bentley,

M. and

E.J.

Press.

Varon

An

1933

accessory study of phonetic symbolism.

76-86.

American Journal of Psychology 45

Berlin, B.

Some

1963

semantic features of reduplication in

International Journal of American
211-218.
Linguistics 29
Tzeltal.

Berlin, B.

1972

Speculations on the growth of ethnobotanical

51-86.
nomenclature. Language in Society 1
:

Berlin, B., D.E. Breedlove, and P.H. Raven.

General principles of classification and

nomenclature in folk biology. American
214-242.
Anthropologist 75

197 3

Berlin, B., and P. Kay.

1969

Basic Color Terms: Their Universality and

Evolution Berkley: University of California
.

Press.

Bernard-Thierry,

1960

S.

Les onomatopees en Malagache. Bulletin de

la Societe de Linguistique de Paris. 55
:

240-269.
Bickerton, D.

1990

Language and Species

Chicago Press.

Chicago: University of

Bladon, R.A.

1977

Approaching onomatopoeia.
158-166.
Linguisticum. 8
:

Archivum

267
Boas,

F.,

and E. Deloria

1941

Dakota Grammar. Memoirs of the National

Academy
Boehm,

of Sciences 23

16-90.

C.

1989

Vocal communication of wild chimpanzees.

15-18.
Anthro Quest 39
:

Bolinger, D.

1964

Intonation as universal. In Proceedings

of

Congress of Linguists
Edited by H. G. Lunt. The Hague: Mouton
the Ninth

International

833-848.
Brackbill, Y., and K. B. Little.

1957

Factors determining the

guessing of

meanings of foreign words. Journal of

Abnormal and Normal Psychology 54

312-318.
Brown, C.H.

1976

Wittgensteinian

Linguistics

The Hague:

Mouton.
Brown, C.H.

1979

Folk botanical life-forms: Their universality

and growth. American Anthropologist 79
:

317-342.
Brown, R.

1958

Phonetic symbolism and metaphor. In Words

and Things Edited by B. Roger. New York:
Glencoe Free Press 110-154.
.

Brown,

R., A. Black,

195 5

and A. Horowitz.
Phonetic symbolism
Journal of

50

in

natural

languages.

Abnormal and Social Psychology

388-393.

268
Brown,

and R. Nuttall

R.,

1959

Method

symbolism experiments.

phonetic

in

Journal of Abnormal and Social Psychology

59

441-445.

Burns, A.F.

An Epoch of Miracles:
Yucatec Maya Austin:

1983

Oral Literature of the

University of Texas

Press.
Carroll,

J.

B.

1959

Review of

measurement of meaning.

the

Language 35

58-77.

Chandola, A. C.

1963

Animal commands of Garhwali and

implications.

linguistic

Chastaing,

Word

19

their

203-207.

M.

1958

Le symbolisme des
de

i.

Journal

voyelles,

signification

de Psychologic 55

403-423,

461-481.
Chastaing,

M.

1962

La

brilliance des voyelles.

Linguisticum 14

Archivum

1-13.

Chomsky, N.

1968

Language and Mind

New

York: Harcourt,

Brace and World.

Chusid, J.G.

1970

Correlative

Neuroanatomy and Functional

Neurology Palo Alto,

.

Cal.:

LANGE

Publications.
Costanzo, F.S., N. Markel, and P.R. Costanzo.

1969

Voice quality profile and perceived emotion.

Journal of Counseling Psychology 16

267-270.

269
Cowan, W.

1972

Reduplicated bird names in Algonquian.

International Journal of American Linguistics
229-230.
38
:

Crawford, J.M.

1978

More on Cocopa Baby

Journal

Talk. International

of American Linguistics 44

17-23.

Critchley, E.

1967

Speech Origins and Development

Charles C. Thomas.

Springfield,

111.:

Crockett, D. B.

1970

Secondary onomatopoeia
107-113.
26

in

Russian.

Word

W.W.
1990

Daniel,

Applied Nonparametric
PWS Kent Publishing.

Statistics

Boston:

Darwin, C.J.

1975

On

dynamic use of prosody

in speech
and Process in
Speech Perception Edited by C. A.
Nooteboom. New York: Springer 178-193.

the

perception.

Structure

In

Davis, C.J., R.K. Harding, R.A. Leslie, and P.L.R. Andrews.

1986

The organisation of vomiting as a protective

reflex. In Nausea and Vomiting: Mechanism
and Treatment Edited by C. J. Davis, G. V.
Lake-Bakaar and D. G. Grahame-Smith.
Berlin: Springer- Verlag 65-75.
.

Davis, R.

1961

The

fitness

of

of names to drawings: a crossBritish Journal

study in Tanganyika.
259-268.
Psychology 52

cultural

270
DeCasper, A.

J.,

1980

and

W.

P. Fifer.

Of human bonding: newborns

mother's voices.

prefer their

Science 208

1174-1176.

Dellow, P.G.

1976

The general physiological background of

chewing and swallowing. In Mastication

and

Swallowing: Biological and Clinical

Correlates Edited by B. J. Sessle and A. C.
Hannam. Toronto: University of Toronto
Press 6-17.
.

De Reuse, W.J.
1986

The

lexicalization of sound

symbolism

in

Santiago Del Estero Quechua. International

54-64.
Journal of American Linguistics 52
:

Diffloth, G.

1972

Notes on expressive meaning.

440-447.
Chicago Linguistic Society 8
:

Diffloth, G.

1979

Expressive phonology and prosaic phonology

in Mon-Khmer. In Studies in Tai and MonKhmer Phonetics and Phonology Edited by
.

Thongkum. Bangkok: Chulalongkorn

University

Press.

Doke, C. M.

1935

Bantu Linguistic Terminology

London:

Longmans Green.
Driver, H.E.

1966

/ntroduction

to

in

Methodology Edited by
.

Haven:

HRAF

for

statistics

research. In Readings

comparative

Cross-Cultural
F.

W. Moore. New

Press 310-338.

271
Durbin, M.

1969

Sound symbolism

Mayan language
Symposium on Mayan

the

in

family. Experimental

Ethnosemantics.

New

Orleans, Louisiana:

Mimeograph.
Emeneau, M.

B.

1969

Onomatopoetics

Language 45

in the Indian linguistic area.

274-299.

Emeneau, M.B.

1978

Review of Masica's Defining

Area. Language 54

a Linguistic

201-210.

Ertel, S.

1969

Psychophonetik: Untersuchungen uber

Lautsymbolik und Motivation Guttingen,
Germany: Verlag Fur Psychologie.
.

Falk, D.

1990

Brain evolution in

Homo: The

theory. Behavioral

and Brain Sciences 13

"radiator"

333-381.
Farb, P.

1974

Word

New

Plav:

What Happens When People Talk

York: Knopf

Farb, P. and G. Armelagos

1980

Consuming Passions: The Anthropology of

Eating Boston: Houghton Mifflin Company
.

Ferguson, C.A.

1964

Baby talk in six languages. American

103-114.
Anthropologist 66
:

Fernald, A., and P. Kuhl.

1987

Acoustic determinants of infant preference

motherese speech. Infant Behavior and
Development 10 279-293.

for

272
Fink, B.R.

1975

The Human Larynx:

York: Raven Press.

Functional Study

New

Firth, J.R.

1935

The use and

Fischer,

J.

English

distribution of certain

sounds. English

Studies 17

8-18.

L.

1983

Magical imitation
In Glossogenetics

in
.

the origin of language.

Edited by E. DeGrolier.

New

York: Harwood Academic Publishers

313-328.
Fischer-J0rgensen,

1978

On

the

universal

character of phonetic

symbolism with special reference

Studia

Fonagy,

Linguistica 32

to

vowels.

80-90.

I.

1979

La Metaphore en Phonetique

Ottawa: Marcel

Didier.

Fudge, E.

1970

Phonological
Journal

and

structure

of Linguistics 6

"expressiveness."

161-188.

Gamble, G.

1975

Consonant symbolism
International

41

Journal

in

Yokuts.

of

American Linguistics

306-309.

Ganong, W.F.

1983

Review of Medical Physiology Palo

.

LANGE

Publications.

Alto, Cal.

273
Gardner, B.T., and R.A. Gardner

Two-way communication

1971

with an infant
chimpanzee. In Behavior in Nonhuman
Primates: Modern Research Trends Edited
by A. M. Schrier and F. Stollnitz. New York:
Academic Press 117-184.
.

Gebels, G.

An

1969

investigation

different

of phonetic

Journal

cultures.

symbolism

Behavior 8

Learning and Verbal

in

of Verbal

310-312.

Genette, G.

Mimologiques: Voyage en Cratylie

Gallimard.

197 9

Paris:

Geoffrey, V.C., J.E. Bernthal, T.E. Bertozini, and J.F. Bosma.

Interaction of experimentally elicited
19 84
auricular

Folia

Goudsblom,

cough with voluntary phonation.

Phonetricia 36

31-39.

J.

1983

The domestication of

fire

and the origins of

language. In Studies in Language Origins.

Edited by

Benjamins
Gouzoules, H.,

1986

S.

Wind. Amsterdam: John

E. A.

J.

159-172.

Gouzoules, and P. Marler

Vocal communication:

vehicle for the

study of social relationships. In

The Cayo

Santiago Macaques .Edited by R. G. Rawlins

and M. J. Kessler. Albany State: University of
New York Press 111-129.

Grammont, M.
1901

Onomatopees

et

mots expressifs.

In

Trentenaire de la Societe pour I'Etude des

Langues Romanes Montpellier, Fr.: 261-322.
.

274
Greenberg, J.H.

1964

Some

word

universals of

order.

In

Proceedings of the Ninth International

Congress of Linguists Edited by H. G. Lunt.
The Hague: Mouton 418-420.
.

Greenberg, J.H.

1966

Language Universals. with Special Reference

to Feature Hierarchies The Hague: Mouton.
.

Gudschinsky, S.C.

1964

The ABC's of

lexicostatistics. In

Culture and Society

York: Harper and

Haas,

M.

Row

Types of reduplication
Linguistics

M.

Language

Edited by D. Hymes.

in

New

612-623.

R.

1942

Haas,

in

Thai. Studies

in

1-6.

R.

1970

Consonant symbolism in northwestern

California: A problem in diffusion. In
Languages and Cultures of Western North
America Edited by E. H. Swanson. Pocatello:
.

Idaho

State

University

Press

86-96.

Hall, K.R.

1952

The

fitness of signs to words. British

of Psychology 42

Hamano,

Journal

21-33.

S. S.

1986

The sound-symbolic system of Japanese.

Ph.D. Dissertation, University of Florida,
Gainesville.

Heise, D. R.

1966

Sound-meaning correlations among 1000

English words. Language and Speech 9
:

14-27.

275
Henry,

J.

1936

The linguistic expression of emotion.

250-256.
American Anthropologist 38
:

Hewes, G.W.

1973

Primate communication and the gestural

origin of language. Current Anthropology
14

5-24.

Hewes, G.W.

The invention of phonemically-based

language. "In Glossogenetics: The Origin and
Evolution of Language V.l. Edited by E. De
Grolier. New York: Harwood Academic

1983

Publications

143-162.

Hockett, C. F.

1985

Distinguished lecture:

Anthropologist 87
Hockett, C.F.

f.

American

263-281.

and R. Ascher.

1964

The human revolution. Current

5

Anthropology

135-168.

Holander, M., and D.A. Wolfe.

197 3

Nonparametric Statistical Methods

York: John Wiley and Sons.

New

Householder, F.W.

1946

On

the

English

Hudak,

problem of sound and meaning, an

83-84.
phonestheme. Word 2
:

T.J.

1990

The World's Major Languages Edited

by B. Comrie. New York: Oxford University
Press 757-775.

Thai. In

276
Jakobson, R.

Why

1960

In Perspectives

'mama' and 'papa'?

in

Psychological Theory. Essays in Honor of

Heinz Werner Edited by B. Kaplan and S.
.

New

Wapner.

Universities

York: International
124-134.

Press

Jakobson, R.

1978

Sound and Meaning

MIT

Cambridge, Mass.

The

Press.

Jakobson, R., and L. Waugh.

The Sound Shape of Language

197 9

University

Indiana
Jenkins,

J. J.,

W.A.

195 8

Bloomington:

Press.

Russell, and G.J. Suci.

An

atlas of semantic profiles for 360 words.

American Journal of Psychology 71
688-699.
:

Jerison, H.J.

1976

Discussion paper: The paleoneurology of

language. Annals of the New York Academy
370-382.
of Sciences 280
:

Jespersen, O.

1918

Negle men-ord (Sound gestures nest

vocabulary). Studier
49-55.
Tegner

tillegnade

in

Esaias

Jespersen, O.

1921/1947

symbolism In Language: Its Nature.

Development, and Origin Edited by O.
Jespersen. Woking, England: Unwin Brothers
Limited Chapter XX, 396-411.

Sound

277
Jesperson, O.

1922/1949

Symbolic value of the vowel

Jespersen

Writings of Otto

In

Selected

Edited by D.

Mandelbaum. London: Allen and Unwind Ltd

557-577.
Johannesson, A.
195 2

Gestural Origin of Language: Evidence from

Six "Unrelated" Languages Reykjavik,
.

Iceland: H.F. Leiftur.

Johnson, R.

C, N.

1964

S.

W. K. Olds
Phonetic symbolism in an artificial language.
Journal of Abnormal and Social Psychology
233-236.
69

Suzuki, and

Jonas, D.F., and A.D. Jonas.

1975

Gender differences

in

Anthropology 16

mental function:

to the origin of language.

clue

Current

626-630.

Jonckheere, A.R.

195 4

distribution-free

ordered alternatives.

k-smaple

test

against

Biometrika 41

133-143.
Judson, L.S., and A.T. Weaver.

1942

Voice Science

New

York: Appleton-Century-

Crofts Inc.

Kahlo, G.

1960

Der Irrtum der Onomatopoetiker. Phonetica

5

35-42.

Karwoski, T.F., H.S. Odbert, and C.E. Osgood.

1942

Studies in

synesthetic thinking:

II.

The

of form in visual responses to music.

The Journal of General Psychology 26

199-222.

role

278
Kingston,

J.

1990

Five exaptations in speech: Reducing the

arbitrariness of the constraints on language.
738-739.
Brain and Behavioral Sciences 13
:

Koehn,

E.,

and

S.

Koehn.
Apalai. In Handbook of Amazonian
Languages Edited by D. C. Derbyshire and
G. K. Pullum. The Hague: Mouton 33-127.

1986

Kunihira, S.

1971

Effects of the expressive voice on phonetic

symbolism. Journal of Verbal Learning and

427-429.
Verbal Behavior 10
:

Langdon, M.

1978

Animal talk in Cocopa. International Journal

10-16.
44
of American Linguistics
:

LeCron

Foster,

1978

M.

The symbolic
language. In

Perspectives

structure

Human
.

of primordial

Evolution:

Edited by

S.

Biosocial

Washburn and
California: The

L.

McCown. Menlo Park,

Benjamin/Cummings Publishing Company
77-122.

E. R.

Lehrer, A.

1974

Extended meanings of body-part terms.

International Journal of American Linguistics
135-137.
40
:

Leith, D.E.

1977

Cough. In Respiratory Defense Mechanisms

J. D. Brain, D. F. Proctor and L. M.
Reid. New York: Marcel Dekker 545-592.

Edited by

Lenneberg, E.

1967

The Biological Foundations of Language New

.

York: John Wiley.

279
Lieberman,

P.

The Biology and Evolution of Language

Cambridge, Mass.: Harvard University Press.

1984

Lieberman,

P.,

198 8

and S.E. Blumstein.

Speech Physiology. Speech Perception, and
Acoustic Phonetics Cambridge, U.K.:
Cambridge University Press.
.

Lund,

J.P.

1976

Oral-facial

sensation in the control of

mastication and voluntary movements of the

jaw. In Mastication and Swallowing: Biology

and Clinical Correlates Edited by B. J. Sessle

and A. C. Hannam. Toronto: University of
Toronto Press 145-153.
.

Maddieson, L

1984

Patterns of Sounds

Cambridge, Mass.: M.I.T.

Press.

Mahapatra, K.

1976

Echo-formation in Gta? In Austroasiatic

Studies Edited by P. N. Jenner, L. C.
Thompson and S. Starosta. Honolulu:
University of Hawaii Press 815-831.
.

Malkiel, Y.

1963

The interlocking of narrow sound change,

broad phonological pattern, level of
transmission,

areal

sound
Linguisticum 15, 16

configuration,

symbolism. Archivum
144-173, 1-33.
Malkiel, Y.

1990a

Diachronic Problems in Phonosymbolism

Philadelphia: John Benjamins Publishing

Company.

280
Malkiel, Y.

1990b

From phonosymbolism

morphosymbolism.

to

Diachronic Problems in Phonsymbolism

Edited by Y. Malkiel. Philadelphia: John
Benjamins 157-175.
In

Malkiel, Y.

1990c

Language-universal vs. language-specific

phonosymbolism: /t/ as the carrier of the
idea of stubborn resistance. In Diachronic
Problems in Phonosymbolism Edited by Y.
.

Malkiel.

Philadelphia:

Publishing

Maltzman,

I.,

195 6

Company

John Benjamins
69-80.

L. Morrisett, and L. O. Brooks.

Critique and notes: an investigation of

phonetics symbolism. Journal of Abnormal
249-251.
and Social Psvcholoev 53
:

Marchand, H.

1959

Phonetic symbolism in English wordformation. Indogermanische Forschungen

146-168, 256-277.
64
:

Markel, N.

1966

The psycho-morph: Two exploratory studies.

The Human Use of Symbols Conference.
Tampa, Florida. Unpublished Manuscript
:

1-10.
Markel, N.

1990

Speaking style as an expression of solidarity:

words per pause. Language in Society 19
81-88.
:

Markel, N.

1990

Hypertense individuals and negative

vocabulary use. Unpublished research.
University of Florida: Gainesville.

281
Markel, N.

1991

Psycho-Semiotics

Under review. University

of Florida: Gainesville.

Markel, N., M.F. Bein, and J.A.

The

197 3

Phillis.

of-voice.

Markel, N. and E.P.

between words and tone15-21.

Language and Speech 16

relationship

Hamp
Connotative meanings of certain phoneme
47-61.
sequences. Studies in Linguistics 15

1960

Marks, L.E.

1975

Synesthesia:
senses.

The lucky people with mixed up

Today 9 48-50.

Psychology

Marks, L. E.

1982

Bright sneezes and dark coughs, loud

sunlight and soft moonlight. Journal of
Experimental Psychology. Learning, Memory,
177-193.
and Cognition 2
:

Martin, S.

1962

Phonetic symbolism in Korean. In Uralic and

Altaic Series Edited by N. Poppe.
Bloomington, Ind.: Indiana University Press
.

177-189.
Masson, D.

I.

1952

Synesthesia and sound spectra.

Word

39-41.
McClure, E.F.

1976

Ethno-anatomy: The structure of the domain.

78-88.
Anthropological Linguistics 17
:

McMurray, G.

1958

study of "fittingness" of signs to words by

means of semantic differential. Journal of
310-311.
Experimental Psychology 56
:

282
McWhirter, N.
Guinness Book of World Records

1978

New

York:

Sterling.

Miles, H.L.

Apes and language: The search for

communicative competence. In Language
Primates Edited by J. deLuce and H. T.
Wilder New York: Springer- Verlag 43-61.

1983

in

Miron, M.

S.

1961

cross-linguistic

investigation

of phonetic

symbolism. Journal of Abnormal and Social

623-630.
Psychology 62
:

Mithun, M.

1982

The synchronic and diachronic behavior of

plops,

squeaks, croaks, sighs, and moans.

Journal of American Linguistics

International

48

49-58.

Mori, A.

Comparison of the communicative

1983

and behaviors of group ranging

chimpanzees and pygmy
486-500.
chimpanzees. Primates 24
vocalizations
in

eastern-gorillas,

Newman, P.
1989

Reduplication and tone

in

Hausa ideophones.

Proceedings of the 15th Annual Meeting of

the Berkley Linguistic Society Edited by K.
Hall, M. Meacham, and R. Shapiro Berkeley,
Cal.: Berkeley Linguistic Society Incorporated

248-255.

Newman, S.
1933

S.

Further

experiments

in

phonetic

symbolism.

American Journal of Psychology 45

53-75.

283
Nichols,

J.

1971

Diminutive consonant symbolism

826-848.
America. Language 47

in

North

Norwood,

F.S.

1978

Medico-botanical vocabulary: A special case

of semantic development. Archivum
107-122.
Linguisticum 9
:

Odbert, H.S., T.F. Karwoski, and A.B. Eckerson.

Studies in synesthetic thinking: I. Musical
1942
and verbal associations of color and mood

The Journal of General Psychology 26

153-173.

Odeh, R.E.

1972

On

power of Jonckheere's k-sample test

alternatives. Biometrika 59
467-471.
the

against

Ohala,

ordered

J.J.

1984

An

language utilization of
41

Ojemann,

Fo of

voice.

cross-

Phonetica

1-16.

G., and C. Mateer.

1979

Motor-phoneme
205

Orr,

common

on

ethological perspective

identity

systems. Science

1401-1403.

J.

1944

On some sound
Journal

values in English. British

of Psychology 35

1-8.

Osgood, C.E.

1960

The

cross-cultural

verbal

synesthetic

Sciences 5

generality

tendencies.

146-169.

of visual-

Behavioral

284
Osgood, C.E., GJ. Suci, and P.H. Tannenbaum.

The Measurement of Meaning

195 7

Urbana: The

University of Illinois Press.

Osgood, C.E., W.H. May, and M.S. Miron.

Cross-Cultural Universals of Affective
197 5
Meaning Urbana: University of Illinois Press.
.

Oyama,

T.,

and

J.

1963

Haga.

Common

factors

between figural and

phonetic symbolism. Psychologia 6

131-144.
Paget, R.

193

Human Speech New

.

York: Harcourt, Brace.

Patterson, F., and E. Linden.

1981

The Education of Koko New York:

.

Rhinehart and

Holt,

Winston.

Pecjak, V.

1970

Verbal synesthesiae of colors, emotions, and

days of the week. Journal of Verbal Learning
623-626.
and Verbal Behavior 9
:

Pederson, E.

1986

Expressive language in White Hmong.

Proceedings of the Annual Meeting of the
Berkley Linguistic Society V. Nikiforidou,
M. VanClay, M. Niepokuj, and D. Feder eds.
Berkely: Berkeley Linguistic Society 472-

484.
Pinker, S., and P.

1990

Bloom
language and natural selection.
707-784.
Behavioral and Brain Sciences 13

Natural

Premack, D.

1971

Language
808-822.

in

Chimpanzee. Science 172

285
Pullman, G.K. and Ladusaw, W.A.
Phonetic Symbol Guide. Chicago: University
1986
of Chicago Press.

Reichard, A., R. Jakobson, and E. Werth.

Language and synesthesie.
1949

Word

224-233.
Reichard, G.

Composition and symbolism of Coeur d'Alene

1945

verb stems. International

AmericanLinguistics

11

Journal
:

of

47-63.

Roper, C.W., P.W. Dixon, E.H. Ahern, and V.L. Gibson.

Effect of language and sex on universal
1978

phonetic symbolism. Language

387-397.
19

and Speech

W.
1967

Samarin,

J.

Determining the meanings of ideophones.

West African Languages 9 35-41,

Journal of

W.
1970

Samarin,

J.

Inventory and choice in expressive language.

Word 27 153-169.
:

Samarin, W.J.

1971

Appropriateness and metaphor in the use of

356-369.
ideophones. Orbis 20
:

Sapir, E.

1911

Diminutive and augmentative consonantism

in Wishram. In Handbook of American
Indian Languages Edited by F. Boas.
Washington D. C. Smithsonian Institution,
Bureau of America Ethnology Bulletin.
.

286
Sapir, E.

1929

study in phonetic symbolism. Journal of

225-239.
Psychology 12

Experimental
Sapir,

J.

D.

1975

Big and

thin:

Two

terms.

linguistic

Diola-Fogny meta-

Language and Society

1-15.

Saussure, F. de

1959

Course

General Linguistics

in

Philosophical

New

York:

Library.

Schuchardt, H.

Keltorom, frog- frogn- Lautsymbolik.

fur romanische Philogien
199-205.
21

1897

Zeitschrift
:

Shafer, R.

Some Uto-Aztecan Sino-Tibetan comparisons

1964

and their significance. Orbis 13

104-109.

Shanks, L., and C.T. Velanti.

Form and

1990

Summer

function in Ndiuka ideophones

Institute

of Linguistics.

Paramaribo,

Suriname.
Shields, K. C.

On

1976

the origin of normal reduplication in

37-43.
Orbis 25

Indo-European.
Siegel, A.,

I.

1967

Silverman, and N. Markel.

On

the effects of

mode

of presentation on

phonetic symbolism. Journal of Verbal

171-173.
Learning and Verbal Behavior 6
:

Siqueland, E.R., and C.A. DeLucia.

1969

Visual
in

reinforcement

human

infants.

of

sucking
1144-1147.

nonnutritive

Science 165

287
Smithers, G. V.

1954

Some

English ideophones.
73-111.

Linguisticum 6

Archivum

Snowdon, C.T.

1986

Vocal communication. In Comparative

Primate Biology. Edited by G. Mitchell and
New York: Alan R. Liss 495-530.
J. Erwin.

Solomon, L.N.

1958

Semantic approach

to

the

perception of

complex sounds. The Journal of the Acoustical

421-425.
Society of America 30
:

Solomon, L.N.

1959

Semantic reactions

to

systematically

varied

sounds. The Journal of the Acoustical Society

986-990.
of America 31
:

Swadesh, M.

1971

The Origin and Diversification of Language

Chicago: Aldine.
Tanner, N., and A. Zihlman.

1976

The evolution of human communication:

What can

primates

tell

New York Academy

us? Annals of the

of Science 280

467-480.
Tarte, R. D.

1974

Phonetic symbolism in adult speakers of

87-94.
Czech. Language and Speech 17
:

Tarte, R. D., and L. S. Barrit.

1971

Phonetic symbolism in adults speakers of

158-168.
English. Language and Speech 14
:

Taylor,

I.

K.

1963

Phonetic symbolism re-examined.

200-209.
Psychological Bulletin 60
:

288
Taylor, I.K.

Anatomy

1967

phonetic

General
Taylor,

I.

K.,

1962

of words used in a word-matching

symbolism experiment. Journal of
231-239.
Psychology 76
:

and M. M. Taylor
Phonetic

symbolism in four unrelated

languages. Canadian Journal of Psychology
16

344-356.

Terpstra, T.J.

1953

The exact probability

distribution

of the

against trend and

testing

statistic

for

normal

approximation. Indagationes
433-437.
15

Mathematicae

its

Thorndike, E.L.

1945

On Orr's hypothesis concerning the front and

back vowels- British Journal of Psychology
10-14.
36
:

Tiwary, K. M.

1968

The echo-word construction

Anthropological

Tsuru,

S.,

and H.

193 3

in

Linguistics 10

Bhojpuri.
:

32-38.

S. Fries.

problem

in

Psychology

meaning. Journal of General

281-284.
:

Uhlenbeck, E.M.

1950

The structure of the Javanese morpheme.

Lingua 2 239-270.
:

Ullmann,

S.

1966

Semantic universals.

In

Universals

of

Language Edited by J. H. Greenberg.

Cambridge, Mass.: M.LT. Press 217-262.
.

289
Ultan, R.

Size-Sound Symbolism. In IJniversals of

Human Language Edited by J. H. Greenberg.

1978

Stanford, Cal.:

Stanford University Press

525-568.
Waser, P.M., and C.H. Brown.
Is there a "sound window" for primate
communication? Behavioral Ecology and
73-76.
Sociobiology 15

1984

Washburn, S.L.
Tools and human evolution. Scientific
63-75.
American 203

1960

Weiss,

J.

H.

Phonetic symbolism re-examined.

454-458.
Psychological Bulletin 61

1964

Weiss, J.H.

study of the ability of English speakers to

guess the meanings of nonantonym foreign

1966

words. The Journal of General Psychology

74

:.97-106.

Wentworth, H., and S.B. Flexner.

Dictionary of American Slang

I960

New

York:

Thomas Y. Crowell Company.

Wertheimer,

M.

1958

between the sound of a word

meaning. American Journal of
412-415.
Psychology 71

The

relation

and

its

Wescott, R.W.

1971a

Labio-velarity and derogation in English:

study in phonosemic correlation.

123-137.
Speech 46
:

American

290
W.
1971b

Wescott, R.

Linguistic iconism.

Language 47

416-428.
Wescott, R.W.

1980a

Ideophones

in

Sound and

Bini and English. In

Sense: Linguistic Rssavs on Phonosemic

Subjects Edited by R. Wescott. Lake Bluff,
Illinois: Jupiter Press 195-209.
.

Wescott, R.W.

1980b

origin of speech. In Sound and Sense:

Essays in Phonosemic Subjects Edited by
R. Wescott. Lake Bluff, Illinois: Jupiter
Press 100-117.

The

W.
1980c

Wescott, R.

Sound and Sense: Linguistic Essavs on

Phonosemic Subjects Edited by R. W.
Wescott. Lake Bluff, Illinois: Jupiter Press.
.

Wescott, R.W.

1980d

Metaphones in Bini and English. In Sound

and Sense: Linguistic Essavs on Phonosemic
Subjects Edited by R. W. Wescott. Lake Bluff,
.

Illinois:

Wescott, R.

Jupiter

Press

210-220.

W.

1987

Holesthemes or phonesthemes twice over.

67-72.
General Linguistics 27
:

Williams, K.

1976

Synaesthetic adjectives:

semantic change.

possible law of

Language 52

461-478.

Wissemann, H.

1954

Untersuchungen zur Onomatopoiie

Heidleberg, Germany: C. Winter.

291
Witkowski,

S.

Guttman scaling of kinship semantics.

In Kinship Studiesin the Morgan Centennial
Year Edited by P. Reining. Washington D.C.:
Anthropological Society of Washington
167-188.

197 2

Witkowski,

S.,

1977

and C.H. Brown.

An explanation of color nomenclature
universals. American Anthropologist 79

50-57.
Witkowski,

S.,

197 8

and C.H. Brown.

Lexical universals. Annual Review of
427-451.
Anthropology 7
:

Witkowski, S.R., and C.H. Brown.

1982

Culture,

18

environment, and polysemy.

Man

72-89.

Wyke, B.

1967

Recent advances in the neurology of

phonation: Phonatory reflex mechanisms

in

the larynx. British Journal of Disorders of

Communication 2 2-14.
:

Wynne,

J.D.

1982

Learning

Approach

A Common Sense
York: Macmillan Publishing.

Statistics:
.

New

BIOGRAPHICAL SKETCH
Nick Ciccotosto was born

in Elmhurstjllinois in

1955. His

parents are Donald Tosta and Irene Tosta of Ingelside, Illinois.

He

received his B.A. in cultural anthropology from Northern Illinois

University in

his

M.A. from

the

same university

in

1984.

married to Carol P. Costoff in June 1989. His current

He was
interests
states,

1978 and

include

languages origins,

bio-cultural

evolution,

trance

primate vocal communication, educational techniques for the

mentally

handicapped,

bilingual

education,

translated

J. P.

Sarte's

comparative religion, improved

methods of

and sound symbolism. He has recently

timely play, Les

292

Mains

Sales, into English.

I have read this study and that in my opinion

standards of scholarly presentation and
acceptable
conforms to
fully adequate, in scope and quality, as a dissertation for the

certify that

it

is

degree of Doctor of Philosophy.

Linda D. Wolfe^ Chairperson

Professor of Anthropology

certify that

have read

study and that in my opinion

of scholarly presentation and

this

to acceptable standards
adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.

conforms

it

is

fully

/2Ja^xA>1i^ ^^ Wl MxfYKOc
Christiana M. Leonard

A/

Professor of Neurosciences
I have read this study and that in my opinion
acceptable standards of scholarly presentation and
fully adequate, in scope and quality, as a dissertation for the
degree of Doctor of Philosophy.

conforms

certify that
to

it

is

Norman N. Markel
Professor of Communication
Processes and Disorders
it
I have read this study and that in my opinion
is
and
presentation
scholarly
of
standards
acceptable
conforms to
fully adequate, in scope and quality, as a dissertation for the
I

certify that

degree of Doctor of Philosophy.

Robert

Lawless

Associate Professor of

Anthropology

certify that

conforms

to

have read

this

acceptable standards

study and that in my opinion

presentation and

of scholarly

adequate, in scope and quality, as a dissertation for the

degree of Doctor of Philosophy.

it

is

fully

^^

Ronald Kephart
Assistant

Foreign

Professor of

Languages

University of North Florida

This dissertation was submitted to the Graduate Faculty of the

Department of Anthropology in the College of Liberal Arts and
Sciences and to the Graduate School and was accepted as partial
fulfillment of the requirements for the degree of Doctor of

Philosophy.

December

1991

Dean, Graduate School

"3 1262 08553 5770