This paper aims at exploring how the views on language acquisition have developed over time, where they conflict each other and whether it is possible to position oneself on either side.
The “nature vs nurture” debate is an everlasting discussion, not only in the area of linguistics, but science in general. This is what makes it particularly interesting because there is always new information to expect. Furthermore, it is an issue not limited to academic discourse only, but one that appears in everyday conversation.
An interesting aspect of language acquisition, more specifically the knowledge of language, is the question if children arrive at such knowledge by the mere exposure to “positive evidence”, i.e. the language of adults they are surrounded with, or if negative evidence also has an impact, i.e. children’s knowledge of language is based on explicit or implicit feedback to whether an utterance was in some way incorrect. The two terms will be defined and looked into in more detail in the second chapter.
In the following, the correlation between the “nature vs nurture” debate and language acquisition will be explained first. The respective chapter provides a brief insight into how the nativist approach developed, mainly based on Chomsky’s (1988) notion of “Universal Grammar” (UG), and the type of criticism it brought along. It further explains the connection between the respective theories and the grammatical aspect of language acquisition. In the last section, the main problems of experience-based language learning will be displayed and defined.
The third chapter starts with a section dedicated to where the particular view that humans need innate structures for a correct knowledge of language stems from. It continues with a detailed look on the “(no) negative evidence” problem. The (non-)existence of negative evidence and which role it plays in terms of language acquisition will be explored from different standpoints by nativists as well as non-nativists.
Table of contents
1. Introduction
2. Language acquisition: The nature vs nurture debate
2.1 Theoretical background
2.2 Acquisition of grammar
2.3 Negative and positive evidence
3. The nativist approach
3.1 Why nativist views exist
3.2 The “(no) negative evidence” problem
3.2.1 Inconsistency of negative evidence
3.2.2 Negative evidence as resource for grammar acquisition
4. General discussion
5. References
1. INTRODUCTION
In modern science, one of the most discussed issues is the “nature versus nurture” debate. This also relates to the question about how we acquire language. During the last century, there have been mixed opinions and theories supporting either a “nativist” or an opposing approach (e.g. “social interactionism”) or both, that is: are we born with innate mechanisms that make language learning possible or is it entirely based on our social environment? This paper aims at exploring how the views on language acquisition have developed over time, where they conflict each other and whether it is possible to position oneself on either side.
The “nature vs nurture” debate is an everlasting discussion, not only in the area of linguistics, but science in general. This is what makes it particularly interesting because there is always new information to expect. Furthermore, it is an issue not limited to academic discourse only, but one that appears in everyday conversation.
An interesting aspect of language acquisition, more specifically the knowledge of language, is the question if children arrive at such knowledge by the mere exposure to “positive evidence”, i. e. the language of adults they are surrounded with, or if negative evidence also has an impact, i.e. children’s knowledge of language is based on explicit or implicit feedback to whether an utterance was in some way incorrect. The two terms will be defined and looked into in more detail in the second chapter.
In the following, the correlation between the “nature vs nurture” debate and language acquisition will be explained first. The respective chapter provides a brief insight into how the nativist approach developed, mainly based on Chomsky’s (1988) notion of “Universal Grammar” (UG), and the type of criticism it brought along. It further explains the connection between the respective theories and the grammatical aspect of language acquisition. In the last section, the main problems of experience-based language learning will be displayed and defined.
The third chapter starts with a section dedicated to where the particular view that humans need innate structures for a correct knowledge of language stems from. It continues with a detailed look on the “(no) negative evidence” problem. The (non-)existence of negative evidence and which role it plays in terms of language acquisition will be explored from different standpoints by nativists as well as non-nativists.
Finally, the discussion will be summarized and followed by the author’s point of view. Then, an outlook will be given, which deals with the question whether language acquisition can be explained by means of either a nativist or a contrasting approach respectively and what further research to tackle to explore the presented theories in a different context.
2. LANGUAGE ACQUISITION: THE NATURE VS NURTURE DEBATE
This chapter starts with an introduction to Chomsky’s (1988) nativist theory and an opposing theory, namely “social interactionism”. It continues with a section explaining the relation between these theories and a particular aspect of language acquisition, which is the development of correct knowledge of grammar. The final part of the chapter presents specific problems pertaining to the notion of acquiring language through experience, which build the base for a nativist approach to language acquisition. One of these problems involves the role of negative evidence, which will receive special focus throughout the paper.
2.1 Theoretical background
The “nature vs nurture” debate is primarily concerned with children’s syntactical development, as the main focus in context is the acquisition of grammar and how it is accomplished in an unbelievably short amount of time.
Research into a child’s syntactical development involves knowledge about its individual stages and problems he or she faces. Such knowledge is the basis for proper analysis and discussion of specific language acquisition theories.
In most of the 20th century, there was a strong belief in the idea of a “blank state of mind” (cf. Pinker 2004: 5). This refers to the assumption that nothing is innate, i.e. man has no nature. Before Chomsky, there was behaviorism. Behaviorists explain psychology only by what is observable, more specifically, behavior is seen as a response to certain stimuli. Their approach is heavily based on environmental factors and other factors determining behavior are hidden in the “black box”, i.e. difficult internal cognitive processes that are hidden or cannot be accessed.
This resulted in an attack on part of Noam Chomsky, who in relation to language believed that the behaviorist approach cannot adequately account for learning a concept as complex as language. He backs his opinion with the following reasons:
1. Behaviorism cannot explain certain structural aspects of language acquisition, for example naming explosion, which typically occurs at the age of 18 - 24 months (cf. Pruett 1999).
2. Language acquisition depends on more than simple reinforcement; grammar is not explicitly taught, but it is still acquired.
3. There is an infinite number of possible responses to an utterance.
4. Linguistic abilities are underdetermined by input.
Chomsky thus developed a particular innateness theory, which has influenced other nativists and builds the foundation for this paper. According to Chomsky, there is an innate machinery responsible for language comprehension and production, which is also referred to as the “Language Acquisition Device” (LAD). He proposes a “generative grammar”, an innate feature in our brain specifically about grammar. Language acquisition is therefore genetically predetermined; the Chomskyan theory suggests that we are born with a set of rules, i.e. UG, a property of all humans which the development of language is based on. The environment determines only which particular language will be acquired.
The concept of UG bears some problems, though. Research shows that children start using specific grammatical features of their mother tongue at a very early stage (cf. Bowerman 1989: 161), which contradicts the idea that all children start with UG, which then develops into the grammar of their mother tongue.
“Social interactionism” is an opposing response to nativism. It constitutes the idea that language acquisition is based on experience and input as we grow. Therefore, social interactionism stresses the role of external factors, as opposed to innate principles. In contrast to Chomsky’s original theory, where language perception and production is a cognitive ability independent from others, social interactionism emphasizes the interrelation of the mind, body and language. Social interactionists rely on the matter that humans interact with the world via sensory experience and motor movement. It is by now well known that the ability to acquire language interacts with other cognitive abilities and does not function independently, as nativism proposes.
Social interactionism mainly focuses on parent-child interactions in language acquisition research. This will be further discussed in the third chapter with regard to nativist and nonnativist views.
2.2 Acquisition of grammar
“Do children eliminate their grammatical errors solely on the basis of internal mechanisms, or do they require external feedback from their parents?” (Marcus 1993: 54) This quote perfectly forms the research question for the paper at hand. Acquiring a language requires the production and the subsequent elimination of grammatical errors. The question above highlights the opposing assumptions that children learn language based on either innate structures or the influence of the speech community in their environment.
The “nature vs nurture” debate, especially the notion of UG, is primarily concerned with how we learn grammar, or more specifically, how we distinguish between what is grammatical and what is ungrammatical. The main problem lies in the fact that despite having different input, all children and adults alike arrive at the same end result, which is the knowledge about the correct use of a language.
An important point is made by Crain and Pietroski (2001: 141): “We assume that language learning involves the acquisition of a grammar, where a grammar is a cognitive resource that allows its possessor to generate and recognize endlessly many expressions of the language”. This is in line with Chomsky’s understanding that one can produce an unlimited number of possible correct expressions, including ones that were never even heard before. Ambridge, Pine, Rowland, Jones and Clark and Kasper (2009: 1302) give the following example and explain:
[A]ll native English speakers are able to produce utterances with novel verbs such as The man pilked the ball on the basis of hearing utterances such as The ball pilked, yet, at the same time, are able to avoid producing *The clown giggled the girl despite hearing The girl giggled. Furthermore, children are known to actually produce overgeneralization errors of this type (e.g., *Don’t giggle me; *1 ’mjust gonnafall this on her; from Bowerman, 1982) before “retreating” from such overgeneralizations later in development.
The question whether children need external input or internal mechanisms to acquire language is related to the fact that they use the target language incorrectly (such as overgeneralizations), but are capable to recover from the errors they make. In terms of external input, children would need to be explicitly informed about the ungrammaticality of their linguistic output. Reaching the level of the adult grammar system is not driven by learning (cf. Crain and Pietroski 2001: 167), though, it happens rather naturally. In their study about parental feedback and child grammaticality, Brown and Hanlon (1970: 48) found out that even though approval and disapproval is present, “they are almost always semantic or phonological. Explicit approval or disapproval of either syntax or morphology is extremely rare in [their] records and so seems not to be the force propelling the child from immature to mature forms”. The apparent lack of interrelation between parental feedback and child grammaticality raises an issue whether negative evidence is required for children to eliminate errors from their system.
This leads back to the concept of UG. White (1989: 4)justifies its existence, claiming that there is a mismatch between the primary linguistic input and the system actually attained, for which UG has to be the solution. In other words, linguistic input alone does not suffice for children to arrive at the end result of an adult-like grammar system. There seems to be a need for principles “which constrain the form and functioning of grammars” and “interaction between the innate UG and the linguistic input from the language acquired” (White 1989: 5). White lists three major problems relating to the assumption that language acquisition is a result from the influences of the speech environment alone: “underdetermination”, “degeneracy”, and negative evidence. The next subchapter looks into these problems, the main focus being on negative (and positive) evidence.
2.3 Negative and positive evidence
White is a strong nativist who distinguishes between three problems regarding acquisition solely based on input, which therefore, as reported by her, lead to the existence innate linguistic mechanisms: underdetermination, degeneracy and negative evidence.
Underdetermination is in accordance with the above standing quote by Ambridge et al. (2009):
A number of aspects of language are underdetermined by the input. That is, the linguistic competence of children and adults includes properties which are not immediately obvious and which are not explicitly taught; the grammar that underlies our language use, which has to be acquired by children, goes far beyond the actual sentences that an individual learner may happen to have been exposed to. (White 1989: 5)
If language acquisition is solely based on input a child receives, he or she would be likely to make overgeneralizations. White (1989: 6) refers, among others, to the contraction of“want to” to “wanna”; it is possible in some cases, as in “Who do you -want totwanna see?”, while in some it is not: “Who do you -want to/*wanna feed the dog?” As children are not made aware of which contraction is grammatically correct or incorrect, they still develop an accurate judgement of its grammaticality. If learning depends on input alone, the chances of the child contracting “want to” in the latter sentence would be higher, as a result of overgeneralization. White’s (1989: 7) explanation for why this does not happen is UG:
Wh-questions are derived by movement of the w^-word from an underlying position to the front of the sentence. An empty category (or trace t) marks the position from which the wh- question word has moved. (This proposal accounts for a wide range of syntactic phenomena and was not made specifically to deal with the wanna phenomenon). [...] Wanna- contraction is sensitive to the presence of w^-traces. If a trace intervenes between want and to, contraction is impossible.[...] None of this information is obviously present in the input, since traces are an abstraction.
The respective proposal is a common rule, though, as stated by White herself, so even though it sounds plausible, one has to question if the language learner was exposed to traces regardless of -wanna-contractions and is thus able to apply it correctly in other contexts.
The second problem White (1989: 12) mentions is that of degeneracy. This refers to the fact that children are exposed to any kind of linguistic input, may it be grammatical or ungrammatical: “Adults make mistakes, hesitate, change their minds about what they are going to say, etc.” She continues to argue that the lack of knowledge about whether sentences are grammatical or not “could make the extracting of generalizations extremely difficult”.
[...]