Review

Reviewed Work(s): Psychology and Language: An Introduction to Psycholinguistics by

Herbert H. Clark and Eve V. Clark
Review by: Lynn H. Waterhouse
Source: Language , Jun., 1979, Vol. 55, No. 2 (Jun., 1979), pp. 436-439
Published by: Linguistic Society of America

Stable URL: https://www.jstor.org/stable/412600

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

is collaborating with JSTOR to digitize, preserve and extend access to

Language

This content downloaded from

139.135.38.55 on Sun, 21 Jan 2024 10:46:12 +00:00
All use subject to https://about.jstor.org/terms
436 LANGUAGE, VOLUME 55, NUMBER 2 (1979)

Psychology and language: an introduction to psycholinguistics. By HERBERT H

CLARK and EVE V. CLARK. New York: Harcourt Brace Jovanovich, 1977. Pp.
xvi, 608. $15.95.
Reviewed by LYNN H. WATERHOUSE, Trenton State College
This text is an extremely well-organized and lucidly written discussion of nearly
all the research areas and theoretical constructs of importance for psycholinguistic
in the past five to ten years, and it contains many clear and succinct summaries of
critical theories and findings. The Clarks are concerned with 'the study of thr
mental processes-the study of listening, speaking, and of the acquisition of these
two skills by children' (vii): and in their consideration of these three topics through
out the five major parts of the book-' Language',' Comprehension',' Production
'Acquisition', and 'Meaning and thought', they maintain a consistent level of
analysis, keeping their discussion tightly focused on efforts to model systematical
the many theorized mental operations that may contribute to the general processe
of speaking, listening, and acquisition.
However, within this frame, the rigor of C&C's approach in discussing such
psycholinguistic process modeling devolves on a continuing brief for a particul
viewpoint within psycholinguistics: the notion that the PROPOSITION is a 'psycho-
logically real' unit of analysis for most measurable behaviors which arise from
language-bound cognitive processes. According to C&C, 'For speakers the proble
is how to express propositions in strings of words, and for listeners, it is how to
reconstruct the underlying propositions' (31). For them, comprehension, memory
storage, speech production, and acquisition are all processes based on propositions:

'What is comprehension? ... it is the process by which listeners come to an interpretation

for a stream of speech, and this is called the construction process ... In the construction
process, listeners take in the raw speech, isolate and identify the constituents of surface
structure, and build propositions appropriate to each. As they build each proposition, they
add it to the interpretation of the sentence they have formed so far, and the propositions
taken together constitute the finished interpretation.' (84-5)
'Over the long term, information is often represented as a network of propositions ...
Yet ... information is incorporated into global representations that ... may or may not
consist of a network of propositions.' (172)
' In planning what to say, speakers are faced with a series of interrelated problems ... In
choosing a sentence, speakers have to decide on the propositional content, illocutionary
content, and thematic structure. They have to pick propositions that mirror the ideas they
want to get across. They have to focus on salient verbalizable events and states and their
participants, and find propositions to fit them.' (257)
'When children use their first words ... they begin by talking about what they already
know. In effect, the "here and now" provides the propositional content of one and two
word utterances.' (330)

While C&C's propositional model provides a powerful template for measuring,

trimming, and shaping interpretations of data from others' research-generating a
tour-de-force synthesis of an immense amount of current research-nonetheless it
remains true that not all psycholinguists subscribe to this model. Furthermore, as
C&C's discussion of others' research often is presented as an argument that such
research can BEST be interpreted by the propositional model, their text cannot be

This content downloaded from

139.135.38.55 on Sun, 21 Jan 2024 10:46:12 +00:00
All use subject to https://about.jstor.org/terms
 REVIEWS 437

said to give an unbiased representation of the current state of affair

linguistics. If this were not meant as a text, the synthesis provided by th
brief for C&C's theory would be an asset, not a problem. However, this
written as a general text for 'undergraduate and graduate students in ps
linguistics, and related fields' (vii); as such, it may have undermined its
by the presentation of such a forceful and consistent argument for
model. Students may well be convinced that the proposition is 'psyc
real' when they have finished this volume, but they may also come
sense that much of psycholinguistic research has been smoothly a
leading to the establishment of C&C's model, rather than being awar
argument constitutes one model among several.
Part One, 'Language' (pp. 3-40), provides an excellent introduction
sideration of the processes of listening, speaking and acquisition; it offe
discussion of what modeling itself entails, introduces a simple sch
structure of language, and outlines three theorized functions of languag
act goals, propositional goals, and thematic-structure goals. For their con
of what modeling entails, C&C use the analogy of bicycle building i
separate clearly for the reader the nature of structure, function, and pr
then proceed to consider structure and function. In their discussion of t
structure of language, C&C do not explain the state of current competin
of the structure of language (English), nor do they suggest in any w
claims for various models of the grammar of English have had majo
lines of research within psycholinguistics-something which would gi
perspective, and provide context for currently valued models. C&C d
present a very clean schema of the structure of language: 'Every sentenc
two levels of structure. Its SURFACE STRUCTURE is its linear arrangemen
phrases, words, and sounds. Its UNDERLYING REPRESENTATION is its mean
consists of propositions that are interrelated in a particular way. Loosely
the surface structure says how the sentence is to be spoken, and the
representation says how it is to be understood' (12).

C&C claim that language has three central functions, two of which are social and
-speech acts and thematic structure; the other is more an internal cognitive fun
creating propositions. Speech acts are classified into direct and indirect varieties, a
of speech acts is discussed. Propositions are claimed to have three functions, within
function of providing the 'meat' or ideas for sentences: (1) to provide informat
states and events; (2) to provide information about such states and events; and (3) t
qualify other propositions (cf. Zeno Vendler, Linguistics in philosophy, Ithaca, 196
structure means that speakers and listeners must cooperate in order for the v
speech interaction to be done. Following Halliday, this notion provides that (a) liste
to information in speech (sentences) with an either/or decision, judging whether w
hearing is a given or new piece of information: (b) speakers and listeners pay criti
to subject and predicate division; and (c) initial or introductory material in a s
provide a frame for insertion of subsequent parts.

Part Two, 'Comprehension', contains Chapter 2, 'Comprehension of sentences'

(43-85); Chapter 3, 'Utilization of sentences' (87-132); Chapter 4,' Memory for
prose' (133-73); and Chapter 5,' Perception of speech' (175-220). C&C believe that,
when listeners hear a sentence, they employ a 'construction process', the goal of

This content downloaded from

139.135.38.55 on Sun, 21 Jan 2024 10:46:12 +00:00
All use subject to https://about.jstor.org/terms
438 LANGUAGE, VOLUME 55, NUMBER 2 (1979)

which is to build an underlying representation of the sentence based on

ponent propositions. Having done this, listeners then turn to the 'utiliz
process', which involves them in stepwise mental operations which follow t
functions of the three proposed functions: on hearing an utterance, li
identify the speech act, propositions, and thematic structure, searching
memory for any information they may have which matches the given info
then dealing with the remaining new information typologically. In addition
the utilization process calls for the listener to behave according to thre
principles:
(1) The reality principle: Listeners can expect that speakers are actually ta
about something of which they can make sense.
(2) The cooperative principle: Listeners expect that speakers really are try
communicate honestly with them.
(3) The congruence principle: Listeners are continually searching their me
for stored information congruent to that which is incoming.
Critical to Part Two is a discussion in Chap. 2 as to whether comprehension is a bott
or top-down process. C&C suggest that comprehension may be both proposition-driven
(top-down) AND constituent-driven (bottom-up); they argue that, depending on the situation or
conditions under which they are listening, listeners may look for surface features to support a
developing propositional hypothesis about the incoming sentence, or they may build up a
notion of what propositions are involved through steps of combining constituents. C&C's
consideration of these issues is well served by their propositional model, and clearly points the
way to further research.
The chapter on prose memory follows the traditional division of short-term and long-term
memory. C&C argue that the proposition is the basic unit of long-term memory, with memory
being served by the integration of new propositions with old, and the creation of 'global
representations' which, they argue, may be a network of propositions.
The final chapter in Part Two, 'Perception of speech', provides an exceptionally clear but
brief introduction to phonetics, phonological theory, and the current state of speech perception
research. Close attention is given to the motor theory of speech and to a consideration of
factors which may influence the perception of continuous speech.
Part Three is devoted to production. In Chapter 6, 'Plans for what to say'
(223-58), C&C argue (1) that speech has an instrumental goal; and (2) that speech
arises from a nested set of plans wherein discourse plans encompass sentence plans,
and sentence plans encompass constituent plans. After the generation of such plans,
speech is executed (Chapter 7, pp. 259-92) by means of a set of motor orders called
the articulatory program. C&C argue that five considerations are critical to speech
plans: (1) knowledge of the listener; (2) the cooperative principle; (3) the reality
principle; (4) the social context; and (5) the linguistic devices available.
As promised in the preface, Part Four deals with child language development
through the filter of the model established in Parts Two and Three. Chapter 8,
'First steps in the child's language' (295-331), Chapter 9, 'Later growth in the
child's language' (333-73); and Chapter 10, 'First sounds in the child's language'
(375-404), provide concise, systematic summaries of current research; together,
they are as good an introduction to the whole field of child language development
as one could ask for. Especially noteworthy is the integration of social interaction
findings with the body of research on early stages provided in Chap. 8. Chap. 9 on
acquisition examines overgeneralization, the development of speech-act knowledge,

This content downloaded from

139.135.38.55 on Sun, 21 Jan 2024 10:46:12 +00:00
All use subject to https://about.jstor.org/terms
 REVIEWS 439

and children's sensitivity to given/new distinctions in the course o

Chap. 10 covers the range of perception and production research on the
of the sound system, examining the seeming paradoxes of perceptio
representation.
Part Five, 'Meaning and thought', contains Chapter 11, 'The repre
meaning' (407-48); Chapter 12, 'Uses of meaning' (449-83); Chapter 1
in the child's language' (485-514); and finally, Chapter 14, 'Language
(515-58). In the first of these, C&C examine various possible forms of t
ship between the hypothesized internal encyclopedia and hypothesi
lexicon. A componential determination of the internal lexicon is review
semantic differential and procedural semantics are discussed. In 'Uses of
C&C show propositions as operational at the lexical or word level:
itself be a proposition. The role of such word-propositions in comp
considered, covering logical and categorical relations between such
role in production is also considered, and here C&C argue that vario
procedures are critical toward selection. The notion that there is a
between perception and the nature of the relationship between word-pr
is introduced, to be considered further in the final chapter. The chapte
tion of meaning centers on children's continuing adjustment of their ow
standings of word meanings to those of adult usage: children 'derive str
producing and interpreting words ... their initial hypotheses often over
adult meaning, but their use is over-extended or under-extended compa
adult's. At other times, they fail to hit on any part of the adult me
The final chapter of Part Five (and of the text), on 'Language and t
presents discussions of non-human primate language-training exper
discussions of the natural-category research presently going on; both of
used to buttress C&C's contention that language most likely arises f
cognitive abilities-which, though not specific to humans nor specific fo
behavior in humans (cf. findings of language skill development in
primates), are nonetheless perhaps uniquely developed in humans (cf. th
of natural-category research). Implicit in these discussions is the n
cognitive-perceptual skills determine language skills, and that the differe
humans and the great apes is one of degree and not kind. The case
evolution of language in humans is not well presented; it is, in fact
a-priori by an argument that the label 'innate' has the power to close off
gation of the processes of behavior. This particular discussion is one
enlightened in the entire volume, and gives short shrift to an area of c
importance for psycholinguistics.
This volume is a compelling synthesis of the greater part of current r
psycholinguistics. This synthesis, with the notion of the proposition at
both the book's great strength and its weakness. I would recommend th
to all students and scholars interested in the area, with the simple p
readers maintain an awareness that the Clarks, in creating this powerfu
have filtered the research of others through the patterns of their own

[Received 12 April 1978.]

This content downloaded from

139.135.38.55 on Sun, 21 Jan 2024 10:46:12 +00:00
All use subject to https://about.jstor.org/terms