Subject Omission in Early Child Language

Table of Contents

1.0 Introduction

2.0 State of Art
2.1 Principles and Parameters
2.2 The null-subject parameter
2.3 Studies on that field

3.0 Data
3.1 The pro-Hypothesis
3.2 The dual-value-solution

4.0 Discussion

5.0 Conclusion

Appendix

Bibliography

1.0 Introduction

The null-subject parameter or pro-drop parameter is one of the most discussed parameters in language acquisition. It is said that all children omit subjects until a certain age. Some studies claim that this happens due to the initial setting of the null-subject parameter. In this research paper different factors pertaining to the phenomenon of the so-called null-subject parameter will be discussed.

During the essay the majority of the focus will be will be placed on the discussion of that parameter. Therefore, different studies, which examine and try to explain the phenomenon of the null-subject parameter and the omission of subjects in early child language will be consulted, discussed and reflected upon. The main goal of the paper will be to answer the question whether the non-null-subject parameter is set from the beginning in the grammar of children who acquire English as a first language.

Firstly, a brief explanation of the Principles and Parameter Theory will be given and the null-subject parameter or pro-drop parameter and its importance in language acquisition will be explained. To facilitate the explanation of the said parameter some examples of sentences from non-null subject languages, such as English and null-subject languages, such as Italian, Greek and Spanish will be provided. Then, an overview of different approaches, studies and claims considering the null-subject parameter and the omission of subjects will be provided in order to outline the further processing of that essay.

Secondly, it will be moved on to the next part of the paper where two opposed studies, which try to explain why children omit subjects and if the reason for that omission is the null-subject parameter, will be presented and described. In order not to go beyond the scope of this paper, not all the available studies on that topic will be discussed.

In the next part of the paper, the selected studies’ findings will be stated and reflected. Furthermore, these studies will be compared to each other, to show possible differences and similarities regarding the different approaches. Based on that, their importance concerning the omission of subjects and the null-subject parameter will be outlined and discussed critically.

In the last part of the paper, the most important issues will be summarized and evaluated. Based on all findings of this paper and on the preceding discussion, a conclusion will be drawn in order to answer the research question.

2.0 State of Art

This section of the paper will lay out the theory of the Principles and Parameters (Chomsky 1981) in order to explain the phenomenon, which is called null-subject parameter. Afterwards a brief outline about the studies that will be discussed regarding the connection of the subject omission in early child English and parametric approaches will be given. In connection with that, there will be a succinct explanation of the terms competence and performance.

2.1 Principles and Parameters

This part of the paper will explained what the principles and parameters are and how the null subject parameter can be defined. According to Noam Chomsky (1981) there exist principles and parameters in Universal Grammar. There exists a set of principles, which are those aspects of grammar that are given in all languages and are therefore innate. In combination with that there exists a set of structural parameters “which impose severe constraints on the range of structural variation permitted in natural languages […]” (Radford 1997: 21).

2.2 The null-subject parameter

The null subject parameter or pro-drop parameter, which will be discussed in this paper, is the very same parameter that determines if a language allows a subject to be omitted, for example as in Spanish, Italian and Greek, or not, as in English, thus it has “[…] only two possible settings […]” (Radford 1997: ibid.). Taking that assumption into account we must suppose that the null subject parameter has two values and one of these has to be chosen correctly by the child. So, one of those parameters has to be set, which can be done by the child rapidly through minimal linguistic experience (cf. Radford 1997: ibid.). The setting of each parameter is accomplished by triggering. That means that each child has to receive certain linguistic input to choose the right parameter value, which would fit best into the language the child is acquiring (cf. Ayoun 2003: 41). Parameters are arranged in clusters. Thus, a language is restructured and therefore a parameter reset when a property of a parametric cluster is lost. That means that a speaker may keep both values of a parameter at the same time over a period of time, but then a restructuring occurs which leads to the adoption of the final parameter setting (cf. Ayoun 2003: 12).

To explain the phenomenon of the null subject parameter better it will be looked at sentences comparing English and French, which are non-null subject languages, with languages such as Greek, Spanish and Italian, which are null subject languages.

Abbildung in dieser Leseprobe nicht enthalten

Neither in English, nor in French is a construction like that in example 3 possible. There it can be seen that in null subject languages the verb is even marked morphologically (here: -a suffix, -ei suffix) and so the person or subject that is meant can be deciphered by the verb-marker. In English, such a construction would be merely ungrammatical so the difference is that languages such as Spanish (pro-drop languages) do not require an overt subject, whereas the English language does.

In addition to that, according to Haegeman, there are further aspects in which pro-drop languages differ from non-pro-drop languages. The first one is, that „the overt subject can occupy a post-verbal position.” (Haegeman 1994: 20ff.):

Abbildung in dieser Leseprobe nicht enthalten

(cf. Haegeman 1994: 20ff.).

In this example it becomes clear that there are not only morphological differences between null subject languages and non-null subject languages but also syntactic differences are existent. Based on that findings and having explained the main differences between non null subject languages and null subject languages that paper will concentrate on the question whether the non-null subject parameter is already set in early child English or not.

2.3 Studies on that field

In order to explain the phenomenon of the omission of subjects in early child English and furthermore to come to a conclusion if the non-null-subject parameter is innate or if it has to be acquired during the acquisition of a language the concept of competence and performance will be explained. According to Noam Chomsky (1965) competence is the knowledge of language, which usually native speakers possess, thus the innate ability of speaking that language. Performance is the way in which knowledge is put to use in understanding and uttering sentences and in the production of actual utterances, i.e. in conversations or when communicating with others. According to his theory, the performance is not only the use of a language but also the use of competence (cf. Chomsky 1965: 5ff.). On the field of the null-subject phenomenon, there are competence- based approaches as well as performance- based approaches. The two studies which will be examined more detailed in this paper, is the approach of Nina Hyams (1986) on the one hand, who claims that children omit subjects in the early stages of language acquisition due to competence - deficits, so their sentences lack subjects because their grammar differs from the adult’s grammar and thus has to be “restructured“ at some later point (cf. 1986: 6ff.) and on the other hand, the study of Virginia Valian (1990, 1991) who argues that children omit subjects due to performance - deficits. She claims that children effectively have the same grammar as adults and know that subjects are required, but due to processing limitations they cannot apply their knowledge properly (cf. Valian 1991: 22). Nevertheless both of them base their assumptions on how the parameter may be set in early child English.

3.0 Data

In this part of the paper, the data of two approaches will be provided and described. The data presented by Hyams (1986) will be described first and afterwards the data presented by Valian (1990, 1991) follows. Hyams and Valian do not share the same opinion regarding the omission of subjects in early child language; therefore it is interesting to look at both assumptions. On the one hand, Hyams (1986) states that English-speaking children do not have the same grammar as English-speaking adults, which she considers as a reason to explain why subjects are omitted by children (cf. Hyams 1986: 6). Due to that assumption, she claims that all children start acquiring a language by using grammatical structures similar to those in Italian^[1]. Valian, on the other hand, claims that the single-value solution, which means that the parameter is either set to the English-like value or to the Italian-like, cannot work because of the inability of the child’s parser to decode the input. Therefore, Valian suggests that both values of the parameter have to be provided, so that the child decides during his language acquisition which value would be best choice according to the language the child is speaking (cf. Valian 1990: 105). In the following the data of both assumptions will be provided, starting out with Hyams.

3.1 The pro-Hypothesis

Hyams claims that children start out with the pro-drop parameter, which is set to the value that allows null-subjects, i.e. the Italian-like value (cf. Hyams 1986: Preface x). She proposes a variation of the null-subject parameter, which she calls AG/PRO-parameter. In languages were AG=PRO, subject omissions are allowed (cf. Hyams 1986: 32). AG means agreement and includes agreement for person, number and gender in respect to the subject (cf. Hyams 1986: 12). The assumption of Hyams is that children omit subjects together with expletives such as it and there, but also modal verbs. She claims that as soon as children stop to omit subjects, they acquire as well the correct usage of expletives as well as of modals and that those elements are the trigger to provoke the resetting of the parameter. The data she examined is taken from Bloom, Lightwood, and Hood 1975. The children are aged from 1-1/2 to 3 years. She notes that sentences such as: “Eating cereal.” and: “Outside cold.”, which are taken from the corpora of Eric during taping session Time II and III, are very frequent in early child speech, but she also notes that subjectless sentences also coincide with sentences that contain subjects (cf. Hyams 1986: 63-64), as she shows by using the data of Kathryn during taping Time I until time III. She compares these examples with Italian examples. (7) Kathryn I throw away Mommy throw it away helping Mommy Lois coming

(8) Kathryn II no want this no Kathryn want play with self no like celery, Mommy Kathryn no like celery

(9) Kathryn III want go get it I want take this off see Lois n face I see Kathryn in mirror

Italian: (10) Taglio (11) Questo è mio papa ‘cut’ (= I cut) ‘This is my daddy

(English examples taken from Bloom, Lightwood, Hood 1975 cited in Hyams 1986: 65-66; Italian examples cf. Hyams 1986:111).

That leads Hyams to assume that the coexistence of subjectless sentences and sentences containing subjects, excludes a performance-deficit for the explanation of null-subjects. She also states that the missing expletives are properties, which are assigned for null-subject languages, and therefore she takes that as evidence to suppose that English-speaking children start out with the null-subject parameter set as the initial value. (Hyams 1986: 70). To illustrate it, she presents utterances taken from Bloom, Lightwood and Hood (1975): (12) Outside cold (‘It’s cold outside’) No morning (‘It’s not morning’) No more cookies (‘There’s no more cookies’)

(examples taken from Bloom, Lightwood, and Hood in Hyams 1986: 63). Hyams explains that expletives are lacking because pronouns are used in null-subject languages only to emphasize something or due to pragmatic reasons. So, mostly they are not necessary. She links her assumption to the “Avoid Pronoun Principle” (Chomsky 1981), which predicts that lexical pronouns can be omitted whenever they are not required, that is whenever “null pronominal is possible” (Hyams 1986: 72). In order to show that the omission of pronouns is possible, Hyams compares the following sentences:

[...]

^[1] Henceforth, the parameter-value that allows null subjects (+pro-drop) will be termed Italian-like value. The parameter-value that does no allow null subjects (--pro-drop) will be termed English-like value.