"Every language has the CAPACITY to take the form that is users require". (Bolinger, 1980: p.ix)
1.0. This paper intends to relate processes in language acquisition of a first language with some of the linguistic processes that take place in pidginization, creolization and decreolization. It is argued that the process of simplification in the initial stages of first language acquisition is similar to the process of simplification in pidginization; the phenomenon of innovation is common to the later stages of first language acquisition and creolization; and the final stages of first language acquisition are similar to decreolization as in both the cases the goal is identical, namely, to arrive at the proficiency of adult language in case of first language acquisition and target language in case of decreolization. It is finally pointed out that the chances of survival of innovations in child language are very low whereas innovations in creoles may survive.
1.1. First, I shall discuss simplification, innovation, transfer of features from one language to the other and admixture in child language on the basis of data of Varnita who is four years old now. The data was collected from the age of 17 months onwards. The child was exposed to Telugu, Hindi and English at home and Hindi and Punjabi outside.
The child's exposure to Hindi deserves a special mention. The Hindi that she was exposed to at home was standard Hindi from two speakers and some kind of pidgninized Hindi from two speakers because these two speakers do not have a good command of Hindi. Some examples are given in the appendix for the type of pidginized Hindi that the child is exposed to.
1.1.1. Simplification : I shall now present some cases in which deletion of certain constituents in the child language takes place. Such a deletion appears to be directed towards simplification.
i) Deletion of the postposition ke liye `for'
1) baby bottle cry kar diyā
baby bottle cry did
`Baby cried for the bottle'.
ii) Deletion of the postposition ke in ke pas `in the possession of; near'
2) bhaiyā pās Campa hai
Salesman possession is
`The salesman has Campa Cola.
3) mitā pās ja rahī hũ
Meeta near am going.
`I am going to Meeta'.
iii) Deletion of verb be as tense carrier. The following sentences are
4) daddy coming
`Daddy is coming'
5) I going
`I am going'
Concerning deletion of be as copula,
(1971) makes some observations which are worth mentioning here. He proposed that "In languages of type a(that is with a copula-KVS), the Copula in equational clauses will tend (emphasis supplied to the omitted in simplification speech such as baby talk and foreign talk" (Ferguson, 1971: p.146). Ferguson further points out that such a hypothesis would suggest a pidgin language is likely to omit the Copula if the source language has one. Since am, were in English are also forms of the verb be, Ferguson 's argument can be extended to those cases in which be is a tense marker, and it can be proposed that deletion of the tense marker be is also a case of simplification. Ferguson
iv) Deletion of the dative
The postposition ko in the dative subject construction with predicates such as cāhiye `need' in Hindi is also deleted in child language as in (6).
6) tum lagāna cāhiye
you to put must
`You must put it'
It should be pointed out that the deletion of the postposition could be either an instance of simplification or it could be due to the pidgin Hindi that she is exposed to at home which does not have dative in such constructions.
The dative postposition with ditransitive verbs is also deleted as in sentences (7).
7) pi dedo tayyā
apple give Talli
`Give the apple to Talli'
v) Deletion of the instrumental se `by'
8) Mommy office scooter gayī
`Mommy went to the office by scooter'
vi) Absence of the genitive ka `of'
The child when she was around two, did not use the genitive postposition.
9) akkā sūz
10) ūpar guriyā dedo
up doll give
`give the doll up there'
The absence of the genitive in (9) can be interpreted in two ways: this can be an instance of simplification or it could be due to the influence of Telugu. In Telugu, the genitive marker is not overtly present in such cases.
The absence of postpositions in general can be interpreted in two ways. It can be hypothesized that child has `simplified' the grammar and therefore postpositions are not overtly present. Another view could be that the child has not acquired the "proper ability" to use postpositions. Either way, the sentences about form part of the `inter-language'.
i) Reduplication is a common feature of all the Indian languages. To express that an action takes place iteratively, the child used a reduplication form as in (11).
day too day too day too all day
sandwich lātī hai mere ko bhī din ko
brings to me too day
din ko sandwich de do
`she brings sandwich everyday all these days. Give sandwich to me too'.
The child did not know the expression har din or rōz `every-day'. In Hindi and Telugu a reduplicative form of the noun is used to express total set relation as opposed to sub-set relation or a reduplicative form can express iteration.
12) ghar me - `in the house'
13) ghar ghar me - `in every house'
14) ek bār - `once'
15) bār bār - again and again (repeatedly)
The child must have internalized the semantic principle behind reduplication and innovated the sentence as in (11).
ii) When the child was two years old, she had taken the nouns buwwa `food' from Telugu and manna (pānī of Hindi) - `water and added karnā `for do' to make the verb `eat' and `drink', i.e., buwwa karnā `to eat' manna karnā `to drink'.
16) ide buwwa kar diyā
he food did
`He ate food'
17) ide manna kar diyā
He water did
`He drank water'
Making a new verb out of a noun plus an action verb such as karnā is an innovation of the child.
iii) In Telugu pada means `move' (imperative). To express that she wants to go out for a walk, the child camp up with the expression in (18) which is a combination of the Telugu imperative and Hindi imperative of denā to give'.
18) pada dedo
Literally `give me a walk' i.d. `take (me) for a walk'.
It appears that the verb pada was conceived as a nominal by the child.
iv) The child generated sentences with productive causatives with the intransitive verb from Telugu and the casual do verb from Hindi as in (19).
19) pappol kar diyā
fell down did
style="tab-stops:-14.2pt 0in 7.1pt 21.3pt 42.55pt"> (Tel.) (Hindi)
style="tab-stops:-14.2pt 0in 7.1pt 21.3pt 42.55pt"> `I dropped something'
Notice that pappol `fell down' has the form padi poyīndi `fell down' in adult language in Telugu. Notice that Hindi does have a productive causative construction and Telugu like other Dravidian languages, has one whose range of usage is limited. The child had until then not been exposed to productive causative constructions. Such a construction reflects the INNOVATIVE and CREATIVE ABILITY of the child to generate structure, SENTENCE TYPES to be more specific, which she had never heard before and which are entirely NOVEL.
Some further examples in which karnā isused as a causal verb are worth mentioning.
20) get down karo
`Get me down'
21) kursī ko stand karo
`chair (acc) do
`Make the chair stand'
22) guriyā ko sōnā karnā hai
doll (acc) to sleep has to do.
`The doll has to perform the act of sleeping'
i.e., `the doll has to sleep'.
23) mai gill karūgā
I pinch will do
(Hindi) (Tel.) (Hindi)
`I' ll pinch you'.
24) kaun song kartī hai
who song is doing
`who is singing'
The verb karnā of Hindi in child language appears to be very productive. It is used as in (25) below in place of the verb kahnā `to say'.
25) khalo, bagundi kardo
eat, is good do (imp.)
(Hindi) (Tel.) (Hindi)
`Eat it and say it is good'.
I shall discuss in the last section in detail the significance of these innovations in child language.
It is very interesting that the child came up with structure which have either Telugu or English lexical items but have Hindi morphological endings and sometimes even Hindi compound verbs.
26) food, bananas eat liyā
`Mommy and daddy ate food and bananas
27) mai tipp + ū
I turn optative
(Hindi) (Tel.) (Hindi)
`Should I turn on the TV?
28) so jāo kallu mūsū nahī
sleep eyes close not
(Hindi) (Tel.) (Tel.) (Hindi)
`Lie down, do not close your eyes'.
29) mai ceyyi kadukkor-kar gay
I hands wash-conjunctive went
`I went to wash my hands'
(Hindi) (Tel.) (Tel.) (Hindi)
30) baby bottle cry kar diyā
`The baby cried for the bottle'
31) anni buk ūpar le liyā
all book upstairs took
(Tel.) (Hindi) (Hindi)
`We took all the book upstaris'
32) talli juttu Lūkas lāg raha hai
Talli hair Lucas pull progr. is
(Tel.) (Tel.) (Hindi)
`Lucus (a dog) is pulling Talli's hair'
33) jākī ide tokki yā
jackey this stepped on past
(Tel.) (Tel.) (Hindi)
`Jackey (dog) stepped on it'
From the above data, it can be clearly seen that:
Hindi seems to play a dominant role in acquisition.
In some instances, the child used English plural markers with lexical items from Hindi or Telugu.
34) kanna-s - rabbits
35) kelā-s - bananas
Some times the Punjabi plural marker ā is used as in (35)
36) kitābā - books
1.1.4. Syntactic Extension
1. The range of a syntactic construction is `extended' in child language. These extensions would appear to be a violation of adult language rules. I shall now discuss some cases.
184.108.40.206 The kar construction in Hindi normally requires identical subjects except in case of time adverbials. In the child's language sentences of the following type were observed.
37) soup roz banā ke cāhiye
soup everyday having made needed
`I need the soup made everyday'
38) mujhe dūdh ubāl kar cāhiye
to me milk having boiled is needed
`I need the milk boiled'
Sentence (38) is a very marginal sentence in Hindi. This sentence can be an answer to a question such as
39) ap ko dūdh kaisā cāhiye - thanda
to you milk how is needed - cold
`How'd you like to have your milk - cold or hot?
220.127.116.11. The vālā construction in Hindi also is extended. Vālā in the child's language was used in palace of the hua of the perfect participle. Let us consider the following examples.
40) vah yahā se niklā vālā hōtá hai
it here from the one that has come out is
`The band of the frock normally hangs from here (the waist)'.
42) merā pet niklā vālā ho
my stomach the one that has come out became
`(after eating) my stomach has popped out'.
Such sentences with vālā are perfectly grammatical if the vālā is replaced by huā in Hindi. In the child's language the vālā seems to have been used as an agentive modifier whereas in adult language vālā is a modifier as in (42)
42) cār baje vāli gāRi
18.104.22.168. The child has used a vālā expression as in (43) for specific whereas the vālā construction in adult language however denotes + generic.
43) āj nīce vālō se khelī thī
today kinds downstairs with played
`I played with the kinds who were downstairs'
When the child spoke sentence (43) she was standing on the second floor. What she meant was that she played with the girls who were at that particular moment downstairs but who actually live in the top floor apartment. In adult Hindi sentence (43) would be interpreted as the child played with a person who lives downstairs. This extension of generic to specific could be attributed to an extension of the specific usage as in (44) below
44) nīce vālī kitāb do
bottom one book give
`Give the book at the bottom'.
22.214.171.124. Another extension with vālā is worth mentioning. In Hindi vālā and the genitive kā alternatively are used after a nominal as in (45) and (46).
45) pūjā ki doll
pūjā genetive doll
46) pūjā vālī doll
puja of doll
Consider a sentence such as (47)
47) jo choTī Tīvī hotīī hai usme diwālī
which small TV is in that Divali
ke me nahī ātī baRī
comes big of in not come
`Diwali celebrations are shown in the small TV
not in the big one (cinema theatre)'
This use of the genitive in place of vālā after an adjective is ungrammatical from the adult language point of view. This replacement is an illustration of rule generalization, that is, the use of the genitive with a noun phrase is generalized to an adjective too.
1.2. I have presented above sentences which clearly exhibit the processes of simplification, innovation, extension of syntactic constructions, rule generalization admixture. The literature of pidgins and Creoles is abundant with examples on simplification, innovation and admixture.
A close look at the sentences given in 1.1.3 on admixture reveals that the verbal morphology is that of Hindi but English and Telugu verb stems, nouns, quantifiers were freely used. Interestingly, the child was able to choose the exact underlying verb stem of Telugu in sentences such as (27), (28) and (32) and then the Hindi tense markers were attached. The fact that she could arrive at the correct verb stems in Telugu indicates that the child has the capacity to internalize rules of the language and thus has the capacity to abstract the forms and use them.
The question that arises now is: Why Hindi verbal morphology and why Hindi syntax were chosen? Though the child's exposure to Hindi in the initial stages was much less in comparison to Telugu, Hindi in my opinion; was chosen due to peer group pressure.
The phenomenon of admixture is the child's language is interesting. The child was exposed to three languages simultaneously. She had to choose one as the dominant or main language. As mentioned above, she chose Hindi. The other two languages Telugu and English also played an important role in acquisition. Once Hindi was chosen, Hindi became her `primary' language. Since she was exposed to three languages, the child started constructing sentences choosing freely from three languages though all her sentences reflect the structure of the `primary' language Hindi. It can be hypothesized that for the child closer to that of a pidgin. For a pidgin speaker, there exists a dominant language, his mother tongue and then he is exposed to the other languages. A pidgin evolves in a evolves in a bilingual or multilingual situation. It should however be emphasized that the language which the child came up with should not be considered as a pidgin but the point to be taken note of is that the phenomenon of admixture is common to pidgins and the child's language.
Since the publication of Hymes (1971) on pidginization and creolization, it has been generally accepted that pidgins are simplified codes of the source language and Creoles learnt by the children of the pidgin speakers as first language are complicated and expanded formed of the pidgins Andreson (1979) and Brickerton (1977 a, b) hold the view that the raw material (linguistic data ) for pidginization is the target language input and the primary linguistic data for creolization is the output of pidginization. A pidgin speaker is exposed to the target or source language and he attempts to `abstract' those structures which are sufficient for him to be able to communicate. As his goal is not to acquire the target language, but only functional communication, he attempts to simplify the code. When the children of the pidgin speakers learn the pidgin as the first language, the functions that the creole has to perform also increase, because, the creole is not only a contact language with the target language group but also a contact language with the target group but also a vehicle of communication within the family, with the peer group and also the language of emotion. According to Smith (1972), language has three general functions: Communicative integrative and expressive. The communicative function operates in the transmission of referential, denotative information between persons. By the integrative function, what Smith means is that in order to be a member of a particular social group, a speaker has to sound like the member of that particular group, that is, the speaker attempts to establish his identity with a group by his speech. The speaker becomes a valued member of the group by his linguistic skills such as creative writing, story telling oratory etc. This is the expressive function. Since the goal of a pidgin language is basic communication, that is, to merely convey denotative, referential information and since pidgins are second languages, Smith feels that pidgin languages are restricted to the communicative function only. According to Smith, this functional restriction results in a simplified and reduced code. A creole on the other hand can perform the integrative and referential functions.
The parallels between child languages in the initial stages and a pidgin from the point of view of functions proposed in Smith (1972) are quite clear. The basic goal of the child too is to be able to communicate. In child's goal is to convey denotative and referential information. Group identity or literal creativity do not form part of the functions of the child language.
The restriction in function in a pidgin is due to `lack of social solidarity between speaker and addressee' according to Martin Joos (1971). Schumann (1976) feels that in addition to lack of social solidarity, the lack of `actual or prospective psychological solidarity between the parties' is also a factor. In child language acquisition, however, those factors do not play any role.
There are syntactic innovations in child language and Creoles and also in pidgins to some extent. When a speaker acquires a pidgin, he already has at least one base language and therefore a pidgin is a non primary language. It can hence be predicted that the pidgin would be syntactically, morphologically and lexically influenced by the base language. The learning situation of a creole on the other hand is entirely different. The creole learner does not have a base language and the creole is his first language. Thus, there is no question of interference or influence of another language on the creole. Bickerton (1975) points out that in situation where children do not have sufficient input from any existing language to learn it as their native language, the children create a new language out of the fragments of the languages they hear plus their knowledge of universal grammar.
The reason why a creole has complicated structures, according to Bickerton, is that the lower level complexity of the pidgin leaves a deficit or a gap for the first generation creole speakers and in order to fulfill their communiation needs', the children innovate. Thus, innovation plays a great role in creolization in comparison to pidginization. Sankoff's (1979) view is similar to Bickerton's. She feels that the structural expansion is a creole is due to the native speaker's `need' and is also concomitant with the increasing functional load of a second language' (Sankoff, 1979, p.35). Muhlhausler's (1976) comments on structural expansion are also quite relevant. According to him, linguistic expansion only occurs in situations where large numbers of first language speakers are involved. Creolization is thus the product `communal language acquisition competence'.
's (1979) view creolization as a case of first language acquisition. Anderson 's view seems to be only partly tenable because the similarity between first language acquisition and creole acquisition is that in both the cases the child is exposed to a code. However, the first generation creole speaker is exposed to a pidgin, which is a simplified code with restricted functions, whereas in first language acquisition, the child is exposed to a full fledged code. Anderson
In Varnita's language acquisition process, there is an instance of generation of the periphrastic causatives which were not found in Hindi. The generation of a novel sentence type can be attributed to the capacity of the child to draw from the knowledge of their Universal Grammar. There are instances of such similar developments of morphological causatives in Tok Pisin, a creole in Papus New
. The morphological causative developed in Tok Pisin according to a `specific hierarchy' with the suffix-im being applied successively to stative intransitive, then to true adjectives' ad then to `nonstatives' (Sankoff, 1979, p.30). Guinea
There is similarly between the final stages in the child language acquisition and the process of decreolization. Decreolization is the process where a creole speaker exposed to the target language attempts to master the target language. In such an attempt he has to modify his structures, may be, discard some of the newly evolved syntactic strategies. The goal for the child too is to arrive at the level of competence of the adult. The child gets `corrected' if his tructures do not conform to those of an adult. Varnita is 4 years old now and most of her sentence patterns conform to the adult Hindi language. This clearly indicated that the process of simplification in her language is no longer present. There are certain innovations made in her language but the direction of her language acquisition is more towards conformity to adult language.
1.3 I shall now discuss the implications of innovations, to linguistic change.
It appears that irrespective of the quantum of the input and the nature of the input, that is, whether the child is exposed to one language or more than one, each child is endowed with the innate capacity to innovate. These innovations get `corrected' by adults and therefore one does not normally find novel structures transferring from child language to adult language. Jackobson's (1968) remarks made forty years ago on child language are worth considering. Jackobson points out that the child does not create out of nothingness nor does a child imitate adults mechanically. According to him, `the child creates as he borrows'. The child's language may contain elements which are foreign to the adult model and the child may resist `every attempt at correction'. Child's language with its innovations, Jacobson continues, may remain `frozen' if such children remain isolated. Jackobson mentions of one such case on an isolated Estonian form where some three children retained the `frozen' speech of their early childhood. The innovations in child language can be labeled as what Sassure calls the `particularist spirit' and the driving adult force towards the norm as the `unifying force'. Jackobson conjectures that certain characteristics of child language may be picked up by women and these characteristics may be extended to normal adult language, there by bringing a change in adult language. If Jackobson's conjecture is correct, then child language can bring about linguistic change.
Muhlhausler's (1976) observation that structural expansion occurs only when a large number of first language speakers interact is not a complete statement. A child innovates as and how he learns. As the emphasis in first language acquisition is towards `acculturation to the target language group' (
, 1979), the innovations may die out. In creolization it is the community need that is to fulfilled and therefore the chances of survival of innovations and retention are very high. Thus, innovations are present in first language as well as in Creoles, but so far as the retention of innovations is concerned, it is the creole that is more likely to retain them than the language. Anderson
Further, the creole, in view of its paucity of syntactic structures is more ready to accept innovations than the language which is far more developed than the creole.
Once decreolization starts taking place, and the creole has to conform to the norms of the language, innovations in a creole also would disappear.
1.4. In this paper, I had discussed some syntactic processes which operate in child language and I've attempted to show the similarities of such processes with those in pidgins, Creoles (?). I have also discussed the chances of survival of innovations in child language and Creoles, and their implications to linguistic change.
Given below are some examples of the Hindi (of two speakers) that the child is exposed to at home. SH stands for standard Hindi.
I. Absence of the dative with the dative subjects :
1. tum jana hai to jao
you have to go then go
`go if you have go'.
SH: tum ko jana hai to jao
2. ap karna cahiye
you to do should
will have to
`You to do'
will have to
SH: Ap ko karna cahiye
II. Absence of the genitive
3. ap jane ke bad vah aya
`you to to after he came
`He came after you left'
SH: ap ke jane ke bad vah aya
4. telli pahuncte hi bataugi
Telli right after will telle
`I' ll tell you right after Telli arrives'
SH: telli ke pahuncte hi bataugi
5. ap bat karne se phayde hoga
you to talk by will be helpful
`It' ll be helpful if you talk'
SH: ap ke bat karne se phayda hoga
III. Absence of ne `the ergative'
6. ham khana khaye
we food ate (masc. pl.)
`We ate food'
SH: ham ne khana khaya
IV. Extra ne hypercorrection
7. kis ne mara
who erg. Died
SH: kaun mara
V. Agreement :
8. tum nahī khāyege?
You not will eat
(fam.) (polite, masc.)
`Won't you eat?'
SH: tum nahī khāogā?
VI. Absence of the possessive reflexive
9. sudhita ko mera dawa diya tha
Sudhita to my medicine gave (masc. sg.)
`I gave my medicine to Sudhita'
) sudhita ko apni dawāī dī thī maine
I erg. my
(Poss. refl.) (fem. sg.)
Andersen, R.W. 1979.
`Expanding Schumann's Pidginization Hypothesis'
in Language Learning 29:105-120.
Bickerton, D. 1975.
Dynamics of a Creole system.
Press. Cambridge University
Bickerton, D. 1977a.
`Pidginization and Creolization: Language
Acquisition and Language Universals' in A.
Voldman (ed.) Pidgin & Creole Linguistics,
: Bloomington Press. Indian University
Bickerton, D. 1977b. Creole Syntax. Vol. 2 of Final Report.
Social Sciences and Linguistics Institute,
. Universityof Hawaii
Bolinger, D. 1980. Language, The Loaded Weapon, (The Use
And Abuse of Language Today), Longman.
1971. `Absence of Copula and the Notion of Ferguson, C.A.
Simplicity: A Study of
NormalSpeech, Baby Talk,
Foreigner Talk, and Pidgins' in D. Hymes (eds)
Pidginization and Creolization of Languages'
Press. Cambridge University
Hymes, D. (ed.) 1971. Pidginization & Creolization of
Press. Cambridge University
Jackoson, R. 1969.
Child Language, Aphasia & Phonological
Joos, M. 1971. `Hypotheses as to the, Origin and Modification
of Pidgins' in D Hymes (ed.) Pidginization
and Creolization of Languages, 187-189,
Meisel, J. 1975. The Language of Foreign Workers in
, Germany , Xerox. Wuppertal
Muhlhaesler, P. 1976
Growth & Structure of the lexicon of
Pidgin. Unpublished doctoral dissertation,
. (As quoted in Australian National University
Sankoff, G. 1979 Original not consulted).
Samarin, W.J. 1971. `Salient and substantive pidginization'
In D. Hymes (ed.) Pidginization & Creolization of
Press. Cambridge University
Sankoff, G. 1979 `The Genesis of a Language' in K.C. Hill
(ed.) The Genesis of Language 23-48.
. Ann Arbor
Schumann, J.H. 1974. `The Implications of Inter-language,
Pidginization and creolization for the Study of
Adult Second Language Acquisition'
In TESOL Quarterly 8: 145-152.
Schumann, J.H. 1976. `Second Language, Acquisition: The
Pidginization Hypothesis' in Language Learning
Schumann, J.H. 1978. `The Relationship of Pidginization
Creolization and Decreolization to Second
Language Acquisition' in Language Learning
Smith, D.M. 1972. `Some Implications for the Social Status
Of Pidgin Language in D.M. Smith and R. Shuy
(eds.) Sociolinguistics in Cross Cultural
, Washington D.C. Georgetown
De Linguistique N.5, 543-550.
Subbarao, K.V. 1979. `On the Acquisition of Syntax, in a
Multilingual Situation in Revue Roumaine
De Linguistique N.5, 543-550.
Subbarao, K.V. 1980. `Convergence in Child Language',
To appear in the Proceedings of the Tenth Annual
Conference and the Decennial Celebrations of the
Dravidian Linguistics Association.
Whinnom, K. 1971. `Linguistic Hybridization and the "Special
Case" of Pidgins and Creoles, `in D. Hymes (eds.)
Pidginization & Creolizations of Languages.
Press. Cabridge University