Relationship Between Standard And Non Standard Grammar English Language Essay

Published: Last Edited:

This essay has been submitted by a student. This is not an example of the work written by our professional essay writers.


The relationship of standard and nonstandard grammar is a matter of crucial materials for teacher and language learners. However, the distinction between them is not always clearly understood. Therefore, this paper will attempt to clearly distinguish them and analyse the relationship between them. In doing so, we should discuss corpora because not only they have much influenced the relationship between them, but they play a very important role in pedagogy. Thus, this paper will be arranged as follows: explaining chronologically the relationships between them in the first half, and analysing thematically corpus data in the other half.

The relationship between Standard and Non-standard Grammar

Grammar can be classified into two parts: standard and nonstandard grammar. Although slightly different, we can say approximately that intuition-based grammar, prescriptive grammar, and written grammar can be categorised as standard grammar, while observation-based grammar, descriptive grammar, communicative grammar and spoken grammar can be categorised as nonstandard grammar. As mentioned in the introduction, the relationship and distinction have been attempted to be explained by linguistics with many theories.

1.1 Intuition based grammar and Observation based grammar

Aarts (1991) describes an intuition-based and an observation-based grammar in order to explain the relationship between standard and nonstandard grammar. In his description, an intuition-based grammar produces "grammatical" sentences, while an observation-based grammar produces "acceptable" sentences (p. 47). Although His explanation of the relationship between them seems to be different, he strongly argues that the two types are not different.

Aarts (1991) argues that through its confrontation with corpus data which can enable us to use a great amount of evidence of actual occurrences produced by its native users, intuition-based grammar becomes an observation-based grammar. That is to say, intuition-based grammar which is based on a set of structure rules contradicts grammars which occur in a corpus- a large collection of different types of spoken and written discourses produced in real-life situations (Aarts, 1991). For example, when second language learners speak with native speakers, they notice that their utterances which are based on their grammatical knowledge from grammar classes are not appropriate (Aarts, 1991). Therefore, if we keep observing a corpus, we can find a grammar which accounts for acceptable sentences.

Furthermore, in order to explain the relationship between an intuition-based and an observation-based grammar, Aarts (1991) describes three types of sentence. First, a sentence is grammatical and acceptable, which appears very frequently in corpus data. Second, a sentence is grammatical but unacceptable. Lastly, a sentence is ungrammatical but acceptable. Aarts (1991) focuses on the last type which occurs when a writer knows the standard rule and makes constructions ungrammatical for a particular purpose. In this case, Aarts (1991) seems to argue that the initial process of every construction must begins with the basis of a standard grammar even if the construction is going to be changed for being used appropriately in real situations. Therefore, Aarts (1991) concludes that an intuition and observation based grammar are not different, but complementary.

In addition, Odlin (1994) supports Aarts' view. He insists that performance, similar with the intuition-based grammar, and competence, regarded as the observation-based grammar, can interrelate in various ways. In fact, Odlin (1994) argues that this has created many specific achievements in grammar. However, Odlin and Aarts have a slightly different view of the distinction of the two grammars.

1.2 Prescriptive grammar and Descriptive grammar

With the change of a language, some rules of grammar change or disappear in the standard area, such as thou and you. However, Odlin (1994) states that some grammarians disapprove these changes and emphasise the rules because the grammar as prescription has maintained the standardisation to decide whether a form is correct or incorrect. This is very important to us, as teachers, because if this standardization disappeared in language use, there would be serious deviation from proper communication in society (Odlin, 1994), especially in teaching and learning a second language. The prevention of a deviation by teaching a prescriptive grammar, Odlin (1994) argues, would make learners produce better or more intelligible discourses.

On the other hand, Odlin (1994) describes that many patterns and structures are considered in descriptive grammar, and that descriptive grammarians deal with non-standard language areas such as dialects. Odlin (1994), moreover, claims that this detailed perspective seems to help second language learners tackle problems associated with their target language. In fact, these problems might be often posed by ESL text books which seem to make sense to only native speakers (Odlin, 1994). Therefore, the descriptive grammar can allow teachers to acknowledge the different ways of language acquisition between non-native speakers and native speakers, and thus, if teachers teach the descriptive grammar to learners, the learners can reduce or solve the problems related to the gap between standard and nonstandard grammar (Odlin, 1994). Overall, Odlin has a tendency to keep a balance on the relationship between standard and nonstandard grammar, whereas Aarts (1991) emphasises the standard rules of grammar.

1.3 Communicative grammar

In the consideration of a more practical use of grammar, Leech (1994) emphasises grammar for communication skills in order to explain the relationship between standard and nonstandard grammar, while Odlin (1994) maintains a balanced relationship between them. Leech (1994) presents that communicative grammar has a goal to formulate the relationship between the standard, referred to syntax and morphology and nonstandard, referred to semantic and pragmatic. This is similar with the statement from Aarts (1991) which an intuition-based grammar is not different from an observation-based grammar but complementary.

However, Leech (1994) points out the difficulty learners encounter that the grammar, acquired by the explicit learning from grammar books, contradicts itself in a variety of actual texts in corpus data. Therefore, he tends to emphasise that communicative skill should be more dealt more with in linguistics and in the classroom than it was.

1. 4 Spoken grammar and written grammar

Written grammar, as mentioned above, has been regarded as a standard grammar. Therefore, if we identify the specific features of spoken grammars, the identification can allow us to realise the differences between standard and non-standard grammar.

McCarthy and Carter (2002) attempt to distinguish the grammar of spoken and written English by their samples. The samples consisted of the several examples including the ungrammatical utterances but frequently produced in real situations. From these samples, McCarthy and Carter (2002) argue that "grammar is 'not in there' (internal, in the grammarian's or informant's head); rather it is 'out there' (external, recorded as used, and preferably supported by widespread occurrences across a number of speakers)" (p. 55). In other words, when the products of written grammar are compared to spoken corpus data, they are not identical with each other.

In addition, McCarthy and Carter (2002) employ two terms: deterministic grammar and probabilistic grammar. They described that deterministic grammar concentrates on correctness, while probabilistic grammar focuses on possibility of language. The main point of their distinction between them is that unlike the deterministic grammar, probabilistic or spoken grammar allows speakers to express their interpersonal feeling by a variety of choices of grammar forms (McCarthy and Carter, 2002).

However, McCarthy and Carter (2002) also point out that we should consider a balance between written and spoken discourses in corpus data. Well written utterances might not be included in a corpus because a corpus deals with more frequent utterances. In other words, the utterances can be ignored because people do not often produce such discourses. Therefore, McCarthy and Carter (2002) allow us to realise that even some spoken grammars are not different from written grammars.

Overall, McCarthy and Carter (2002) declare that even though spoken discourses are in conflict with written discourses and conventional grammar, we should not just ignore spoken grammar, but pay attention to it because it plays a very important role in producing written and spoken discourses. However, McCarthy and Carter (2002) suggest that learners should not simply adopt the spoken grammar because of its flexibility, but be explicitly instructed that some patterns occurring in spoken discourses are very commonly used in their target language and do not have no problems to be used.

1.5 The way to reduce uncertainties about grammar

No one can deny the fact that although it has long been taught to language learners by in an explicit way, a standard grammar has not contributed to the growth of speaking skill in an effective way (McCarthy and Carter, 2002). Therefore, the determination of whether or not a sentence in a spoken corpus is correct becomes very important to second language teachers because they should tackle uncertainties about grammar (Leech, 1994). The uncertainties are often caused by ungrammatical and acceptable constructions.

In order to reduce the uncertainties, Aarts (1991) claims that if certain constructions are frequently used and accepted by a large number of language users, such structures can be included in the grammar. Similarly, McCarthy and Carter (2002) argue that an aspect should be incorporated in the grammar, if there are enough examples of it. Therefore, if teachers teach these acceptable grammars, learners can reduce the uncertainties about grammar.

Another way to reduce the uncertainties is to teach grammar rules to students at the earliest level although there are a great number of variations in grammar (Leech, 1994). In addition, McCarthy and Cater (2002) claim that, resulting from the awareness of the differences between spoken and written grammar or from the wide range and change of grammar forms in a spoken area, the flexibilities, such as options of a relevant form, should be accepted in pedagogical grammar. These flexibilities allow learners not only to reduce the uncertainties, but to develop the ability to communicate and to create their interpersonal reaction depending on real circumstances (McCarthy and Carter, 2002). If it can be said that modern linguistics tend to concentrate more on the production of speaking and writing, McCarthy and Carter's attitude toward grammar is more radical and makes more progress than Aarts and Leech.

Corpus data

No one would deny the contribution of the corpus data to grammar studies which has allowed linguists and language teachers to analyse a tremendous amount data of natural occurrences in discourse (Conrad, 2000). In addition, with emphasis on communication in modern society, a pragmatic grammar has been significantly considered in the content of language courses (Widdowson, 2003). From these perspectives, we may infer that corpus data plays a significant role in teaching grammar in language classes. However, even the best method or theory cannot be free from criticism, and thus corpora have been criticised.

2.1 The importance and weaknesses of corpus data

At least once, all language teachers have answered 'it depends…' to their students' questions, which reflects their uncertain intuition about the language (Widdowson, 2003). However, a corpus has helped the teachers not rely on the uncertainty because it can enable them to use a great amount of evidence of actual occurrences produced by its native users (Conrad, 2000). Widdowson (2003) states that corpus data provides teachers with chances to reduce the gap between prescriptive and descriptive grammar, and thus, they can instruct their students to produce an accurate and acceptable discourse in a variety of circumstances.

Despite the importance of a corpus, we need to know its weaknesses. Thus, it is inevitable to enquiry whether or not learners can generate the applicable model from corpora. Gavioli and Aston (2001) suggest that corpora are valuable for learners as a way of helping them to create their own discourse models, rather than as a way of encouraging them to reproduce the constructions from corpora.

In addition, the leaner's standpoint in corpus data should not be a participant, but an observer because the learners have not produced the occurrences which they have, indeed, never experienced, or will not experience in the future (Widdowson, 2003). In other words, corpus data, as described above, helps the observers to produce appropriate discourses in different situations with their own models, not to mimic the data as a participant (Widdowson, 2003). However, Gavioli and Aston (2001) argue that the "observer as well as participant roles can allow learning: observation allows strategies of interaction to be noticed, while participation allows such strategies to be tested" (p. 241).

Widdowson (2003) argues that there are three limitations which corpus data often does not cover. First, in general, language users have more knowledge of their language, such as old fashioned vocabularies, than they can realise. In fact, because the nature of corpus data is not to include words which are not currently being used, corpus linguistics may not be used to measure how large a person's vocabulary may be. In addition, the current corpora do not "either reflect prototypicality, the relative cognitive salience of words and structures" (Widdowson, 2003: 86). Lastly, due to another nature of corpus data which is just to include attested occurrences and normal usages, language users cannot use the language in an innovative way (Widdowson, 2003).

It is very significant for teachers and learners to be aware of the importance and weaknesses of corpus data because corpus data can provide effective and efficient help for them when they can use it in an appropriate way.

2.2 Corpus data in a pedagogical grammar: Teachers

As it was often pointed out in the preceding section, corpus linguistics changed and contributed to pedagogical grammar. With the advantage of availability to access to the practical data, a corpus has contributed to teacher-training programmes (Conrad, 2000). Therefore, teachers can be trained by the corpus based programme, and then lead to a great impact on pedagogy, as Conrad (2000) explains. Furthermore, the corpus based programme in pedagogical grammar enables teachers not only to identify differences between many grammar books, but to develop their own materials. (Conrad, 2000)

On the other hand, when teachers apply the corpus based materials in their classes, they need to be careful because of the two reasons. First, although certain constructions are not produced frequently, such constructions should not be ignored because they can be useful under certain circumstances (Conrad, 2000). Moreover, even frequent construction is hard to be generalised in real discourses (Aarts, 1991). As a consequence, corpus findings do not provide what we have to teach, but help us to instruct learners in more accurate contents (Gavioli and Aston, 2001).

2.3 Corpus data in a pedagogical grammar: learners

In fact, the main purpose of corpus is to give learners the chance to be involved in real discourse, which can improve their language skills (Gavioli and Aston, 2001). The process of the improvement, Gavioli and Aston (2001) imply, begins with the doubt whether or not a writer's construction makes sense. After that, the writer can be aware of a variety of patterns or structures from corpora. Therefore, the writer tries to use his findings in his own discourses. Although they may not meet his intention at the beginning, he can use more findings to establish the proper model he desires.

Odlin (1994) points out that language learners should have an attitude to strive to independently examine and analyse their target language because instruction cannot cover all the problems they encounter in real situations. Therefore, learners should need teachers' help and corpus data in order to tackle those problems. First, a teacher might instruct their students to notice the components of the system of grammar as well as to provide them with the way of exploring the system (Odlin, 1994). Moreover, as pointed out in the preceding parts, corpus data can be helpful to avoid the problems through the establishment of their own model of discourse (Gavioli and Aston, 2003).


The several theories analysed above go some way towards clarifying the relationship between standard and nonstandard grammars, and thus, could help us to decide what contents should be included in pedagogical grammar. However, when we teach grammar in class, we should remember that there is no perfect theory or standardisation. In this sense, corpus linguistics has made valuable contribution to pedagogy. However, the data needs to be handled carefully. In other words, even if corpus data shows a certain construction is frequent, we should consider it is very hard to generalise it, as shown above. Although a corpus can be used as an effective approach, it can also sometimes lead us to create confusion based on an incorrect collection from corpus data. Therefore, recognising that the previous theories or information about grammar could be incorrect or inaccurate in our teaching, we need to research or invent our own materials and appropriate methods by very accurate analysis of corpus data, rather than just relying on the data itself.