Journal of Language and Linguistics Volume 1 Number 3 2002 ISSN 1475 - 8989

1. Introduction

One of the major issues in grammar pedagogy has been the instructional balance between prescriptive and descriptive grammars in classroom implementations (Celce-Murcia & Larsen-Freeman 1999). Traditional grammar pedagogy, which has often been described as rule-based and single-sentence oriented, tends to overemphasize prescriptive rules and underestimate such functional aspects of language as discourse or context features (e.g. Celce-Murcia & Larsen-Freeman 1999; Nelson 1993). Thus, several pedagogical implications were advocated in the past. For instance, Ellis (1993, 1995), Fotos (1993, 1994), Rutherford (1987), Sharwood Smith (1988), and Yip (1994) claimed that grammatical consciousness-raising or input analysis exercises should be incorporated into classroom instructions. In addition, Celce-Murcia (1991a, 1991b) and DeCarrico (2000) suggested that grammar teachers teach grammar in a discourse context. Descriptive, functional aspects of grammar in instructional materials and activities became primary characteristics in current grammar pedagogy. Accordingly, today's English teachers are required to be familiar with and to be able to teach not only accurate sentential form and meaning but also accurate use of a sentence at suprasentential, discourse level (Celce-Murcia & Larsen-Freeman 1999; Conrad 2000; DeCarrico 2000). Furthermore, English teachers are also expected to be able to explain "why speakers of English choose to use one form rather than another when both forms have more or less the same grammatical or lexical meaning" (Celce-Murcia & Larsen-Freeman 1999:5).

A relative clause in English, "a type of complex postnominal adjectival modifier used in both written and spoken English" (Celce-Murcia & Larsen-Freeman 1999:571), is said to have two main functions: to modify a noun and to give detailed information about the person or thing the noun refers to (e.g. Thomson & Martinet 1986). Generally, relative clauses in English follow the nouns they modify and are introduced by such a relative pronoun as 'who,' 'which,' or 'that.' However, when a relative pronoun is not placed in the subject position, the relative pronoun can be deleted in restrictive relative clauses. Thus, as we can see in Example A below, all of these sentences are grammatical without relative pronouns.

Hurford (1994) mentions that since we can use a relative pronoun 'that' before a relative clause in each sentence above, "the relative pronoun can be omitted altogether anywhere where 'that' can be used, except when the 'shared' position inside the relative clause is the subject position" (219). Accordingly, it is possible to delete the relative pronoun that replaces an object of a verb or a preposition altogether except the case that "the preposition has not been fronted along with the relativized object" (Celce-Murcia & Larsen-Freeman 1999:581). Therefore, it is practically possible to delete the relative pronoun in sentence (d) only (see Example B). Sentence (g) is ungrammatical because, as DeCarrico (2000) points out, there is the 'structural gap' that has to be filled up by moving 'whom' to the beginning of the clause. In order to make the sentence (g) grammatical, we should either place 'about' right after 'talked' or insert 'whom' after the preposition 'about' so that the preposition is placed right next to the relative pronoun.

Although a few structural restrictions can be found in the relative pronoun deletion, whether we should delete or retain relative pronouns in restrictive relative clauses has been left unclarified. In fact, most of the grammar reference books and textbooks touch on the relative pronoun deletion although they merely claim that deletion is more preferable in speaking than writing, and that one of the main reasons why deletion is preferred in speaking is that speaking, in general, is less formal in style than writing (see Azar 1992, 1999; Hurford 1994; Thomson & Martinet 1986). In addition to the discourse modality (speaking vs. writing) and formality of respective discourse, however, there may be some other factors involved in the relative pronoun deletion and retention in writing. According to Celce-Murcia and Larsen-Freeman (1999), one possible factor is a usage tendency or preference affected by discourse genres or topics, and the other is the animacy or humanness of the referent. The present article reports on a preliminary corpus analysis on native-speakers' preference of relative pronoun deletion/retention in relation to discourse-genre specific features. Based on research findings, some implications for English teachers are discussed.

2. Research Background

Olofsson (1981) examined the Brown University corpus to clarify lexical co-occurrence in relative clauses. An interesting finding of the study is the preference to use 'that' for the relativized subject found in the relative clauses including body parts as head nouns. The study also reported that when 'every' was used to modify the head noun, the relativized object was always deleted as in 'Every student I met two days ago studied English very hard.' Besides these findings, three syntactic environments in which a certain relative pronoun is preferred over another are also presented: (a) 'that' is preferred when there is relativized nonhuman [- human] subject; (b) relative pronoun is preferably deleted when there is relativized nonhuman [- human] object; and (c) 'which' is preferred when there is relativized prepositional object. Even though these environments seem to be valid, Olofsson did not take into consideration discourse genres or topics that might change one or more of the above environments.

The structural environment (a) was not considered relevant to the present study because the relative pronoun deletion is not possible for the relativized subject as we have previously discussed. As for (c), it is, of course, impossible to find the relative pronoun deletion when the preposition has been fronted along with the relative pronoun. However, the use of 'that' or the relative pronoun deletion is an option when the preposition has not been fronted along with the relative pronoun. As for (b), since discourse genres or topics were not fully investigated in the Olofsson's study, it is probable that the retention of the relative pronoun over deletion may be preferred in different discourse genres or topics. On the basis of these speculations, an attempt was made to examine the relative pronoun deletion and retention in restrictive relative clauses using such different linguistic corpora as letters exchanged between magazine editors and readers published in The Humanist and scientific articles in Scientific American. Examined texts from The Humanist and Scientific American were randomly selected from issues published in 1999 and 2000, respectively. Approximately, a total of 20,000 words were investigated.

3. Data Analysis and Discussion

3.1. Letters in The Humanist

A corpus analysis on the letters exchanged between editors and readers of The Humanist revealed that the deletion of the relative pronoun is preferred when the relative pronoun has a [- human] antecedent. Thus, the structural environment (b) in the Olofsson's study was confirmed. Excerpts from the letters in The Humanist are as follows. As we can see in these excerpts, the relative pronoun deletion is applied in a restrictive relative clause that has a [- human] relativized object.

There is only one exceptional sentence, however, which retains 'that' even though its relative clause has a relativized [- human] object. What follows is an excerpt that shows this exception.

Although the relative pronoun 'that' is not deleted in the excerpt above, the fact that any other relative pronouns for the relativized [- human] object are deleted illustrates that the relative pronoun deletion is more preferable than retention in the specific discourse genre 'letters' in The Humanist.

Furthermore, the Olofsson's structural environment (c) was also confirmed in the corpus analysis on The Humanist. Namely, every preposition is fronted along with the relative pronoun that functions as an object of the preposition, and thus the relative pronoun (i.e. which) is retained as in the following excerpt.

3.2. Scientific Articles in Scientific American

A corpus study on the scientific articles in the Scientific American did not fully confirm the Olofsson's structural environment (b). Some example cases are as follows:

In all the above excerpts, even though the relativized object is [- human], the relative pronoun 'that' is not deleted. Namely, this fact may imply that the retention of the relative pronoun is more preferred over deletion in the specific genre 'scientific article' of Scientific American. Although the number is fewer than that of the retained relative pronouns, several deleted relative pronouns were also found. Nevertheless, most of the cases were found in interview transcripts cited in the articles. Thus, the relative pronoun deletion might have been done because of the modality (speaking). The following excerpts show two examples of the deleted relative pronouns found in such interview transcripts.

Moreover, the deleted relative pronouns were also found in several sentences that are typically long and complex and have already contained one or more relative pronouns. The following excerpts show a few examples of this finding.

Additionally, the Olofsson's structural environment (c) was well confirmed in the corpus analysis on Scientific American. That is, all prepositions but one exception are fronted along with the relative pronoun that functions as an object of the preposition, and thus the relative pronoun 'which' is retained. The exception can be found in the following excerpt where the relative pronoun is deleted due to the fact that the preposition has not been fronted.

4. Conclusion and Implications for English Teachers

In the present study, an attempt was made to examine the relative pronoun deletion and retention in restrictive relative clauses using the corpora of the two different genres: published letters exchanged between magazine editors and readers of The Humanist and scientific articles in Scientific American. Despite the preliminary nature of the study, the present corpus analysis verified some discourse genre-specific effects on the deletion and retention of the relative pronoun. In what follows, major findings are summarized in comparison with Olofsson' (1981) research.

The present study also depicts the fact that the description on the relative pronoun deletion that almost all grammar reference books and textbooks give to readers (e.g. the major reason why deletion is preferred over retention in speaking is that speaking, in general, is less formal in style than writing) is not appropriate. There are other factors affecting the relative pronoun deletion in addition to the discourse modality; such factors as structural/grammatical restrictions and discourse genres or topics should be taken into account. Accordingly, students in grammar class or writing class should be exposed to various types of spoken and written linguistic patterns in authentic materials (see Biber, Conrad & Reppen 1994) to become familiar with various use conditions which prefer the relative deletion/retention; ESL/EFL teachers should take into account not only the discourse modality but also genres or topics to teach the relative pronoun deletion (or retention) effectively. Moreover, even though it turned out to be less important than genres or topics in the present study, the animacy or humanness of the referent should also be taken into consideration.

Current grammar pedagogy requires English teachers to know and teach accurate use conditions and native-speaker preferences. Considering the fact that language is functional and systemic in nature (Christie 1999), this is a natural requirement for language teachers. However, there may be some difficulty in the real teaching situation: explanations about accurate use conditions or context-specific native-speakers' preference of one form over another might be challenging to non-native teachers of English: especially, when they teach in EFL (English as a Foreign language) contexts (e.g. Japan, Korea, Taiwan, etc.), there may be no immediate, sufficient access to native speakers of English. According to Conrad (2000), corpus-based research is a key solution for English teachers to cope with the difficulty. Without any access to native-speakers, for instance, non-native teachers of English can analyze functional aspects of English lexicogrammar and use some findings for pedagogical practices. In this regard, non-native teachers of English should consider corpus-based research a research option for action research, which is generally conducted to pose and solve problems found in the status quo of teaching, to generate some change, and to reflect on their teaching practice (e.g. Bailey, Curtis & Nunan 2001; Gebhard & Oprandy 1999). Needless to say, native English teachers can, of course, benefit from corpus-based research because much broader samples of typical uses of grammar structures and lexicon will be available for instructional purposes (Conrad 2000).

About the Author

Toshinobu Nagamine is currently a doctoral candidate in Composition and TESOL at Indiana University of Pennsylvania, USA.

Email: tn_73@yahoo.co.jp

References

Andersen, J.L., Schjerling, P. & Saltin, B. (2000, September). Muscle, genes and athletic performance. Scientific American, 283, (3), 49-55.

Azar, B.S. (1992). Fundamentals of English grammar. (2nd ed.). Englewood Cliffs, NJ: Prentice Hall.

Azar, B.S. (1999). Understanding and using English grammar. Englewood Cliffs, NJ: Prentice Hall.

Bailey, K.M., Curtis, A., & Nunan, D. (2001). Pursuing professional development: The self as source. Boston: Heinle & Heinle.

Biber, D., Conrad, S. & Reppen, R. (1994). Corpus-based approaches to issues in applied linguistics. Applied Linguistics 15, (2), 169-189.

Burton, J., (1999, July/August). Economic enlightenment. [Letters]. The Humanist, 59, (4), 2.

Celce-Murcia, M. (1991a). Discourse analysis and grammar instruction. Annual review of applied linguistics, 11, 459-480.

Celce-Murcia, M. (1991b). Grammar pedagogy in second and foreign language teaching. TESOL Quarterly, 25, 135-151.

Celce-Murcia, M., & Larsen-Freeman, D. (1999). The grammar book. (2nd ed.). Boston, Mass.: Heinle & Heinle.

Christie, F. (1999). Genre theory and ESL teaching: A systemic functional perspective. TESOL Quarterly, 33, (4), 759-763.

Conrad, S. (2000). Will corpus linguistics revolutionize grammar teaching in the 21st century? TESOL Quarterly, 34, (3), 548-560.

DeCarrico, J. S. (2000). The structure of English: Studies in form and function for language teaching. Ann Arbor: University of Michigan Press.

DeTar, R. P. (1999, March/April). The menace of 'Jury Nullification.' [Letters]. The Humanist, 59, (2), 2.

Edwords, F. (1999, November/December). Retraction and correction. [Letters]. The Humanist, 59, (6), 2.

Ellis, R. (1993). The structural syllabus and second language acquisition. TESOL Quarterly, 27, 91-113.

Ellis, R. (1995). Interpretation tasks for grammar teaching. TESOL Quarterly, 29, 87-105.

Ezzell, C. (2000, July). Beyond the human genome. Scientific American, 283, (1), 64-69.

Fotos, S. (1993). Consciousness and noticing through focus on form: Grammar tasks performance versus formal instruction. Applied Linguistics, 14, 385-407.

Fotos, S. (1994). Integrating grammar instruction and communicative language use through grammar consciousness-raising tasks. TESOL Quarterly, 28, 323-351.

Gebhard, J.G., & Oprandy, R. (1999). Language teaching awareness: A guide to exploring beliefs and practices. New York: Cambridge University Press.

Howard, K. (2000, July). The bioinformatics gold. Scientific American, 283, (1), 58-63.

Hurford, J. R. (1994). Grammar: A student's guide. Cambridge: Cambridge University Press.

Nelson, E. S. (1993). Adding discourse-level practice to sentence-level exercises. TESOL Journal, 2, (3), 30-32.

Olofsson, A. (1981). Relative junctions in written American English. Gothenburg Studies in English 50. Gothenburg: Acta Universitatis Gothoburgensis.

Rutherford, W. (1987). Second language grammar: Learning and teaching. London: Longman.

Sharwood Smith, M. (1988). Consciousness raising and the second language student. In M. Sharwood Smith (Ed.), Grammar and second language teaching (pp. 51-60). New York: Newbury House.

Thomson, A. J., & Martinet, A.V. (1986). A practical English grammar. (4th ed.). Oxford: Oxford University Press.

Yip, V. (1994). Grammatical consciousness-raising and learnability. In T. Odlin (Ed.), Perspectives on pedagogical grammar (pp. 123-138). Cambridge: Cambridge University Press.

Zeilinger, A. (2000, April). Quantum teleportation. Scientific American, 282, (4), 50-59.