The role of universal grammar in Second Language Acquisition (SLA)

1  Introduction  2

Contents

1  Introduction  3

2  The idea of Universal Grammar  4

2.1  Introducing UG  4

2.2  UG vs. Input  5

2.2.1  Underdetermination  5

2.2.2  Degeneracy  7

2.2.3  Negative evidence  7

2.3  Principles and parameters of UG  8

2.3.1  Principles  8

2.3.2  Parameters  10

3  UG in Second Language Acquisition  12

3.1  Interlanguage grammar  12

3.2  L1 and L2 in comparison  13

3.3  Major positions  14

3.3.1  UG is completely available in SLA  14

3.3.2  UG is no longer available in SLA  16

3.3.3  UG is partially available in SLA  17

4  Conclusion  19

5  References  20

1 Introduction 3

1 Introduction

This piece of work intends to present the most cogitable positions concerning the functioning of universal principles in Second Language Acquisition (SLA) as they are commonly considered to work within First Language Acquisition (FLA). It will transpire that various aspects of SLA compared to FLA do not allow for a lucid and irrevocable answer to be given for this question.

In order to give the reader an idea of what Universal Grammar (UG) deals with, the first part of the paper sets up a conception of FLA to work on the basis of innate knowledge. Special emphasis is put on the content of Input as well as the realization of UG in terms of principles and parameters.

Following SLA is scrutinized more closely with respect to the acquisition process and differences concerning the preconditions of both FLA and SLA before turning to the discussion of the major positions in linguistic research.

2 The idea of Universal Grammar 4

2 The idea of Universal Grammar

In the first part of this paper the basic issues Universal Grammar is concerned with will be introduced. This knowledge is necessary in order to understand why the idea of UG seems to hold true for First Language Acquisition but undergoes a rather controversial debate in the field of Second Language Acquisition.

2.1 Introducing UG

Languages people use to make themselves understood to each other are generally divided into First Language (L1) and Second Language (L2). The L1 of a learner is also called her mother tongue, i.e. the language children are exposed to in their social environment, in most cases realized as the social structure of a family. People who speak a certain language as their mother tongue are considered to be “native speakers” of this language. L2 refers to any language other than the native that is acquired or learnt for various reasons, e.g. as a part of school education, personal interest or integrating into another speech area.

The knowledge of a first language is thought to be the “grammar” of the language, i.e. a set of rules which by means of the lexicon (the mental vocabulary of a language) allow speakers of the language to communicate verbally. Surprisingly at first sight, all L1 acquirers of a certain language appear to reach roughly the same level of grammar, being capable of understanding and producing sentences they might not even have heard before.

A commonly accepted explanation for this phenomenon among linguists is the “innateness hypothesis”, a theory introduced by Noam Chomsky, proposing that all individuals dispose of an innate knowledge that determines underlying principles of grammar all languages have in common. Therefore this mental system is called “Universal Grammar”.

“UG provides a genetic blueprint, determining in advance what grammars can (and cannot) be like.” (White 2003:2).

2 The idea of Universal Grammar 5

This doesn’t mean that all languages to some extent have the same grammar, but rather that there are certain “principles” which allow children of any speech area to judge the accuracy of certain utterances and reach the complete knowledge of the respective grammar of their L1, even if they are exposed to less input than others. UG can be considered to be a “built in” mediator between the input data of a language and the final grammar representing the complete competence of the same language:

(1) (White 1989:5)

2.2 UG vs. Input

So far it was proposed that the fact of adults arriving at a very similar grammar structure after having completed the L1 acquisition process can be explained in terms of some kind of innate assistance. Still the question is not yet answered which role input data plays and why it is considered to fail as the only force to bring out knowledge of grammar.

Linguists try to describe the divergence of the adult grammar from the actual input by means of the so called “projection problem”. This mainly includes three circumstances which show that grammar rules cannot only derive from input:

2.2.1 Underdetermination

Certain features of a language are said to be “underdetermined by the input” (White 1989:5). In other words, these features are neither explicitly taught by parents nor abstracted from the data children are exposed to since the core of the

2 The idea of Universal Grammar 6

features’ relevance is outside the utterances the learners hear. If the learners would make generalizations about language barely on the basis of input, thus only using their cognitive capacity, one would expect them not to succeed when exceptions of the generalizations made occur.

An example for this is the distribution of reflexive pronouns. These always involve a noun phrase to which they refer, i.e. they are coreferential. In (2) the intricate and at first sight apparently random possibilities of coreference of constituents are illustrated. Their complexity support the statement that it is rather implausible that children try to find out the rules when to apply for a certain type or not on a trial and error basis, since children simply do not make errors such a strategy would evoke. The words in italics are tested to be coreferential:

(2) (White 1989:9)

a. John saw himself

b. *Himself saw John

b. Looking after himself bores John

c. John said that Fred liked himself

d. John told Bill to wash himself

e. John showed Bill a picture of himself

Several possibilities can be drawn from these examples:

the reflexive pronoun usually follows the antecedent (a., b.) - but not always (c.)

the reflexive usually is in the same clause as the antecedent (a., c.) - but not always (d.)

there can be more than one antecedent (e.) - but there does not have to (a., c., d.)