THEORIES OF LANGUAGE ACQUISITION I N RELATION TO

BEGINNING READING INSTRUCTION'

Ronald Wardhaugh
University of Michigan

Various theories of language acquisition a r e discussed: behavior-

istic, nativistic, and cognitive. Linguistically-oriented theories
a r e compared with learning-oriented theories, and four contro-
versial issues of frequency of stimuli, imitation, expansion, and
meaning a r e reviewed. The theories a r e all rather unsubstantial
a t present. Reading acquisition s e e m s to be very different from
language acquisition, and the theories of language acquisition ap-
pear to have little to offer anyone in coming to a better under-
standing of how beginning reading should be taught.

INTRODUCTION

Numerous summaries of research in language acquisition exist:

McCarthy (1954) summarizes work completed prior to 1950; Elkonin
(1958), Brown and Berko (1960), Carroll (1960), Brown (1965), Ervin-
Tripp (1966), Ervin-Tripp and Slobin (1966), McNeill (1966, 1970a,
1970b), and Slobin (1967) summarize parts, o r the whole, of more
recent work; and Kelley (1967) provides one of the most interesting
discussions of many of the major issues. The purpose of this paper
is to isolate and assess major theories of language acquisition and
to relate these theories to beginning reading instruction. The par-
ticular focus is the acquisition of syntax. The theories are also
reviewed from the perspective of the linguistic knowledge available
today; consequently, certain cognitive and affective factors a r e mini-
mized. These factors are not to be considered unimportant in be-
ginning reading instruction; rather they are to be considered beyond
the scope of this paper.

LANGUAGE ACQUISITION
Atheoretical Studies
A reading of McCarthy's summary article induces mixed feel-
ings in anyone trained in linguistics. She reports on a wide variety
1Preparation of this paper was supported through an Office of Education Contract
OEC-0-70-4790 (508) to Rutgers, The State University. The author is grateful to Patricia Dishuck
and Mary Ann Gatten for their assistance.

1
2 LANGUAGE LEARNING, VOL. 21, NO. 1

of descriptive and normative studies, but all seem unrevealing inso-

far as current interests in language acquisition are concerned. The
studies reported appear atheoretical today because the investigators
made little attempt to formulate and test fruitful hypotheses and to
handle data other than quantitively. Consequently, no coherent ac-
count of language acquisition emerges from the studies reviewed by
McCarthy. Instead, child language appears to drift somehow from
a prevocalic stage, through various stages replete with e r r o r s and
deficiencies, toward the clearly articulated speech of a n ideal
speaker of Standard English. As a result, sounds “emerge” in
ways that a r e never specified, “ f i r s t words” are uttered a t a char-
acteristic time, grammatical distinctions are “acquired, ” often
through the elimination of various “ e r r o r s , ” and vocabularies
‘‘expand” as the child’s dictionary gains more entries. Gradually,
by some process of making successive approximations, the child’s
language becomes more and more like the language ascribed in
traditional g r a m m a r s to those who speak the language “properly. ”
Working in such a way, investigators may t r y to discover when
the child learns to distinguish pin from pen and witch from which,
a l l the while ignoring the fact that in certain dialects such distinc-
tions are not made a t all. Or they may t r y to count various sen-
tence types using formulae for sentence description that derive
from analyses of writing and studies of rhetorical devices rather
than from any close observation of spoken language. Or they may
calculate word frequencies and compute type-token ratios without
defining the concept of “ a word” or devising the most appropriate
elicitation procedures. Such investigators often collect considera-
ble quantities of data which can be neatly inventoried and displayed
in tables and figures (for example, tables of e r r o r s in articulation
which show a gradual reduction in frequency as age increases).
However, the data are essentially unrevealing because the investi-
gators do not a s k why it is that one linguistic skill is acquired
before another, or what is the nature of the linguistic ability of
the child a t various stages in his linguistic development.
Only in recent years have such questions been asked by psy-
chologists and linguists engaged in the study of language acquisition.
They have realized that inventories are unrevealing unless they
show which items contrast with each other within the inventories.
They no longer disregard regional and social variations in speech
and developments in modern linguistics. They insist that it is im-
possible to describe language acquisition without first spelling out
either a specific theory of language or a general theory of learning.
Recent work on language acquisition therefore confronts these
theoretical issues. It does so at the expense of large scale data
 LANGUAGE ACQUISITION 3

collection, investigators preferring to test out hypotheses on as few

a s two or three children, as Brown and Bellugi (1964) did with
Adam and Eve, or on a single phonological, grammatical, or se-
mantic distinction, as Klima and Bellugi (1966) did with negation.
One issue that has never been dealt with satisfactorily, even
in recent work, is the specification of the ultimate linguistic knowl-
edge or ability that is being acquired. Obviously, more is involved
than knowledge of a dictionary or of a n inventory of sentence pat-
terns, or the ability to combine words and patterns. N. Chomsky
(1965) has proposed the t e r m competence, as distinguished from
performance, to describe this knowledge. However, this t e r m has
become more of a slogan than a well-defined concept in linguistics.
Since research in language acquisition must focus on such issues
as “increasing complexity” and ‘(developing competence,” a cer-
tain vagueness results when the end-point toward which the child
is assumed to be progressing still remains largely hidden from
view. Menyuk (1969) discusses some of the problems that result
in attempts to interpret data in such circumstances. Fortunately,
many of the data a r e not in dispute among those who study language
acquisition, for all agree that certain stages or trends can be ob-
served: babbling ends around 18 months; holophrastic utterances
precede two- and three- word utterances; early speech is “tele-
graphic”; control of word order antedates control of inflections;
and comprehension outstrips production. The interpretation of the
data is the crucial issue.

Behavioristic Theories
In his book Verbal Behavior (1957), .. Skinner -proposes
- acompre-
hensive theory of language acquisition and language behavior in
which specific linguistic behaviors a r e acquired through operant
conditioning and then extended through response generalization.
N. Chomsky’s devastating review (1959) of the book demonstrates
the inappropriateness of Skinner’s proposal. His criticisms re-
iterate e a r l i e r arguments from Syntactic Structures (1957) that
existing theories of language a r e inadequate for almost any purpose
and that the kind of theory he himself proposes is needed. The
review a l s o attacks the adequacy of reinforcement theory and the
notion of generalization, as formulated by Skinner, in explaining
either language acquisition or language behavior. Chomsky claims
that the theory is illusionary, that most of its concepts are irrele-
vant i n explaining linguistic behavior, and that the real issues a r e
never confronted. Chomsky is particularly critical of Skinner’s
failure to recognize the contribution the child makes to language
acquisition, declaring that:
4 LANGUAGE LEARNING, VOL. 21, NO. 1

.
. . a refusal to study the contribution of the child to language
learning permits only a superficial account of language acquisi-
tion, with a vast and unanalyzed contribution attributed to a step
called ‘generalization’ which in fact includes just about every-
thing of interest in this process. If the study of language is lim-
ited in these ways, it seems inevitable that major aspects of
verbal behavior w i l l remain a mystery. (1959:58)

However, in spite of Chomsky’s criticisms of the inadequacy

of conditioning or reinforcement theories to explain language acqui-
sition such theories are still proposed. Staats and Staats (1962,
1963, 1968), f o r example, use such t e r m s as operant learning, r e -
inforcing stimuli, time and scheduling of reinforcement, successive
approximation, chaining, extinction, and discrimination and general-
ization to explain how language is acquired. Such concepts can
only weakly explain why all children exhibit much the same pattern
of development, how they construct novel utterances even in the
earliest days of language use, and in what ways they master the
abstract relationships that a r e not readily apparent in the utter-
ances they hear. This last point is extremely important because,
as Garrett and Fodor (1968) argue, the facts of language are ab-
stractions which children must acquire from masses of highly vari-
able data. Language is a mentalistic phenomenon, and S-R theories
a r e unable to account for either its acquisition or use. The theory
proposed by Staats and Staats involves the learning of a finite s e t
of responses according to certain probabilities of occurrence. On
the other hand, the current view is that a language is an infinite
set of responses that are available to a speaker, and that language
use is essentially creative. Probability has little to do with lan-
guage use, although, of course, certain linguistic usages can be
conditioned to events in the world once such usages have been ac -
quired.
Jenkins and Palermo (1964) propose a theory of language ac-
quisition that recognizes some recent linguistic advances. The basic
problem they see in language acquisition is that of explaining how
the child acquires the frames of a phrase-structure grammar and
the ability to substitute items within these frames. They propose
that the child learns the stimulus and response equivalences that
can occur in the frames. They heavily emphasize imitation, either
overt or covert, as a force in establishing bonds between stimuli
and responses, and they claim that the child generalizes to form
classes of responses. However, they do not explain how control
of such classes allows the child to construct longer sentences.
Their theory does not attempt to analyze complex issues; it merely
hints at them. The linguistic theory that Jenkins and Palermo
 LANGUAGE ACQUISITION 5

propose is one N. Chomsky (1957) criticizes for being inadequate

in that it does not account for the abstract nature of linguistic
knowledge. Weksel (1965) is also critical of their proposal, claim-
ing that it is linguistically inadequate and nowhere comes to grips
with its central concept of generalization.
Another theory of language acquisition cast in the behavioristic
mold comes from Braine (1963a, 1963b, 1965). This theory involves
the principle of ‘‘contextual generalization,” according to which the
child observes that certain s e t s of items occur in certain positions.
He makes generalizations about positions rather than about the s e t s
of items that occupy them. The positions themselves are not simply
linear, but may be hierarchical. Consequently, the theory attempts
to explain how the child acquires the hierarchical grammatical
structures of sentences. Braine claims that transformations can
be learned through contextual generalization. If they cannot, he
declares that the failure argues as much for a reshaping of lin-
guistic theory as it does for a reshaping of the principle of con-
textual generalization:

If there is a possibility that the simpler of two possible gram-

matical solutions might require the more complex acquisition
theory, then the domain over which simplicity is taken cannot
be restricted to grammar alone and must include acquisition
theory-otherwise the grammarian merely purchases simplicity
a t the psychologist’s expense. (1965:491).

Slobin (in p r e s s ) objects to Braine’s proposal, citing evidence

from a variety of languages. Bever, Fodor, and Weksel (1965a,
1965b), argue that no dominant patterns of word order exist for the
child to generalize from, even in a language such as English, and
that word ordering also occurs during language acquisition when the
language has free word order. They say that the child must learn
abstract structures for which no word order patterns exist in the
data to which he is exposed. Answering this last criticism, Braine
(1965) points out that data do exist and that closer attention must
be paid t o how the child uses these data in the process of acquir-
ing language.

Nativist Theories
Lenneberg (1967) proposes a theory of language acquisition
heavily buttressed by biological evidence from studies of normal
language development in children and of abnormal language develop-
ment brought about congenitally, as in nanocephalic dwarfism, or
environmentally, as i n brain damage or aphasia. He emphasizes
the development of the organism’s capacities and shows how these
6 LANGUAGE LEARNING, VOL. 21, NO. 1

mature along a fairly fixed schedule. Language emerges during this

maturational process when anatomical, physiological, motor, neural,
and cognitive developments allow it to emerge. Every child must
learn the specific details of the language of his community, but the
ability to learn language is innate and part of the biological endow-
ment of the organism. The learning mechanisms, such as certain
modes of perception, abilities in categorization, and capacities for
transformation, a r e biologically given. According t o Lenneberg,
the child “resonates” to the language of his environment during
the acquisition process; however, he never clearly specifies exactly
what resonance is. One of Lenneberg’s most interesting observa-
tions is that there is a critical, biologically-determined period for
language acquisition between the ages of two and twelve.
Since Lenneberg is interested in the biological bases of lan-
guage acquisition, he has almost nothing to say about how particu-
lar linguistic items are learned, except to deny that statistical
probability and imitation are important in the process. He claims
that language acquisition is a natural activity, much as learning to
walk is a natural activity. Both activities occur universally unless
a pathological condition exists. Learning, as this t e r m is tradi-
tionally defined, is not involved. Instead, Lenneberg carefully locks
language acquisition into the general biological development of the
organism.
McNeill (1966, 1968, 1970a, 1970b) takes a rather different
nativist position toward language acquisition. He s a y s that anyone
who wishes to study the problem of language acquisition must begin
with a knowledge of what it is that the child must acquire:

A major requirement f o r any theory of language acquisition is

that it explain a known phenomenon, which means that theories
of development must be related to particular grammatical an-
alyses, to particular theories about language itself. (1968:406)

McNeill claims that the child must acquire a generative-transfor-

mational grammar. Following N. Chomsky (1957, 1965), he a s k s
what intrinsic properties must a device, a Language Acquisition
Device (LAD), possess to acquire such a grammar from the corpus
of utterances to which it is exposed:
LAD is, of course, a fiction. The purpose in considering i t is
to discuss real children, not abstract ones. We can accomplish
this because LAD and children present the same problem. LAD
is faced with a corpus of utterances from which i t develops a
grammar on the basis of some kind of internal structure. So do
children. We can readily posit that children and LAD arrive at
the same grammar from the same corpus, and stipulate that
 LANGUAGE ACQUISITION 7
children and LAD therefore have the same internal structure, a t
least within the limits that different children may be said to have
the same internal structure. Accordingly, a theory about LAD is
ips0 facto a theory about children. (1970a:71)
The child must possess certain innate abilities, otherwise it is im-
possible to explain how the random, finite linguistic input into the
child results in the output of linguistic competence.
According to McNeill, one innate property of the LAD is the
ability to distinguish speech sounds from other sounds in the en-
vironment. A second property is the ability to organize linguistic
events into various classes which can later be refined. This ability
allows for the development of both the phonological and syntactic
systems. One of the innate organizing principles is the concept of
the “sentence.” A third innate property is knowledge that only a
certain kind of linguistic system is possible and that other kinds
are not. McNeill claims that the child is born with an innate
knowledge of linguistic universals. He distinguishes (1970a) between
what he calls “weak” linguistic universals (reflections in language
of universal cognitive abilities) and “strong” linguistic universals
(reflections in language of specific linguistic abilities). He is more
interested in the latter and s e e m s skeptical of any claims advanced
by cognitive theorists about the former. A fourth property is the
ability to engage in constant evaluation of the developing linguistic
system s o as to construct the simplest possible system out of the
linguistic data that a r e encountered.
In a n attempt to justify his position, McNeill attacks S-R theory
on the grounds that language acquisition is beyond its domain:

Because S-R theory is s o limited, the problem of language acqui-

sition simply falls beyond its domain. This in itself is not a
serious matter, Not all psychological theories need account for
language acquisition, More serious, however, is the fact that the
application of S- R principles causes theorists to redefine language
in such a way a s to make the phenomenon fit the theory. There
is perhaps some irony in this outcome of modern empiricism.
(1968:412)

McNeill also argues against the importance of the frequency of

stimuli in language acquisition, using examples from Japanese, and
against the importance of imitation. He claims that theories re-
quiring imitation fail to explain why only certain responses occur.
He criticizes Braine for ignoring the essential transformational
nature of grammatical structure. Moreover, to Lenneberg’s notion
of a biological foundation for language, he adds a strong cognitive
“content” component in the form of a structure for the mind that
8 LANGUAGE LEARNING, VOL. 21, NO. 1

allows only certain kinds of language learning to occur. The organ-

ism has the capacity to learn and to generalize, but must realize
this capacity within certain innate constraints that are suggested by
a particular linguistic theory.
McNeill actually s a y s very little about the mechanisms of ac-
quisition. In addition, his claim that in the earliest stages the child
speaks in the universal base structures of a generative-transforma-
tional grammar may not be linguistically sound. His further claim
that the child “honors” grammatical distinctions before actually
making them has been attacked as invalid by Bloom (in press).

Cognitive The or ie s
Like Fodor (1966), Slobin (1966a, 196613) does not subscribe to
nativistic theories of language acquisition. He says:

It seems to me that the child is born not with a s e t of linguistic

categories but with some sort of process mechanism-a set of
procedures and inference rules, if you will-that he uses to process
linguistic data. (1966b:87-88)

Slobin regards language acquisition as a n active process in which

certain abilities of the child develop. One is the cognitive ability
t o deal with the world; a second is the mental ability to retain
items in short t e r m memory, to store items in long t e r m memory,
and to process information increasingly with age. The develop-
ments control the pace of language acquisition. Others are impor-
tant too, such as the ability to segment utterances into sounds and
meanings, and then to combine and recombine these segments, the
ability to isolate meaning units, and the ability to make wide gen-
eralizations before attempting to accommodate exceptions. However,
according to Slobin, general cognitive and mental development is the
critical determinant of language acquisition.
Slobin marshals evidence from a variety of languages to sup-
port his position that language acquisition is one kind of general
development, and that the general principles involved in the latter
must be recognized. He differs from McNeill in the way he uses
linguistic data. McNeill uses such data to postulate the presence
of innate linguistic principles; Slobin u s e s the same data to support
innate principles of cognition. F o r example, in discussing McNeill’s
proposal concerning the child’s innate knowledge of substantive lin-
guistic universals, Slobin says:

Perhaps all that is needed i s an ability to learn certain types of

semantic or conceptual categories, the knowledge that learnable
semantic criteria can be the basis for grammatical categories,
 LANGUAGE ACQUISITION 9

and, along with this substantive knowledge, the formal knowledge

that such categories can be expressed by such morphological de-
vices a s affixing, sound alternation, and so on. The child’s ‘pre-
programming’ for substantive universals is probably not for spe-
cific categories like past, animate, plural, and the like, but con-
sists rather of the ability to learn categories of a certain as-yet-
unspecified type. (1966b:89)

Slobin differs from the behaviorist theories in that he is a cognitive-

learning theorist who regards the human learner as a n active par-
ticipant in learning rather than as a relatively passive reactor to
external stimuli:

The important advances in language development thus seem to be

tied to such variables a s increasing ability to perform a number
of operations in a short time, increasing short-term memory
span, and increasing cognition of the categories and processes of
human experience. In fact, it may be that strictly linguistic ac-
quisition i s completed by age three o r so. Further development
may reflect lifting of performance restrictions and general cogni-
tive growth, without adding anything basically new to the funda-
mental structures of syntactic competence. We have begun to
gather data on the earliest stages of language development. We
have very little data on later stages. And our understanding of
the mental processes underlying the course of this development
is extremely rudimentary indeed. At this point I believe we a r e
in need of much more data on children’s acquisition of various
native languages. . ..
Cromer (1968) provides further evidence of the role of cogni-
tive abilities in determining the language the child can use. From
a study of the development of temporal reference in two children
over a four-year period, he notes that several new types of refer-
ence to points in time begin to occur regularly a t about the age
of 4 to 4$ for each child. Viewed together, these new forms indi-
cate that the child has greatly expanded his range of temporal ref-
erence and increased his sense of the possible relations between
times. Cromer notes that the ability develops to express events
out of chronological order, to make statements about possibility,
and to relate one time to another time. He hypothesizes that a
single factor alone accounts for the observed linguistic changes:
the child suddenly finds that he can free himself from the immedi-
ate situation and the actual order of events and can imagine him-
self at other points in time and view events from that perspective.
This increase in his cognitive ability enables him to express new
meanings, and he immediately m a s t e r s the necessary syntactic
apparatus to do so.
10 LANGUAGE LEARNING, VOL. 21, NO. 1

There are even stronger claims for a cognitive basis to lan-

guage acquisition than those made by Slobin. Schlesinger (in p r e s s )
claims that linguistic structures a r e “. . .
determined by the innate
cognitive capacity of the child,” and Sinclair-de-Zwart (1968) claims
that ‘‘ linguistic universals exist precisely because thought struc-
tures a r e universal.” However, no empirical evidence apparently
exists to confirm either claim.

Linguistically-oriented Theories Versus Learning-oriented Theories

In trying to develop a theory of language acquisition, an in-
vestigator is faced with a- fundamental decision conierning a start-
ing point. Should he begin by accepting certain principles from
linguistics or certain principles from psychology? In other words,
should he begin by saying, as McNeill does, that what must be ex-
plained is how the child acquires a generative-transformational
grammar, o r by saying, as Staats and Staats do, that a behavior-
istic theory employing such principles as association-formation and
stimulus and response generalization should be able to account for
language acquisition? McNeill proceeds t o dismiss current learn-
ing theories as inadequate to explain the special behavior or knowl-
edge which he claims comprises linguistic competence, and Staats
and Staats proceed to ignore certain kinds of linguistic data.
Braine attempts to fasten on to the best in both linguistic
theory and learning theory. He claims that each must, if necessary,
be changed to accommodate the other. The two extremes of the
general positions taken by McNeill and Staats and Staats are prob-
ably equally untenable, for at one extreme the interest is basically
in the linguistic description of child !anguage with very little con-
c e r n for learning principles, and a t the other extreme the interest
is in applying learning principles derived from experiments with
animals to the one behavior that no animal exhibits, linguistic be-
havior. Neither McNeill nor Staats and Staats take these extreme
positions, but sometimes they seem t c be approaching them. In
the circumstances, Braine’s middle ground may appear to be more
attractive; however, both linguists and learning theorists find his
proposed compromises unacceptable.
Fodor acknowledges the necessity for postulating some innate
structure without committing himself as to whether this structure
derives from innate linguistic principles or innate learning princi-
ples:
.
. . the child must bring to the language learning situation some
amount of intrinsic structure. This structure may take the form
of general learning principles o r it may take the form of rela-
tively detailed and language-specific information about the kind of
 LANGUAGE ACQUISITION 11

grammatical system that underlies natural languages. But what

cannot be denied is that any organism that extrapolates from its
experience does so on the basis of principles that a r e not them-
selves supplied by the experience. (1966 :106)

Slobin’s position is l e s s equivocal. He considers the child to be

endowed with the cognitive capacity to perform extremely compli-
cated tasks. The child accomplishes the complicated task of lan-
guage acquisition according to general laws of development, learn-
ing, and perception. Consequently, he brings a particular capacity
to the task rather than knowledge of a s e t of innate linguistic prin-
ciples.

Four Controversial Issues

It is of interest to examine how various theories deal with the
problems of the frequency of stimuli, the place of imitation, the
role of expansion, and the function of meaning in language acquisi-
tion. In this way the theories can be shown to differ in certain
important respects, and some preliminary assessment can be made
of their relevance to beginning reading instruction.
The relative frequency of stimuli must be important in any
behavioristic theory of learning. The most frequently occurring
words and structures in the language should be acquired first by
the child. However, the empirical evidence of language acquisition
contradicts this expectation. Telegraphic speech, for example, omits
the most frequently occurring words in the language, and investi-
gators agree that every child goes through a “telegraphic” stage.
There must be some reason for the existence of such speech, but
it appears to have little to do with the frequency of stimuli in the
environment .
McNeill (1966, 1968) also argues that Japanese children acquire
a l e s s frequent grammatical marker ga before a more frequent
marker wa because ga is important as a deep subject marker
whereas wa is not. He later (1970a:30-31) offers a rather differ-
ent interpretation of the same data in accordance with the kinds
of predicates (intrinsic with w a and extrinsic with g a ) that the child
is capable of forming at the age when w a and ga appear in speech.
Slobin (in p r e s s ) cites similar examples from other languages. If
frequency is not important and certain kinds of learning occur in
a definite progression, then the crucial issue is to account for this
learning and the progression. McNeill argues that the structure of
language and of the child’s mind controls the learning, whereas
Slobin argues that the child’s cognitive and mental capacities at
each stage regulate his ability to learn. However, each agrees
12 LANGUAGE LEARMNG, VOL. 21, NO. 1

with the other that the relative frequency of stimuli is of little

importance in language acquisition.
Imitation in the sense of modeling also holds a n important place
in behavioristic theories of learning in which some kind of modeling
of behavior must occur. While there is evidence that children do
practice language (Weir 1962) and do repeat some of the utterances
of persons around them, they do not imitate indiscriminately. For
example, Weir’s child produced certain imitations but also made
many variations on the imitated utterances. Babies do not imitate
sounds in general, but they do respond quickly to human sounds.
Lenneberg, Rebelsky, and Nichols (1965) a l s o report that the pre-
linguistic vocalization behavior of deaf infants is not different from
that of hearing infants. Therefore, imitation is not a critical factor
in this very early stage of development, as it is, for example, in
Jenkins and Palermo’s theory. Menyuk (196313) notes that the ability
to imitate depends on the acquisition of some prior ability since
children give evidence of various difficulties in imitating utterances.
Utterances such as allgone shoe, allgone lettuce, and allgone vita-
mins reported by Braine (1963b) also argue against imitation and
for some other ability, for no such sentences occurred in the en-
vironment of the child who produced them. Similar evidence is re-
ported by Brown and Bellugi (1964) and by Miller and Ervin (1964).
One obvious constraint upon the child’s ability to imitate is the
limitation imposed by his short t e r m memory span. It is also very
difficult to explain how simple imitation leads to development. Obvi-
ously, some issue has been skirted. Young children a r e actually
rather poor imitators, as McNeill (1966) shows in the following
sample:

The signs a r e that sometimes a child’s tendency to assimilate

adult models into his current grammar is so strong that even
when he makes a deliberate effort to copy adult speech, the ef-
fort may a t f i r s t fail. One child, in the phase of producing double
negatives while developing the negative transformation, had the
following exchange with his mother:
Child: Nobody don’t like me.
Mother: No, say “nobody likes me.”
Child: Nobody don’t like me.
 LANGUAGE ACQUISITION 13

(eight repetitions of this dialogue)

Mother: No, now listen carefully; say “nobody likes m e . ”

Child: Oh! Nobody don’t likes me.

The exchange is interesting because it demonstrates the relative

impenetrability of the child’s grammar to adult models, even
under the instruction (given by the mother’s “no”) to change.
The child behaves a t first a s if he did not perceive the differ-
ence between his mother’s sentence and his own, though later,
when the mother supplied great emphasis, the child recognized a
distinction. With this much delay in introducing changes, spon-
taneous imitations a r e bound not to be grammatically progres-
sive because they consist only of a single exchange. The fact
that a change ultimately was made, however, illustrates that
children can profit from adult models. (1966:69)

McNeill does not deny the importance of models to the child in his
learning, but does show that simple imitation of such models pro-
vides a n inadequate explanation of linguistic development. Ervin
(1964) demonstrates that imitations by children a r e not grammati-
cally progressive, f o r they are l e s s complicated syntactically than
concurrent free utterances. Menyuk (1963a), Lenneberg, Nichols,
and Rosenberger (1964), and Slobin and Welsh (1967) all report that
children produce in imitation only what they produce in spontaneous
speech even to the extent of reducing adult-given sentences to the
forms they are currently producing.
Still another difficulty with relying heavily on imitation in any
theory of language acquisition is the fact that much of the speech
to which the child is exposed is considerably fragmented. Yet he
learns t o filter out poor examples in forming his grammar. This
accomplishment is at least as difficult to explain as is the accom-
plishment of being able to react to more complex utterances than
he can produce. Some factor other than imitation must be involved
in each case. Lenneberg (1962) points out one specific case in
which imitation could not have been involved in language acquisi-
tion, that of a boy with a congenital motor disability that prevented
him from speaking. However, since the same boy could understand
complicated instructions, neither imitation nor reinforcement could
be used t o explain his abilities. The language of the environment
in which the child finds himself is vitally important to him in his
14 LANGUAGE LEARNING, VOL. 21, NO. 1

acquisition of language. But direct imitation of that language s e e m s

not to occur except in rather small amounts.
The role of expansion in language acquisition is a still more
complicated issue. Parents do correct and expand the speech of
their children. However, there is evidence that children are not
particularly receptive to direct instruction in language, as is obvi-
ous in the quotation cited above from McNeill. Although correc-
tions might be expected to extinguish certain undesirable linguistic
behaviors, they a r e unlikely to promote desirable ones. Expansions
might be helpful in stimulating linguistic development, and some
agreement exists that middle-class mothers expand their children’s
speech about 30% of the time and that such use of expansion forms
a part of the normal mother-child relationship. Cazden (1965) tested
the hypothesis that expansions of children’s utterances would aid
language acquisition more than would comments on their utterances,
which she called models, and that both would produce better results
than no expansion or modeling responses. She divided twelve 2;-
year-old children into three groups: the first group received inten-
sive and deliberate expansions; the second group received qualita-
tively equal exposure to well-formed sentences that were models
not expansions; and the third group received no special treatment
a t all. Her experiment lasted twelve weeks. The results do not
show quite the expected differences in that modeling, not expansion,
was more effective. That is, semantically-enriched responses were
more effective than syntactically-enriched responses. However, a
more recent study by Feldman and Rodgon (1970) reports results
a t variance with those of Cazden. In a further study, Brown, Cazden,
and Bellugi (1968) analyzed the conversations of mothers and chil-
dren aged one to four y e a r s to determine what happens during such
conversations. They report that the syntactic correctness or in-
correctness of a child’s speech does not control the mother’s ap-
proval or disapproval. Rather the truth or falsity of the utterance
does. They conclude that parents tend to reward true statements
and punish false ones; however, somewhat surprisingly, the result
is the eventual production of syntactically correct sentences.
Deliberate expansion of children’s language by adults would
seem to be one of the most important possible influences on lan-
guage development. However, the evidence does not confirm this
hypothesis. Having considered the evidence from research in the
use of both imitation and expansion, Slobin (1968) concludes that
there is little evidence t o support imitation. However, he takes a
more positive attitude toward expansion:
 LANGUAGE ACQUISITION 15

It has been suggested that frequency of parental expansion of

child speech may be related to such variables a s social class
and education, and, in turn, be partly responsible for differ-
ences in language acquisition and ability in children of different
socioeconomic backgrounds. The issue is certainly complex, and
w e a r e f a r from being able to determine the functions-if any-
of expansion and imitation in the human child’s remarkable ac-
quisition of language. Until the necessary data a r e amassed, I
would still like to believe that when a child hears an adult ex-
pansion of his own speech he learns something important about
the structure of his language. (1968:443)

The results as a whole argue more for the acceptance of language-

acquisition theories like those of Lenneberg, McNeill, and Slobin
than they do f o r those of Braine and Staats and Staats, and more
for the importance of some kind of innate linguistic or cognitive
structure than of the actual stimuli encountered in the environment.
Studies of language acquisition tend to focus on the acquisition
of phonology o r syntax. The place and function of meaning in lan-
guage acquisition have largely been ignored. However, meaning is
today assuming greater importance in studies of language acquisi-
tion.
Following a comprehensive review of Russian data on language
development in children, Slobin (1966a) suggests that the order of
emergence of various syntactic categories depends on their rela-
tive semantic difficulty rather than on their grammatical complexity.
The f i r s t grammatical distinctions to appear are those like the
singular-plural distinction that make some concrete reference to
the outside world. Later to emerge a r e the diminutive suffixes of
nouns, imperatives, and categories based on relational criteria,
such as the case, tense, and person markings of verbs. Condi-
tional f o r m s of the if-then variety a r e not learned until near the
end of the third year. Still other abstract categories of quality
and action continue to be added until the age of seven. Slobin
argues that semantic complexity rather than grammatical difficulty
determines the developmental sequence. Grammatical gender in
Russian is the most difficult of all the categories f o r the child to
master since it has almost no semantic correlates. No rules exist
that the child can discover to make the learning easier, so the
acquisition of gender is a long, drawn out process. Slobin con-
cludes: “The semantic and conceptual aspects of grammatical
classes thus clearly play an important role in determining the order
of their development and subdivision.” (1966a: 142).
Telegraphic speech is full of “contentive” words. Slobin (1971:
44-46) shows some of the semantic range of telegraphic speech in
16 LANGUAGE LEARNING, VOL. 21, NO. 1

various languages (English, German, Russian, Finnish, Luo, and

Samoan). Following a n analysis of such speech, a reexamination
of the data from the pivot g r a m m a r s of investigators such as
Braine, and some work of h e r own, Bloom (1970, i n p r e s s ) argues
that the evidence indicates that semantic competence outstrips syn-
tactic competence. Her own research showed that noun-noun com-
binations in the speech of very young English children expressed at
least the following five relations: conjunction (block doZZy),attribu-
tion party hat), genitive (daddy hat), subject-locative (sweaterchair),
and subject-object ( m o m m y book). She also found that a n utterance
such as no truck could have various meanings, which themselves
showed a n order of emergence: “nonexistence” (There‘s no truck
here) preceding “rejection” fJ don‘t want a truck), which in turn
precedes “denial” @trsnot a truck ; it‘s something else). She con-
cludes that the child’s underlying semantic competence is more dif-
ferentiated than the surface forms of his utterances, because he is
aware of more types of meaning relationships than he can reveal
through the linguistic devices he controls. Before he develops these
devices, his two-word utterances can be properly interpreted only
through the use of the non-linguistic context. Quite often a young
child must produce a s e r i e s of short utterances in order to convey
information that a n adult or an older child expresses in a single
utterance. For example, he might say raisin t h e r e / buy more
grocery store / raisins / buy more grocery store/ grocery store/
raisin a grocery store instead of one sentence about buying more
raisins at the grocery store. Consequently, Bloom (1970) claims
that three components operate in the development of language com-
pete nc e : cognitive - pe r c e ptual development, linguistic experience ,
and non- linguistic experience. She notes that these components
converge during the child’s development.

An Assessment of the Theories

The studies reported by McCarthy encompass massive quanti-
ties of data but lack clearly defined theories of language acquisition.
A concern f o r such theories is a fairly recent development in
studies of language acquisition. However, all such theories have
a t least the weaknesses of lack of detail and lack of empirical
validation. They are all very general, often being little more than
series of claims about what must be, the claims being supported
by reference to carefully selected data often acquired from no more
than a few children. Consequently, they a r e often hardly any more
convincing than f o r m e r presentations of large quantities of data that
really make no claims a t all.
Recently proposed theories make either a language or learning
 LANGUAGE ACQUISITION 17

component central. Making a language component central requires

postulation of a strong innate predisposition toward the acquisition
of very specific kinds of linguistic facts, for the child is assumed
to “know” much about language in general before any learning of
specific details begins. Environmental factors a r e relatively un-
important in such theories. On the other hand, older behavioristic
learning theories hold the environment to be extremely important
in providing language stimuli and controlling the learning that oc-
curs. According to such theories, language acquisition is achieved
through such processes as association and response generalization.
The child makes little or no active contribution to the total process
and learns language in much the same way as he learns anything
else.
A less extreme position is that language acquisition is unique
because language is different from anything else that is learned,
but that the learning requires u s e of many of the same principles
as other kinds of learning. In this case, the theory may have a
large biological component that emphasizes the importance of cer-
tain kinds of universal neurological and physiological developments.
Or it may assume the availability of this component and emphasize
the kinds of meaningful situations that stimulate language acquisition
and the cognitive limitations that human development places on the
acquisition process. Unfortunately, since meaning has long been a
stepchild in linguistics and cognitive theory a poor relation in psy-
chology, it is difficult a t present to fill out the details of any such
theory.
An evaluation of the importance of such factors as frequency,
imitation, and expansion in language acquisition leads to the rejec-
tion of any kind of monolithic behavioristic theory. However, it
does not eliminate linguistically-based theories nor does it contra-
dict cognitively-based ones. The evaluation reveals how unimportant
each of the factors is in language acquisition, and indicates the
necessity of crediting the child with some kind of innate knowledge
or capacity. The difficulty with the innate knowledge hypothesis is
that investigators like McNeill have very little to say about the
mechanisms through which that knowledge reveals itself, nor do they
t r y to relate language learning to other kinds of learning. The re-
sult is something less than a parsimonious view of total human
development. The advantage of the innate capacity hypothesis is
that general laws of learning, but not exclusively behavioristic ones,
can be used to explain both language acquisition and other kinds of
learning. Sachs (in press) summarizes this problem as follows:
Theories of language acquisition that consider only the linguistic
aspect will not be able to explain why the child learns new forms
18 LANGUAGE LEARNING, VOL. 21, NO. 1

when he does, o r in fact why he ever changes his form of ex-

pression. It i s only through more research on the complex re-
lationship between cognitive development and language acquisition
that we will have a full understanding of either. Hopefully in
the future w e will find more studies of this type, and a closer
communication between psycholinguists and pyschologists study-
ing other aspects of child development.

The linguistically- based theories all have one serious drawback

in that they are concerned with the ideal child. Theories recogniz-
ing individual and group differences are ignored in favor of theories
that t r y to account for the development of abstract linguistic com-
petence. Social, motivational, and cultural variables are all ignored.
The child is said to have acquired his basic linguistic competence
by the age of five o r six. While performance is acknowledged to
vary from child to child, such variability, whatever its cause, is
ignored, often under the guise of “performance” differences, which
are at best of peripheral interest. The result is a deliberate bias-
ing of the theories toward accommodating one set of factors in
language acquisition and ignoring almost all others.

LANGUAGE ACQUISITION AND BEGINNlNG READING

Language Acquisition after Age Six

Although many linguists claim that the major part of language
acquisition takes place in the y e a r s between the ages of one and
four, children who enter school do not have the linguistic abilities
of adults and the linguistic abilities of adults change, and some-
times develop, during their lives. It is of interest to know the
precise differences between the linguistic abilities of children en-
tering school and of adults. Numerous investigators have shown
that significant language development still occurs in all children
after the age of five o r six, among them Harrell (1957), Strick-
land (1962), Loban (1963), Menyuk (1963b), and O’Donnell, Griffin,
and Norris (1967).
In a recent study C. S. Chomsky (1969) points out several
grammatical developments that occur during the y e a r s that follow
six: a grasp of the difference between the eager to s e e and easy
to s e e constructions; a realization that ask and tell require differ-
ent syntactic constructions; the ability to handle relationship requir-
ing and and although; and a control of pronominalizations. Kessel
(1970) used a Piaget-type interview technique similar to that used
by C. S. Chomsky i n further work on some of the same problems.
His study confirms her results but also reports evidence of a
 LANGUAGE ACQUISITION 19

somewhat e a r l i e r mastery of the more complex constructions.

Menyuk (1969) points out other examples in which a more compli-
cated structure is learned later than a less complicated one.
However, in every case it is possible to argue that the linguistic
development has not occurred because the cognitive capacities of
the child do not allow it rather than because the structure which
is learned second is more complicated than the one which is
learned first. Of course, since it is also possible to argue that
the structure learned second is grammatically more complicated,
the temptation is to postulate a linguistic rather than a cognitive
constraint on development, particularly when the investigator is
linguistically- oriented.
Two linguistic abilities that children of about age six appear
to have a r e those to overdiscriminate and to overgeneralize. N.
Chomsky (1964) points out that they have very sharp abilities to
discriminate among phonetically close stimuli. Miller and Ervin
(1964) and Ervin (1964) say that they tend to eliminate from their
language irregular but correct inflections in favor of regular but
incorrect ones for a while. Slobin (in press), citing evidence
mainly from Russian, discusses this same phenomenon, which he
calls “inflectional imperialism.”
Bever, Mehler, and Valian (1968) report that children aged two
to four temporarily overgeneralize newly acquired semantic strate-
gies. There is also some agreement that children do not interpret
“ s a m e ” and “different” in the way that mature adults do, nor a r e
they able to work in a conscious analytic fashion with language, as
many adults can. Slobin (in press) points out that the Russian data
he analyzed provide evidence that any kind of direct instruction in
the analysis of language is rather ineffective with children.
In one crucial area for any kind of reading instruction that
r e l i e s on the relationship of individual sounds to symbols, the ac-
quisition of phonology, six-year olds have not mastered the system
that educated literate adults appear to have mastered (Chomsky and
Halle 1968, C. S. Chomsky 1970). The abilities of the two groups
appear to be quite different. Indeed language acquisition in this
area appears to depend on the acquisition of the ability to read,
but this is the only pluce where this particular dependency occurs.

Some Important Differences Between Language Acquisition and

Beginning Reading
Whatever theory of language acquisition a n investigator sub-
s c r i b e s to, behavioristic, nativistic, or cognitive, he must readily
admit that important differences exist between the acquisition of
language and the acquisition of beginning reading skills. Staats
20 LANGUAGE LEARNING, VOL. 21, NO. 1

and Staats (1962), Carroll (1966), and Natchez (1967) a r e among

those who point out some of the specific differences.
Language is acquired gradually and the acquisition process is
probably never completed, for something always remains to be
learned. The process is also one that had no conscious beginning
point for the child. On the other hand, learning to read often has
a sudden onset for children, although some a r e fortunate to avoid
this kind of introduction. Even though some of the cognitive and
motor skills necessary for reading have been developed for other
activities, the child is often required to put them altogether rather
abruptly, in learning to read in a formal school setting.
The level of anxiety in the context in which learning to read
takes place may also be quite high: the anxiety of the parent,
teacher, and the child. Little such anxiety is manifested during
the process of learning t o talk. Certainly, it is the rare child who
exhibits anxiety, and, if the occasional parent is anxious about a
particular child’s speech, this anxiety seems to have little influ-
ence on the child’s language development. There is a l s o often a
concomitant assignment of blame for any “failure” that occurs in
beginning reading instruction. Children are not “blamed” when
they fail to acquire language; rather they are given special help.
Reading instruction is very formal and deliberate. Language,
however, is learned informally and unconsciously from a wide-
range of stimuli. No deliberate instruction is necessary. Language
is not learned from programmed stimuli, from making conscious
distinctions among stimuli, from learning “about” language, and
from acquiring control of a variety of analytic and synthetic tech-
niques. While controversy does exist as to the function of linguis-
tic stimuli in language acquisition, there is agreement that such
stimuli vary in both form and content in ways that are not well
understood, but which the child is well able to handle.
The usual reinforcements experienced by literate adults for
reading may be irrelevant for many children in the beginning read-
ing stages: the benefits are often too abstract, distant, and mean-
ingless, and the effort to be expended for such remote ends may
seem to be quite wasteful and unpleasant to the child. On the other
hand, the benefits of learning to speak are too obvious to mention.
The two activities a r e also different in certain other ways.
Learning to read depends on the acquisition of special skills in
visual discrimination. The redundancies in the two language sys-
tems that a r e involved are a l s o different, as is quite often the
content, that is, the meanings that are conveyed. Writing is not
simply speech written down: it is more abstract than speech in
content; it usually employs carefully edited and controlled language
 LANGUAGE ACQUISITION 21

for different reasons from speaking; and it functions rather differ-

ently in the lives of the recipients of the message. Vygotsky (1962)
writes as follows on these very points, but in connection with writ-
ing rather than reading:

Written speech is a separate linguistic function, differing from

oral speech in both structure and mode of functioning. Even its
minimal development requires a high level of abstraction. ...
Our studies show that it is the abstract quality of written lan-
guage that is the main stumbling block, not the underdevelop-
ment of small muscles o r any other mechanical obstacles.

Writing is also speech without an interloctor, addressed to an

absent o r an imaginary person o r to no one in particular-a
situation new and strange to the child. Our studies show that he
has little motivation to learn writing when we begin to teach it.
He feels no need for it and has only a vague idea of its useful-
ness. In conversation, every sentence is prompted by a motive.
Desire o r need lead to request, question to answer, bewilder-
ment to explanation. The changing motives of the interlocutors
determine at every moment the turn oral speech will take. It
does not have to be consciously directed-the dynamic situation
takes care of that. The motives for writing a r e more abstract,
more intellectualized, further removed from immediate needs.
In written speech, we a r e obliged to create the situation, to rep-
resent it to ourselves. This demands detachment from the actual
situation.

Writing also requires deliberate analytical action on the part of

the child. In speaking, he is hardly conscious of the sounds he
pronounces and quite unconscious of the mental operations he
performs. In writing, he must take cognizance of the sound
structure of each word, dissect it, and reproduce it in alphabeti-
cal symbols, which he must have studied and memorized before.
(1962 :98-99)

Reid (1966), Meltzer and Herse (1969), and Downing (1970) all point
to the confusion that children often experience in learning to read.
Evidently, many children do not understand what reading is, or what
they are supposed to be doing, o r what the t e r m s mean that are
used in the instructional process.
The usual methods of reading instruction employ imitation,
repetition, control of stimuli, correction, and expansion, exactly
those factors examined earlier in relation to the acquisition of
language. These factors were found not to be very important in
language acquisition; however, they are very important in reading
instruction. Of course, instruction implies some kind of method-
22 LANGUAGE LEARNING, VOL. 21, NO. 1

ology, s o the reason for their existence is obvious. Yet, it would

be well to subject that methdology to periodic critical assessment
in the light of the latest findings from relevant disciplines. Of
course, one can also argue that since language acquisition and
learning to read are quite different tasks, these factors may still
be important in the teaching of beginning reading.
Finally, language acquisition does not cease a t the age six.
Consequently, some kinds of acquisition overlap with learning to
read. However, little is known about the extent of this overlap,
for the later stages of language acquisition a r e even more of a
mystery than are the e a r l i e r stages. It may be that more than one
of these stages depends on the child’s acquiring certain reading
abilities just as beginning reading ability quite definitely depends on
the acquisition of considerable linguistic competence. However, this
acquisition h a s occurred in six-year olds except in r a r e pathologi-
cal cases.

CONCLUSION

The theories of language acquisition that a r e available to us

today are largely irrelevant in deciding issues in beginning reading
instruction or even in devising models of the reading process.
Moreover, reading failure cannot easily be linked to deficiencies
in language acquisition, for children who are asked to learn to read
a r e almost invariably well on the way to linguistic maturity.
Reading methods themselves a r e almost unrelated to theories
of language acquisition. Both phonics and whole-word methods de-
pend on the possession of certain language abilities which all chil-
dren of six apparently do have. What they might not have a r e some
of the cognitive abilities that the methods require: abilities to make
certain kinds of discriminations, to form generalizations, and to
verbalize knowledge. Furthermore, much of what is taught “about”
language in such methods is antiquated and not very useful to any-
one, particularly to six-year olds.
Reading is often taught to improve language. Research has
long demonstrated that such teaching is generally ineffective. Some
linguistic skills apparently derive from the acquisition of the skills
of literacy, but these skills appear to be few and certainly do not
seem to be acquired during the critical period of beginning reading
instruction.
 LANGUAGE ACQUISITION 23

REFERENCES

Bever, T. G., Fodor, J. A., and Weksel, W. 1965a. Is linguistics empiri-

c a l ? Psychological Review 72.492-500.
Bever, T. G., Fodor, J. A,, and Weksel, W. 1965b. On the acquisition of
syntax: A critique of “contextual generalization. ‘ I Psychological R e -
view 72.467-482.
Bever, T. G., Mehler, J. R., and Valian, V. V. 1968. Linguistic capacity
of very young children. Unpublished manuscript, ERIC ED 018-796.
Bloom, L. 1970. Language Development: Form and Function in Emerging
Grammars. Cambridge, Mass. : M.I.T. Press.
Bloom, L. Why not pivot g r a m m a r ? The Journal of Speech and Hearing
Disorders, (in press).
Braine, M.D.S. 1963a. On learning the grammatical order of words. Psy-
chological Review 70.323-348.
Braine, M.D.S. 1963b. The ontogeny of English phrase structure: the first
phase. Language 39.1-13.
Braine, M.D.S. 1965. On the basis of phrase structure: a reply to Bever,
Fodor, and Weksel. Psychological Review 72.483-492.
Brown, R. 1965. Social Psychology. New York: F r e e Press.
Brown, R., and Bellugi, U. 1964. Three processes in the child’s acquisi-
tion of syntax. Harvard Educational Review 34.133-151.
Brown, R., and Berko, J. 1960. Psycholinguistic research methods. In
P. H. Mussen (Ed.), Handbook of Research Methods in Child Develop-
ment. New York: Wiley & Sons.
Brown, R., Cazden, C. B., and Bellugi, U. 1968. The child’s grammar
from 1 to 3. In J. P. H i l l (Ed.), The 1967 Minnesota Symposium on
Child Psychology. Minneapolis : University of Minnesota Press.
Carroll, J. B. 1966. Language development in children. Sn Encyclopedia
of Educational Research. 744-750. New York: Macmillan.
Carroll, J. B. 1966. Some neglected relationships in reading and language
learning. Elementary English 43.577-582.
Cazden, C. B. 1965. Environmental assistance to the child’s acquisition of
grammar. Unpublished doctoral dissertation. Harvard University.
Chomsky, C. S. 1969. The acquisition of syntax in children from 5 to 10.
Cambridge, Mass.: M.I.T. Press.
Chomsky, C. S. 1970. Reading, writing, and phonology. Harvard Educa-
tional Review 40.287-309.
Chomsky, N. 1957. Syntactic Structures. The Hague: Mouton.
Chomsky, N. 1959. Review of B. F. Skinner, Verbal Behavior. Language
35.26-58.
Chomsky, N. 1964. Comments f o r project literacy meeting. Project Liter-
acy Reports 2.1-8.
Chomsky, N. 1965. Aspects of the Theory of Syntax. Cambridge, Mass.:
M.I.T. Press.
Chomsky, N., and Halle, M. 1968. The Sound Pattern of English. New
York: Harper & Row.
Cromer, R. 1968. The development of temporal references during the ac-
quisition of language. Unpublished doctoral dissertation. Harvard Uni-
versity.
Downing, J. 1970. The development of linguistic concepts in children’s
thinking. Research in the Teaching of English. 4.1.5-19.
24 LANGUAGE LEARNING, VOL. 21, NO. 1

Elkonin, D. B. 1958. Razvitie Rechi v. Doshkol’nom vozraste. (The De-

velopment of Speech in Preschool Age.) Moscow: Akad. Pedag. Nauk
RSFSR.
Ervin, S. M. 1964. Imitation and the structural change in children’s lan-
guage. In E. H. Lenneberg (Ed.), New Directions in the Study of Lan-
guage. Cambridge, Mass.: M.I.T. Press.
Ervin-Tripp, S. M. 1966. Language development. In M. Hoffman and L.
Hoffman (Eds.), Review of Child Development Research 2 . Ann Arbor:
University of Michigan Press.
Ervin-Tripp, S. M., and Slobin, D. I. 1966. Psycholinguistics. Annual Re-
view of Psychology 17.435-474.
Feldman, C. F., and Rodgon, M. 1970. The effects of various types of
adults’ responses in the syntactic acquisition of two- to three-year
olds. Unpublished paper. Department of Psychology, University of
Chicago.
Fodor, J. A. 1966. How to learn to talk: some simple ways. In F. Smith
and G. A. Miller (Eds.), The Genesis of Language: A Psycholinguistic
Approach. Cambridge, Mass.: M.I.T. Press.
Garrett, M., and Fodor, J. A. 1968. Psychological theories and linguistic
constructs. In T. R. Dixon and D. L. Horton (Eds.), Verbal Behavior
and General Behavior Theory. Englewood Cliffs, N. J. : Prentice-Hall.
Harrell, L. E., Jr. 1957. An inter-comparison of the quality and rate of
the development of the oral and written language in children. (Mono-
graphs of the Society for Research in Child Development 22.)
Jenkins, J. and Palermo, D. 1964. Mediation processes and the acquisition
of linguistic structure. In U. Bellugi and R. W. Brown (Eds.), The Ac-
quisition of Language. (Monographs of the Society for Research in Child
Development 29.1, Serial No. 92).
Kelley, K. L. 1967. Early Syntactic Acquisition. Santa Monica, Calif.:
The Rand Corporation.
Kessel, F. S. 1970. The role of syntax in children’s comprehension from
ages six to twelve. (Monographs of the Society for Research in Child
Development 35.6, (Serial No. 139).
Klima, E. S., and Bellugi, U. 1966. Syntactic regularities in the speech
of children. In J. Lyons and R . Wales (Eds.), Psycholinguistic Papers.
Edinburgh: Edinburgh University Press.
Lenneberg, E. H. 1962. Understanding language without the ability to
speak: a case report. Journal of Abnormal Social Psychology 65.419-425.
Lenneberg, E. H. 1967. Biological Foundations of Language. New York:
Wiley & Sons.
Lenneberg, E. H., Nichols, I. A., and Rosenberger, E. F. 1964. Primitive
stages of language development in mongolism. Proc. Assoc. Res. New.
Ment. Disease 42.119-137.
Lenneberg, E. H., Rebelsky, F. G., and Nichols, I. A. 1965. The vocaliza-
tion of infants born to deaf and to hearing parents. Vita Humana (Hu-
man Development) 8.23-37.
Loban, W. D. 1963. The Language of Elementary School Children. Cham-
paign, Jll.: National Council of Teachers of English.
McCarthy, D. 1954. Language development in children. In L. Carmichael
(Ed.), Manual of Child Psychology. New York: Wiley.
McNeill, D. 1966. Developmental psycholinguistics. In F. Smith and G. A.
Miller (Eds.), The Genesis of Language: A Psycholinguistic Approach.
Cambridge, Mass.: M.I.T. Press.
 LANGUAGE ACQUISITION 25

McNeill, D. 1968. On the theories of language acquisition. In T. R. Dixon

and D. L. Horton (Eds.), Verbal Behavior and General Behavior Theory.
Englewood Cliffs, N.J. : Prentice-Hall.
McNeill, D. 1970a. The Acquisition of Language: The Study of Develop-
mental Psycholinguistics. New York: Harper & Row.
McNeill, D. 1970b. The development of language. In P. A. Mussen (Ed.),
Carmichael’s Manual of Child Psychology. New York: Wiley.
Meltzer, N. S., and Herse, R. 1969. The boundaries of written words a s
seen by f i r s t graders. Journal of Reading Behavior 1.3-14.
Menyuk, P. 1963a. A preliminary evaluation of grammatical capacity in
children. Journal of Verbal Learning and Verbal Behavior 2.429-439.
Menyuk, P. 1963b. Syntactic structures in the language of children. Child
Development 34.407-422.
Menyuk, P. 1969. Sentences Children Use. Cambridge, Mass.: M.I.T.
Press.
Miller, W., and Ervin, S. M. 1964. The development of grammar in child
language. In U. Bellugi and R . Brown (Eds.), The Acquisition of Lan-
guage. (Monographs of the Society f o r Research in Child Development
29.1, Serial No. 92).
Natchez, G. 1967. From talking to reading without really trying. The
Reading Teacher 20.339-342.
O’Donnell, R. C., Griffin, W. J., and Norris, R. C. 1967. Syntax of Kin-
dergarten and Elementary School Children: A Transformational Analy-
sis. Champaign, Ill.: National Council of Teachers of English.
Reid, J. F. 1966. Learning to think about reading. Educational Research
9.56 -62.
Sachs, J. 1971. The status of developmental studies of language. In J.
Eliot (Ed.), Human Development and Cognitive Processes. New York:
Holt, Rinehart & Winston.
Schlesinger, I. M. Production of utterance and language acquisition. In
D. I. Slobin (Ed.), The Ontogenesis of Grammar: Facts and Theories.
New York: Academic P r e s s , in press.
Sinclair-de Zwart, H. 1968. Sensorimotor action schemes a s a condition
of the acquisition of syntax. Unpublished paper, University of Geneva.
Slobin, D. I. 1966a. The acquisition of Russian a s a native language. In
F. Smith and G. A. Miller (Eds.), The Genesis of Language: A Psy-
cholinguistic Approach. Cambridge, Mass.: M.I.T. Press.
Slobin, D. I. 1966b. Comments on “Developmental psycholinguistics. ” In
F. Smith and G. A. Miller (Eds.), The Genesis of Language: A Psycho-
linguistic Approach. Cambridge, Mass.: M.I.T. Press.
Slobin, D. I. 1968. Imitation and grammatical development. In N. S.
Endler, L. R. Boulter, and H. Osser (Eds.), Contemporary Issues in
Developmental Psychology. New York: Holt, Rinehart & Winston.
Slobin, D. I. 1970. Universals of grammatical development in children.
In G. B. Flores d’Arcais and W. J. M. Levelt (Eds.), Advances in
Psycholinguistics. Amsterdam-London: North Holland Publishing Com-
pany.
Slobin, D. I. 1971. Psycholinguistics. Glenview, Ill. : Scott, Foresman.
Slobin, D. I. Early grammatical development in several languages, with
special attention to Soviet research. In T. G. Bever and W. Weksel
(Eds.) The Structure and Psychology of Language. New York: Holt,
Rinehart & Winston, in press.
26 LANGUAGE LEARNING, VOL. 21, NO. 1

Slobin, D. I., and Welsh, C. A . 1967. Elicited imitation a s a research

tool in developmental psycholinguistics. Unpublished paper, Department
of Psychology, University of California, Berkeley.
Staats, A. W., and Staats, C. K. 1962. A comparison of the development
of speech and reading behavior with implications for research. Child
Development 33.831 -846.
Staats, A . W., and Staats, C. K. 1963. Complex Human Behavior: A sys-
tematic Extension of Learning Principles. New York: Hoft, Rinehart
& Winston.
Staats, A . W., and Staats, C. K. 1968. Language, Learning, and Cogni-
tion. New York: Holt, Rinehart, & Winston.
Strickland, R. G. 1962. The language of elementary school children: its
relationship to the language of reading textbooks and the quality of
reading of selected children. Bulletin of the School of Education 38.
Bloomington, Ind. : Indiana University.
Vygotsky, L. S. 1962. Thought and Language. Cambridge, Mass.: M.I.T.
Press.
Weir, R. 1962. Language in the Crib. The Hague: Mouton.
Weksel, W. 1965. Review of U. Bellugi and R. Brown (Eds.), The acquisi-
tion of language. Language 41.692-709.