SECOND LANGUAGE ACQUISITION:

THE PIDGINIZATION HYPOTHESIS

John H. Schumann
University of California a t Los Angeles

This paper is a case study of the untutored acquisition of

English b y a 33 year old Costa Rican named Alberto. His language
learning was examined longitudinally for a ten month period.
During that time he evidenced very little linguistic growth. Three
causes for Alberto’s lack of development are considered: ability,
age, and social a n d psychological distance. Performance o n a test
of adaptive intelligence indicated that lack of ability is n o t
adequate t o explain his acquisition pattern. Also, due to the
inadequacy of t h e arguments for a biological critical period in
language acquisition, age is also rejected as a cause. However,
Alberto’s English speech showed evidence of pidginization.
Pidginization is seen as the result of t h e learner’s social and
psychological distance from speakers of the target language. Hence,
it is argued that Alberto’s lack of development in English is the
result of his social and psychological distance from native speakers
of English.

In the fall of 1973 a research project (Cazden, Cancino,

Rosansky and Schumann 1975) was undertaken to make a
ten-month longitudinal study of the untutored acquisition of
English by six native speakers of Spanish-two children, two
adolescents and two adults. Data collection involved the recording
of both spontaneous and experimentally elicited speech. This
report is a case study of one of the six subjects, a 33 year-old
Costa Rican named Alberto, who evidenced very little linguistic
development during the course of the project. It was felt that by
attempting t o account for his lack of learning, significant insight
could be gained on what is involved in successful second language
acquisition in general.

Developmental patterns in the negative,

interrogative and auxiliary

The research focused on the subjects’ acquisition of negatives,

wh- questions, and auxiliaries. The analysis revealed several clear

39 I
392 LANGUAGE LEARNING VOL. 26, NO. 2

patterns of development. In the negative all subjects began with n o

+ verb (no V ) constructions in which the negative particle, while
internal t o the sentence, was external t o the verb: I no can see,
B u t n o is m i n e . . ., I no use television. Simultaneously or shortly
afterwards the subjects started using don’t + verb (don’t V )
constructions. Here don’t did not consist of d o + not, but was
simply an allomorph of no which was also kept external to the
verb: I don’t hear, He don’t like it, I don’t can explain. In the
third stage, auxiliary + negative (aux-neg), the subjects learned t o
place the negative particle after the auxiliary. In general, the first
auxiliaries t o be negated in this way were is (isn’t) and can (can’t).
In the final stage (analyzed don’t), the learners acquired the
analyzed forms of don’t (do not, doesn’t, does not, didn’t, did
not): I t doesn’t spin, Because y o u didn’t bring, He doesn’t laugh
like us. At this point don’t was n o longer negative chunk, but
actually consisted of d o plus the negative particle. The stages in
this sequence were not discrete and there was a good deal of
overlap among them. Each stage was defined by the negating
strategy that was used predominately at that time.
The analysis of the acquisition of wh- questions revealed a
developmental pattern which consisted of two stages (undifferentia-
tion and differentiation). The first stage involved three periods
(uninverted, variable inversion and generalization). This develop-
mental sequence is summarized below:
Stage I - Undifferentiation: Learner did not distinguish
between simple and embedded wh-
questions.
a. uninverted: Both simple and embedded wh- questions were
uninverted.
simple: What y o u study?;
embedded: That’s what I d o with m y pillow.
b. variable inversion : Simple wh- questions were sometimes
inverted, sometimes not.
inverted: H o w can y o u say it?;
uninverted: Where y o u get that?
c. generalization: increasing inversion in wh- questions with
inversion being extended t o embedded
questions.
simple: H o w can I kiss her if I don’t even k n o w her
name?;
embedded: I k n o w where are y o u going.
 THE PIDGINIZATION HYPOTHESIS 393

Stage I1 - Differentiation: Learner distinguished between

simple and embedded wh- ques-
tions.
simple: Where d o you live?;
embedded: I don’t k n o w what he had.
(from Cazden, Cancino, Rosansky and Schumann 1975:38).
In the analysis of the acquisition of auxiliaries we found that
is ( c o p ) was acquired first by all the subjects and that generally d o
and can followed shortly afterwards. The other auxiliaries appeared
in a highly variable order.

Alberto’s development

As mentioned above, one of the adult subjects, Alberto,

showed very little linguistic development during the course of the
study. Whereas four stages were found in the acquisition of the
English negative (no V, don’t V, aux-neg, analyzed don’t),
throughout the study Alberto remained in the first stage. Two
stages were found in the acquisition of English wh- questions
(undifferentiation and differentiation); throughout the study
Alberto remained in the first period of the first stage. In addition,
in yes/no-questions he inverted considerably less frequently than
the other subjects. The four inflectional morphemes (possessive,
past tense, plural and progressive) which were studied showed little
or n o growth over time. In terms of auxiliary development, am
(cop), can and are ( c o p ) could be classified as appearing in his
speech (i.e., they were supplied 80% of the time in three
consecutive samples), but only is ( c o p ) approaches the criterion for
acquisition (correctly supplied in 90% of obligatory contexts for
three successive samples). In general, then, Alberto can be
characterized as using a reduced and simplified form of English:
a. in which the negative particle remains external t o the verb
and is n o t placed after the first auxiliary element as
required in well-formed English;
b. in which inversion is virtually absent in questions;
c. in which no auxiliaries [except possibly is ( c o p ) ] can be
said t o be acquired, and using a less stringent criterion
only four auxiliaries [is (cop), a m (cop), can and are
( c o p ) ] can be said to have appeared;
d. in which the possessive tends t o be unmarked;
e. in which the regular past tense ending (ed) is virtually
absent;
394 LANGUAGE LEARNING VOL. 26, NO. 2

f. in which positive transfer from Spanish can account for the

plural inflection being supplied 85% of the time, for is
cop)'^ being correctly supplied to a greater extent than
other auxiliaries and for am (cop), are (cop) and can
reaching criterion for appearance;
g. and in which the progressive morpheme (-ing) is supplied
only about 60% of the time.

Reasons for Alberto’s development

Three explanations are considered in accounting for the lack

of development in Alberto’s speech: ability, age, and social and
psychological distance from speakers of the target language.
Ability. Performance on a Piagetian test of adaptive intelli-
gence (Feldman e t al. 1974) indicated that Albert0 had no gross
cognitive deficits that would have prevented him from acquiring
English more fully. Therefore, lack of ability does not seem
adequate t o explain his acquisition pattern.
Age. I t was once thought that the completion of cortical
lateralization a t puberty was the cause of adult difficulties in
acquiring second languages. However, Krashen (1973) has demon-
strated that the lateralization process which gradually locates
language functions in the left hemisphere of the brain may be
completed by the age of five. Therefore, since we know that six,
seven and eight year olds learn second languages without great
difficulty, we are left with no substantially clear age-related
biological or neurdlogical explanation for Alberto’s lack of
development in English.
Pidginization. Alberto’s essentially reduced and simplified
English contains several features that are characteristic of pidgin
languages. A pidgm language is a simplified and reduced form of
speech used for communication between people with different
languages. The type of pidginization referred to here is secondary
hybridization, not tertiary hybridization (see Whinnom 1971); the
position taken here is that secondary hybridization is legitimate
pidginization. The grammatical structure of pidgins is characterized
by a lack of inflectional morphology and a tendency to eliminate
grammatical transformations. Alberto’s English shared the following
features with other pidgin languages:
a. He used the uniform negative “no” for most of his
negative utterances as in American Indian Pidgin English
(AIPE) (Leachman and Hall 1955) and English Worker
Pidgin (EWP) (Clyne 1975).
 THE PIDGINIZATION HYPOTHESIS 395

b. He did n o t invert in questions as in Neo-Melanesian Pidgin

(N-MP) (Smith 1972) and EWP.
c. He lacked auxiliaries as in EWP.
d. He tended n o t t o inflect for the possessive as in AIPE.
e. He used the unmarked form of the verb as in English-
Japanese Pidgin (E-JP) (Goodman 1967), AIPE and EWP.
f. He deleted subject pronouns as in EWP.
Since Alberto’s English appears t o be pidginized, we want to
answer the question, “What causes pidginization?” The answer lies
in the functions which a pidgmized language serves. Smith (1972)
sees language as having three general functions: communicative,
integrative and expressive. The communicative function operates in
the transmission of referential, denotative information between
persons. The integrative function is engaged when a speaker
acquires language t o the extent that i t marks him as a member of a
particular social group. That is, his speech contains those features
(such as correct noun and verb inflections, inversion in questions,
and correct placement of the negative particle) that are un-
necessary for simple referential communication, but which are
necessary in order t o sound like a member of the group whose
language contains these features. The expressive function goes
beyond the integrative in that through it, the speaker becomes a
valued member of a particular linguistic group. In other words, he
displays linguistic virtuosity o r skill such that he becomes an
admired member of the community. Examples of such people are
storytellers (especially in non-literate societies), comedians, orators,
poets, etc. Since many native speakers d o not command the
expressive functions of their language, in order t o be considered a
fluent speaker of a language, one need only master the
communicative and integrative functions. According t o Smith,
pidgin languages are generally restricted to the first function-
communication. That is, their purpose is merely t o convey
denotative, referential informa tion. Since pidgins are always second
languages, the integrative and expressive functions are maintained
by the speakers’ native languages. As a result of this functional
restriction, pidginization produced an interlanguage which is
simplified and reduced.
The next question t o be answered, then, is “What causes
restriction in function?” Martin Joos (1971:187) suggests that
“the skeletonizing/skeletonized pattern of pidgin-formation . . .
emerges automatically from lack of actual/prospective social
solidarity between speaker and addressee” (emphasis mine). To this
I would also add the lack of actual or prospective psychological
 LANGUAGE LEARNING VOL. 26, NO. 2

solidarity between the two parties. If we turn this formulation

around, restriction in function can be seen as resulting from social
and/or psychological distance between the speaker and addressee.
Placing this notion within the framework of second language
acquisition, we would argue that the speech of the second language
learner will be restricted t o the communicative function if the
learner is socially and/or psychologcally distant from the speakers
of the target language. The extent and persistence of the pidgmized
forms in the second language learner’s speech will result auto-
matically then from the restriction in function.
Social distance pertains t o the individual as a member of a
social group which is in contact with another social group whose
members speak a different language. Hence social distance involves
such sociological factors as domination versus subordination,
assimilation versus acculturation versus preservation, enclosure, size,
congruence and attitude. Psychological distance pertains to the
individual as an individual, and involves such psychological factors
as resolution of language shock, culture shock and culture stress,
integrative versus instrumental motivation and ego-permeability. In
the following two sections each form of distance will be discussed.
Social distance. The following notions about social distance
(Schumann 1976) evolve from the literature on bilingualism,
second language acquisition, sociolinguistics and ethnic relations.
They represent societal factors that either promote or inhibit social
solidarity between two groups and thus affect the way a second
language learning group (2LL group) acquires the language of a
particular target language group (TL group). The assumption is that
the greater the social distance between the two groups the more
difficult it is for the members of the 2LL group t o acquire the
language of the TL group. The following issues are involved in
social distance: In relation t o the TL group is the 2LL group
politically, culturally, technically or economically dominant, non-
dominant, or subordinate? Is the integration pattern of the 2LL
group assimilation, acculturation, or preservation? What is the 2LL
group’s degree of enclosure? Is the 2LL group cohesive? What is
the size of the 2LL group? Are the cultures of the two groups
congruent? What are the attitudes of the two groups toward each
other? What is the 2LL group’s intended length of residence in the
target language area? The above terms are defined as follows:
1. Dominant-2LL group is politically, culturally, technically
or economically superior t o the TL group.
2. Non-dominant-2LL group is politically, culturally,
technically and economically equal t o the TL group.
 THE PIDGINIZATION HYPOTHESIS 397

3. Subordinate-2LL group is politically, culturally, techni-

cally and economically inferior t o the TL group.
4. Assimilation-2LL group gives u p its own life style and
values and adopts those of the TL group.
5. Acculturation-2LL group adapts to the life style and
values of the TL group, but a t the same time maintains its
own cultural patterns for use in intra-group relations.
6. Preservation-2LL group rejects the life style and values
of the TL group and attempts t o maintain its own
cultural pattern as much as possible.
7 . Enclosure-The degree t o which the two groups have
separate schools, churches, clubs, recreational facilities,
professions, crafts, trades, etc.
8. Cohesiveness-The degree to which members of the 2LL
group live, work and socialize together.
9. Size-How large the 2LL group is.
10. Congruence-The degree t o which the cultures of the two
groups are similar.
11. Attitude-Ethnic stereotypes by which the two groups
either positively or negatively value each other.
12. Intended length of residence-How long the 2LL group
intends to remain in the TL area.
I t is argued that social distance and hence a bad language
learning situation (see rows A and B in Tables 1 and 2) will obtain
where the 2LL group is either dominant or subordinate, where
both groups desire preservation and high enclosure for the 2LL
group, where the 2LL group is both cohesive and large, where the
two cultures are not congruent, where the two groups hold
negative attitudes toward each other and where the 2LL group
intends to remain in the target language area only for a short time.
I t is also argued that social solidarity and hence a good language
learning situation (see row C in tables) will obtain where the 2LL
group is non-dominant in relation t o the TL group, where both
groups desire assimilation for the 2LL group, where low enclosure
is the goal of both groups, where the two cultures are congruent,
where the 2LL group is small and non-cohesive, where both groups
have positive attitudes toward each other, and where the 2LL
group intends t o remain in the target language area for a long time.
In comparing Alberto’s social distance from Americans with
that of the other subjects in Cazden, Cancino, Rosansky and
Schumann’s (1975) study, Alberto can be regarded as belonging t o
a social group designated as lower class Latin American worker
immigrants, and the other four subjects can be classified as
 TABLE 1
Analysis o f political, economic, technical, cultural, and structural characteristics for good and bad langua e learning
situations and f o r worker US. professional immigrants from Latin America to the United States 6;
Political, economic,
technical, cultural Culture Structure

TLgp views 2LLgp views TLgp desires 2LLgp desires TLgp desires 2LLgp desires
2LUP itself of 2LLgp for itself for 2LLgp for itself

bM
Y
m al c e
42 U
Y U .-0 .*0
U U Y
al
c
.& c
9 9c e e k
.-e 3 3 P 0 c ‘CI
P P .
BI
’Z0 2 .-M
P I P I a c c
2 2 Ea 2
A. Bad language learning
Situation I * * * * * *
B. Bad language learning
Situation I1 J J J J J J
C. Good language learning
situation X X X X X X
D. Latin American c
workers J J J J J J 0
F
E. Latin American
professionals X X X X X X

‘Corresponding J ’ s and X’s indicate.similarity of situations. Z

0
 TABLE 2
Analysis of second language group characteristics, attitudes and social distance f o r good and bad langua e learning
situations and for worker us. professional immigrants f r o m Latin America t o the United States4
Social
2 LLgp Characteristics Attitudes Distance

TLgp’s 2LLgp’s
Length attitude attitude
Cohesiveness Size Culture of Stay toward 2LLgp toward TLgp
-

A. Bad language learning

Situation I * * * * * * *
B. Bad language learning
Situation 11 J J J J J J J
C. Good language learning
situation X x x X X X X
D. Latin American
workers J J JL J
E. Latin American
professionals X x x X X
lcorresponding J ’ s and X’s indicate similarity of situations.
400 LANGUAGE LEARNING VOL. 26, NO. 2

children of upper-middle class Latin American professional immi-

grants. There was insufficient background on the second adult
subject t o include her in this classification.
Latin American worker immigrants (see row D) are sub-
ordinate in relation t o Americans since they represent an unskilled
labor group whose modal socio-economic status is lower than that
of Americans in general. This view is probably shared by both the
worker immigrants and the Americans. The worker immigrants
probably fall somewhere between preservation and acculturation
with regard t o their desired integration into American society.
American society in general expects them t o assimilate as i t does
all immigrants, but it does not necessarily make the assimilation
easy. In terms of enclosure the Latin American workers have access
t o American institutions, but generally live in immigrant neighbor-
hoods where they share schools, churches and associations with
other immigrants having the same socio-economic status and
usually having the same language and culture. This enclosure by
neighborhood fosters cohesiveness, particularly in Alberto’s case
where Costa Rican immigrants are a small minority within a
Portuguese minority area. The culture of the Latin American
worker immigrants is relatively congruent t o that of the Americans
(both being Western and Christian), but since the Latin American
workers may represent the “culture of poverty” more than does
the modal American culture, there may also be an element of
incongruence between the two cultures (indicated by the arrow, 4,
in Table 2). The attitudes of the two groups toward each other
would have t o be measured before accurate judgments could be
made. I t is also difficult t o assess the intended length of stay in
the United States by Latin Americar? workers.
Upper-middle ciass Latin American professional immigrants
(see row E in the tables) are probably viewed by Americans and
also view themselves as non-dominant in relation t o the English-
speaking TL group because their educational background and
socio-economic status more closely match that of Americans in
general (particularly in the Boston/Cambridge area). The Latin
American professionals are solidly acculturative in their integration
pattern. They have t o be able t o demonstrate culturally appropri-
ate behavior in their relationships with American colleagues and
therefore must adapt to American life styles and values. But since
their length of residence in the United States is often confined t o a
period of postgraduate education, they generally do not choose t o
assimilate. The professionals are generally integrated into the
university and professional communities and do not live in
 THE PIDGINIZATION HYPOTHESIS 401

immigrant neighborhoods. Therefore, their enclosure is low and

they are less cohesive than the worker immigrants. The size of the
professional group is likely to be smaller than that of workers, and
the congruity of the two cultures is relatively high. Once again
attitudinal orientations would have t o be empirically assessed in
order t o be correctly classified.
When both profiles are considered we find that the Latin
American worker immigrant group is a t a considerably greater
social distance from Americans than are the professionals. Thus, we
would expect the workers’ use of English t o be functionally
restricted and to pidginize. This is precisely what we find in
Alberto.
Psychological distance. As the classification of the 2LL group
in either the good or bad language learning situations becomes less
determinant (that is, if a group stands somewhere between the bad
and good situations), then success in acquiring the target language
becomes more a matter of the individual as an individual rather
than of the individual as a member of a particular social group. In
addition, in either a good or a bad language learning situation, an
individual can violate the modal tendency of his group. Thus, an
individual might learn the target language where he is expected not
to, and n o t learn the language where successful acquisition is
expected. In these cases it is psychological distance (Schumann
1975b) or proximity between the learner and the TL group that
accounts for successful versus unsuccessful second language
acquisition. The factors which create psychological distance
between the learner and speakers of the target language are
affective in nature and involve such issues as the resolution of
language shock and culture shock, motivation and ego perme-
ability.
In experiencing language shock (Stengal 1939), the learner is
haunted by doubts as t o whether his words accurately reflect his
ideas. In addition, he is sometimes confronted with target language
words and expressions which carry with them images and meanings
which he interprets differently than do native speakers of the
target language. Also, the narcissistic gratification t o which the
learner is accustomed in the use of his native language is lost when
he attempts t o speak the target language. Finally, when speaking
the second language the learner has apprehensions about appearing
comic, child-like and dependent.
The learner experiences culture shock (Smalley 1963, Larsen
and Smalley 1972) when he finds that his problem-solving and
coping mechanisms do n o t work in the new culture. When they are
402 LANGUAGE LEARNING VOL. 26, NO. 2

used they d o not get the accustomed results. Consequently,

activities which were routine in his native country require great
energy in the new culture. This situation causes disorientation,
stress, fear and anxiety. The resultant mental state can produce a
whole syndrome of rejection which diverts attention and energy
from second language learning. The learner, in attempting to find a
cause for his disorientation, may reject himself, the people of the
host country, the organization for which he is working, and even
his own culture.
Motivation (Gardner and Lambert 1972) relates to the goals
of second language learning. In terms of psychological distance, the
integratively motivated learner would seek maximum proximity in
order t o meet, talk with, and perhaps even become like the
speakers of the target language. An instrumentally motivated
learner would achieve a level of psychological solidarity that would
only be commensurate with his instrumental goals. Consequently, if
the learner’s goal were mere survival, he might maintain a good
deal of psychological distance between himself and the speakers of
the target language.
Another source of psychological distance may be the relative
rigidity of the learner’s ego boundaries (Guiora 1972). Some
experimental evidence indicates that people who have ego
permeability, that is, the ability to partially and temporarily give
up their separateness of identity, are better second language
learners. This essentially psychoanalytic concept is intuitively
appealing and provides another perspective from which the concept
of psychological distance can be understood.
In sum, then, factors causing psychological distance, like those
causing social distance, put the learner in a situation where he is
largely cut off from target language input and/or does not attend
to it when it is available. The language which is acquired under
these conditions will be used simply for denotative referential
communication in situations where contact with speakers of the
target language is either absolutely necessary or unavoidable. The
learner’s psychological distance will prevent him from identifying
with the speakers of the target language such that he will not
attempt to incorporate into his speech those linguistic features that
would help to identify him as a member of the TL group. Hence,
his use of the target language will be functionally restricted and,
therefore, we would expect i t to pidginize.
In order t o get some assessment of Alberto’s psychological
distance from English speakers, a t the end of the study, he was
asked to fill out a short questionnaire which elicited information
 THE PIDGINIZATION HYPOTHESIS 403

concerning his attitude and motivation. In terms of this question-

naire, he seemed t o have a positive attitude and good motivation,
and hence little psychological distance. However, there is some
question as to whether h e was entirely candid in his answers.
Alberto tended not to like t o displease and therefore his answers
may reflect what he thought the experimenter wanted t o hear.
There are several aspects of Alberto’s life-style that appear t o
contradict the positive attitude and motivation expressed in the
questionnaire. First of all, he made very little effort t o get to
know English-speaking people. In Cambridge he stuck quite close
t o a small group of Spanish-speaking friends. He did not own a
television and expressed disinterest in it because he could n o t
understand English. On the other hand, he purchased an expensive
stereo set and tape deck on which he played mostly Spanish music.
Also, he chose t o work a t night (as well as in the day) rather than
attend English classes which were available in Cambridge.
The other subjects were n o t given the attitude and motivation
questionnaire, but in general they seemed t o be psychologically
much closer to Americans. All the children attended American
schools and had American friends. The second adult baby-sat for
American children, studied English on her own and tried to get t o
know and speak with Americans.
The effect of instruction. From the point of view of the
pidginization hypothesis we would argue that Alberto did not seek
o u t instruction in English because his pidginized speech was
adequate for his needs. Nevertheless, it might be argued that with
instruction his simplified linguistic system might have reorganized
and come t o conform more closely with the target language. The
opportunity t o test this idea presented itself after the study was
completed. At the end of the ten-month project, twenty, one-hour
speech samples had been collected. As mentioned earlier, through-
o u t this period Alberto had maintained essentially a no V negation
system.
The experimenter then undertook to teach him how t o negate
in English to see if this intervention would cause him t o alter his
pidginized system of negation. Extensive instruction was provided
during the collection of speech sample 21 and then intermittently
in samples 22 through 32. This program covered a seven-month
period. At the same time in samples 22-30 Alberto was given
extensive sets of positive sentences which he was asked to negate.
These elicited negatives were then compared with the negative
utterances in his spontaneous speech. In elicited speech after
instruction, Alberto’s negatives were about 64% (216/335) correct.
404 LANGUAGE LEARNING VOL. 26, NO. 2

His spontaneous negatives, however, were only about 20% (56/278)

correct. His don’t V utterances were correct only by coincidence
simply because don’t, as an allomorph of no, was occasionally used
in the appropriate linguistic environment. Therefore, we see that
instruction influenced only Alberto’s production in a test-like,
highly monitored situation; it did not affect his spontaneous
speech which h e used for normal communication. This result is
even more striking when we compare i t with spontaneous and
elicited negatives prior to instruction. In samples 16-20, Alberto’s
spontaneous negatives were 22% (23/105) correct and his elicited
negatives were 10% (7/71) correct. This indicates that instruction
has radically improved his performance in an artifical, highly
monitored elicitation task, but that i t had virtually n o effect on his
spontaneous speech which he uses in normal communication with
native speakers of English. Hence we can conclude that instruction
is evidently not powerful enough t o overcome the pidginization
engendered by social and psychological distance.

Cognitive processes in pidginization

The social and psychological forces that cause the persistence

of pidginization in a second language learner’s speech have been
discussed. The term “persistence” is used because, as predicted in
Schumann (1974 a and b), pidginization appears to be character-
istic of early second language acquisition in general. What has been
described as pidgnization in Alberto’s speech corresponds to the
early stages of the acquisition of English by all six learners.
Alberto remained in stage one of negation (the no V stage) and in
stage one, period a of interrogation (universion in both simple and
embedded wh- questions). Since it is reasonable to assume that, as
with Alberto, inflectional marking tended to be absent in the early
speech of the other five subjects (this was not specifically
examined in Cazden, Cancino, Rosansky and Schumann 1975),
evidence exists that pidginization may characterize all early second
language acquisition and that under conditions of social and
psychological distance i t persists. Since pidginization may be a
universal first stage in second language acquisition, it is important
t o explore what cognitive processes either cause or allow the
pidginization t o occur.
Kay and Sankoff (1974) believe that contact vernaculars such
as pidgins and other varieties of incomplete competence such as
child language, second language acquisition, bilingualism and
aphasia are all potential areas for examining linguistic universals.
 THE PIDGINIZATION HYPOTHESIS 405

Referring to contact vernaculars in particular they state that “since

the communicative functions fulfilled by contact vernaculars are
minimal, these languages may possibly reveal in a more direct way
than d o most natural languages the universal cognitive structure
and process that underlie all human language ability and use”
(Kay and Sankoff 1974:62).
Smith (1973:3) notes that the early speech of children is
largely unmarked (hence the term telegraphic speech) and that in
the process of socialization the child learns to mark his language
with those features which characterize his speech community. The
result of this development is that adult speech is naturally and
normally marked. However, pidgin languages which are spoken by
adults are characteristically unmarked. Smith attempts to account
for the fact that pidginization produces a generally unmarked
language by viewing unmarking and marking as part of the same
process. The child a t one point in his development has had the
ability to unmark. Smith speculates that this ability is not lost and
can be retrieved under certain social conditions. One of these
conditions is the pidginogenic social context where the function of
the language is restricted t o communication of denotative
referential information. Both the child in early native language
acquisition and the pidgin speaker reduce and simplify the language
t o which they are exposed into a set of primitive categories which
undoubtedly are innate (Smith 1 9 7 3 : l l ) . These primitive catego-
ries emerge in speech as utterances relatively unmarked by
inflections, permutations and functors. Within this framework
unmarking is not seen as a deficiency, but as a positive cognitive
strategy t o which a language learner turns a t certain development
stages and under certain social conditions.
Corder (1975) maintains a similar position, but argues that
‘simple codes’ spoken by children, neophyte second language
learners, pidgin speakers, and adults using baby-talk or foreigner
talk are not ‘simplified,’ i.e., they are not reductions of a more
complicated and expanded code. Instead they represent a basic
language which, in the process of learning, is expanded and
complicated. Following Kay and Sankoff (1972), Corder (1975:4)
suggests that simple codes “are ‘nearer’, in some sense, t o the
underlying structure or ‘inner form’ of all languages, i.e., more
overtly reflect semantic categories and relations.” He goes on t o
speculate that this basic language, and all intermediate linguistic
systems between basic and complex, once learned are never
obliterated. These approximative systems remain “available both
for special communicative functions in the mother tongue [baby
406 LANGUAGE LEARNING VOL. 26, NO. 2

talk, foreigner talk] and as an ‘initial hypothesis’ in the learning of

second language’’ (p. 9).
Within this framework, pidginization in second language
acquisition can be viewed as initially resulting from cognitive
constraints and then persisting due to social and psychological
constraints. Hence, early second language acquisition would be
characterized by the temporary use of a non-marked, simple code
resembling a pidgin. This code would be the product of cognitive
constraints engendered by lack of knowledge of the target
language. The code may reflect a regression to a set of universal
primitive linguistic categories that were realized in early first
language acquisition. Then, under conditions of social and/or
psychological distance, this pidginized form of speech would
persist.

Conclusion

The pidginization hypothesis predicts that where social and

psychological distance prevail we will find pidginization persisting
in the speech of second language learners. There are several
experimental and several clinical studies that could be undertaken
t o further explore this hypothesis. In order to experimentally test
the social distance aspect of the hypothesis, one might choose a
population of worker immigrants in the United States and compare
its success in the acquisition of English to the success in the
acquisition of English experienced by a group of professional
immigrants. T o experimentally test the psychologcal distance
aspect of the hypothesis one could make an intensive examination
(using questionnaires, interviews, etc.) of those worker immigrants
who d o successfully learn English and the professional immigrants
who fail t o learn it.
T o clinically examine social distance phenomena, a question-
naire might be developed which would be filled out by
experimenters doing research in second language acquisition. In it
they would attempt to classify the subjects with whom they were
working (either groups or individuals) on social distance di-
mensions. The questionnaire would be designed to permit the
researcher to rate a particular 2LL group’s dominance, cohesive-
ness, enclosure, etc., on a numerical scale, to compute a social
distance score for the group and then t o relate that score t o the
extent of pidginization found in his subject(s)’ speech.
Psychological distance might receive clinical examination by
studying a small group of subjects (six t o ten) who will be living in
 THE PIDGINIZATION HYPOTHESIS 407

a foreign language environment for a fairly long period of time.

The subjects might be a group of Peace Corps Volunteers or
foreign service personnel who have a good opportunity t o become
bilingual as a result of training in and exposure t o the target
language. At the beginning of the study the subjects would be
assessed on as many relevant variables as possible, including:
language learning aptitude, attitude, motivation, ego-permeability
(assuming a valid measure is available), experiences in learning
other second languages and general social adjustment. The subjects
would be asked t o keep diaries in which they would describe daily
exposure to the target language, efforts t o learn the language, and
feelings about language learning and the new culture. In addition, ‘

the subjects would be interviewed once every two weeks in order

that the researchers could probe the same issues verbally. Finally
the subjects’ achievement in the second language would be tested
monthly by means of an oral interview which could then be
analyzed for aspects of pidginization. The object of this approach
would be t o develop several case studies in which an individual’s
pattern of second language acquisition could be related longitudi-
nally t o factors involving his psychological distance from speakers
of the target language.
Such research strategies could shed light on the interaction
between the phenomena of social and psychological distance;
uncover new factors contributing t o both phenomena and perhaps
indicate ways in which social and psychological distance can be
overcome and thus free those affected to become bilingual.
Finally, by studying the second language speech of learners
affected by social and/or psychological distance in a variety of
contact situations (for example, Chinese-English, English-Persian,
Italian-French) a further contribution could be made t o our
knowledge of the linguistic aspects of pidginization and the
processes of simplification and reduction in natural languages in
general.

REFERENCES

Cazden, Courtney B., Herlinda Cancino, Ellen J. Rosansky, and John H.

Schumann. 1975. Second language acquisition sequences in children,
adolescents and adults. Final Report, National Institute of Education
(Grant No. NE-6-00-3-0014).
Clyne, Michael G . 1 9 7 5 . German and English working pidgins. Paper presented
at the International Congress o n Pidgins and Creoles, Honolulu, Hawaii.
Corder, S. P. 1 97 5 . ‘Simple codes’ and the source of the second language
learner’s initial heuristic hypothesis. Paper presented at the Colloque
‘Theoretical Models in Applied Linguistics’ IV, University de Neuchatel.
408 LANGUAGE LEARNING VOL. 26, NO. 2

De Camp, David and Ian F. Hancock (eds.). 1974. Pidgins and Creoles:
Current Trends and Prospects. Washington, D.C.: Georgetown University
Press.
Feldman, Carol, B. Lee, J. D. McLean, D. Pellemer, and J. Murray. 1974. The
Development of Adaptive Intelligence. Chicago: Jossey-Bass Publishers.
Gardner, Richard C. and Wallace E. Lambert. 1972. Attitudes and Motivation
in Second-Language Learning. Rowley, Mass.: Newbury House.
Goodman, J. S. 1967. The development of a dialect of English-Japanese
pidgin. Anthropological Linguistics 9.4 3-55.
Guiora, Alexander Z. 1972. Construct validity and transpositional research:
toward an empirical study of psychoanalytic
_ . concepts. Comprehensive
Psychiatry 13.i 39-150.
Hvmes. Dell H. (ed.). 1971. Pidginization and Creolization of Languages.
" Cambridge: Cambridge University Press.
Joos, Martin. 1971. Hypotheses as to the origin and modification of pidgins.
In Dell H. Hymes (ed.), Pidginization and Creolization of Languages, 187.
Cambridge: Cambridge University Press.
Kay, Paul and Gillian Sankoff. 1974. A language-universal approach to pidgins
and Creoles. In David De Camp and Ian F. Hancock (eds.), Pidgins and
Creoles: Current Trends and Prospects, 61-72. Washington D.C.: George-
town University Press.
Krashen, Steven D. 1973. Lateralization, language learning, and the critical
period: some new evidence. Language Learning 23.63-74.
Larsen, Donald A. and William A. Smalley. 1972. Becoming Bilingual. A
Guide to Language Learning. New Canaan, Ct.: Practical Anthropology.
Leachman, D. and Robert A. Hall, Jr. 1955. American Indian Pidgin English:
attestations and grammatical peculiarities. American Speech 30.163-71.
Schumann, John H. 1974. The implications of interlanguage, pidginization and
creolization for the study of adult language acquisition. TESOL
Quarterly 8.145-52.
Schumann, John H. 1975a. The implications of pidginization and creolization
for the study of adult second language acquisition. In Schumann and
Stenson (eds.), New Frontiers in Second Language Learning, 137-152.
Rowley, Mass: Newbury House.
Schumann, John H. 1975b. Second language acquisition: the pidginization
hypothesis. Unpublished doctoral dissertation, Harvard University.
Schumann, John H. 1975c. Affective factors and the problem of age in second
language acquisition. Language Learning 25.209-235.
Schumann, John H. 1976. Social distance as a factor in second language
acquisition. Language Learning 26.135-1 43.
Schumann, John H. and Nancy Stenson (eds.). 1975. New Frontiers in Second
Language Learning. Rowley, Mass.: Newbury House.
Smalley, William A. 1963. Culture shock, laiigclage shock and the shock of
self-discovery. Practical Anthropology 10.49-56.
Smith, David M. 1972. Some implications for the social status of pidgin
languages. In David M. Smith and Roger Shuy (eds.), Sociolinguistics in
Cross-cultural Analyses, 47-56. Washington D.C. : Georgetown University
Press.
Smith, David M. 1973. Pidginization and language socialization: the role of
marking. Unpublished paper, Georgetown University.
Smith, David M., and Roger W. Shuy (eds.). 1972. Sociolinguistics in Cross-
cultural Analyses. Washington, D.C.: Georgetown University Press.
Stengal, E. 1939. On learning a new language. International Journal of
Psychoanalysis 2.471-79.
Whinnom, Keith. 1971. Linguistic hybridization and the 'special case' of
pidgins and Creoles. In Dell H. Hymes (ed.), Pidginization and Creolization
of Languages, 91-115. Cambridge: Cambridge University Press.