Teachers of English to speakers of Japanese (and many other Asian languages) will recognise the problems their learners have with usage of the English definite article. In 99 cases out 100 these problems do not affect comprehension and many L2 speakers of English perform competently without realising they have a ‘problem’ in this area. There are, however, many types of learner who do need to produce patterns of English that approach native-speaker forms. What exactly is the nature of their problem with the article and is it possible to solve it?

During a recent two-year period of teaching a course of English for academic purposes to Japanese postgraduate students, I examined the problems students had with article usage in English. The target population had a background of a long and intensive education in various specialisms, culminating in university studies (not English except in two cases). They were thus of proven academic ability but, because all but two had not spent time outside Japan except on holiday, most were also of medium-to-low ability in both written and spoken English. An analysis of their problems in the written language showed that articles, prepositions and what might be termed ‘general phrasing’ (i.e. *That becomes to the later stage negotiation) caused, in that order, the greatest amount of non-standard production.

There are obvious explanations for the origin of definite article problems that derive from the structure of the Japanese language, but other factors also impinge. The rules for the use of the article in English are so complex that standard usage will best be acquired by prolonged exposure to native speakers. This is unfortunately impossible for most Japanese because they take so few holidays to go abroad. When they do travel, it is generally in groups of other Japanese, so they rarely hear or speak English. Certainly more young Japanese are travelling independently, but it is still the norm for white collar workers to be recruited into firms straight out of university. This denies them the chance to take a year or so to travel in the manner of many European, North American or antipodean young people. Another problem is the reluctance of Japanese learners to involve themselves in learning language with the same lack of inhibition that learners from other cultures manage to achieve. In many ways Japanese culture is still a closed one which inhibits many learners from embracing the language of a foreign culture. This inhibition stems partly from a fear of taking risks before others (output +1), which is an essential part of extending linguistic range.

In linguistic terms the most obvious explanation for the problems with its use in English is the absence of the article in Japanese. There is the additional problem of the concept of countability for Japanese speakers, whose language does not distinguish between singular and plural noun forms except in a few specific cases. As a result of these differences the teacher not only has to introduce learners to:

Little wonder then, that the Japanese learner seeks refuge in a position of :"When in doubt, leave it out". The evidence for this is shown in the statistics drawn from the sample available to this study. These figures are all taken from written work and it is interesting to note that in a similar study of the spoken English of the same groups article errors were third in frequency (after phrases/vocabulary and omission of final [-s] and [-d]). This is probably due to the tendency of the group members to over-correct written work whereas in conversation their production was more natural and they did not have time to over-correct. Nevertheless, one of the problems in oral production was a tendency to repeat phrases, partly to create some "thinking time", partly to correct or over-correct an utterance.

Errors that the students made in phrases involving the articles were collected from the essays produced during one academic year, and the distribution was found to be the following:

Table 1 Percentages of Errors in Written Usage of Articles

* This category consisted mainly of the substitution of definite for indefinite article, or vice versa.

The distribution clearly shows that omission of the articles is the main problem area for Japanese speakers, with the definite article the main source of the trouble. This is familiar to anyone with experience of teaching Japanese learners, but we must now ask if it is possible to analyse the data further in the hope of producing useful teaching techniques or materials. The first breakdown of the types of article problems follows the division given by Quirk and Greenbaum in their "University Grammar of English", with numbers added to the classifications in order to aid coding.

Table 2 Quirk and Greenbaum’s Schema of Article Distribution

	Definite			Indefinite
Specific	1 the tiger		7 a tiger
	2 the tigers	6 the ink			8 (some) tigers	10 (some) ink
	3 the tiger
Generic	4 a tiger		9 ink
	5 tigers

The first conclusion to be drawn when actually coding the errors was the difficulty, even for a native speaker, of deciding the difference between specific and generic, and count and non-count. Some of the distinctions become blurred in real life; nonetheless, a pattern does emerge:

Table 3 Distribution of Article Error Related to Quirk and Greenbaum’s

Schema

From this breakdown we see that the greatest concentration of errors occurs in definite article omission in the area of specific reference to both count and non-count nouns. A hypothetical explanation has been suggested above that omission rather than addition is the main cause of error because of the lack of articles in Japanese. There may also be a confusion about the concept of plurality, but we must also look at the contrast of "specific/generic" around which differentiation turns. It appears that this is a distinction that is clear to the explainer, but not always so to the learner.

If we take the phrase He’s putting wood on the fire we have a straightforward case of:

Now let us assume that a student has written *He’s putting the wood on fire. The teacher tries to explain that wood in this case should not have the article because it has not been referred to before, while fire should be the fire because it is "part of the cultural situation" (Quirk & Greenbaum, 1972:73). The learner is justified in asking how s/he is to learn what constitutes such a "cultural situation". In societies which burn wood rather than oil or coal, the student may feel that wood is a part of the cultural situation.

English in real life is never uncomplicated and an additional problem here is the semantic changes that addition and omission bring about. Firstly the learner may well have partially acquired the phrase on fire as in to set something on fire, in his/her interlanguage. If this semi-acquisition caused the original problem the teacher must explain it from a different point of view. The second change caused by the article, i.e. the wood = an area of countryside covered with trees, is probably not the cause of the problem here because of the context in which the phrase appears. Nevertheless, the learner who may have recently acquired the phrase from other reading may simply write the wood because the article+noun group has been remembered as a block.

We find specificity referred to again in a very thorough analysis of the usage of English articles in a book published in Japan (Brender, 1989) which has clearly been inspired by the author’s struggles with this problem. Using Brender’s concept of specificity we generate examples such as:

SPECIFICITY

And yet we can manipulate these examples; the sentence:

He looked at A clock on THE wall

would mean the same as the original example in most contexts, and if we turn the clock into a plural, answering the question: Where is John?

and

we have, for all practical purposes the same meaning. According to the table of usage given above the teacher may well be able to explain the nuance of difference between the idea that John has probably gone to the shop with an intention of looking for a clock, as opposed to the second sentence in which he is in the shop and has become interested in the clocks he has found there. The problem, however, is one of transfer - how can we expect the learner to internalise this distinction while still remembering that in

and

the article (either definite or indefinite) is not optional?

There are more problems if we go back to THE tallest boy IN THE class and then ask Is Hiroshi IN CLASS today? If questioned, we say that the article is optional in this set phrase, although its omission is not optional in Is Hiroshi IN SCHOOL/COLLEGE today? And, of course, my wife may be IN THE house, but she is AT home because she was brought up to believe that a woman’s place is IN THE home. This brings us to the question of prepositional and adverbial phrases that may or may not use the article and which operate independently of the grid shown above.

Transfer is the crux of the problem: there is a set of rules, but it is an extremely complex one, full of overlapping explanations. Native speakers have trouble explaining this area of grammar to each other and yet the teacher has to find a method of explaining the rules to learners who, as we have seen, have to recognise concepts that do not occur in their mother tongue. Once this recognition has been taught, the learners must be helped to internalise as many as possible of the routes of the mental algorithm that native speakers have been developing since birth.

Grammar reference books and course textbooks imply that it is simple to teach the notion of specificity to our learners so that they can train themselves to identify an article-bearing noun phrase in time to allocate the required article. In practice the English language is not so cut and dried and specificity is an elastic concept. Another elastic concept is that of the count/non-count distinction. One of the errors noted in the sample was: *The Japanese industry is very competitive because.... (This example reminds us that most so-called non-count nouns in English can also be countable; Quirk and Greenbaum use ink as their example of a non-count noun, but we talk of coloured inks.)

In the example quoted above, which occurs frequently, the elastic concepts of specificity and countability are interrelated. The student who has been corrected for writing the phrase may claim that Japanese industry is a specific sector of world industry in general. It is explained to him that Japanese industry, even specific southern Japanese industry, are still general areas of industry (non-count) within which specific industries (countable) operate. All is well until the student re-phrases his sentence to read *Industry of Japan is very competitive because.... and goes on to write about *Global automobile industry is becoming more centralised because.... In this case it appears that we are not dealing with countability or degrees of specificity, but with the nature of the description governing the head word.

Leech and Svartvik (1975:54-5) attempt to formulate a rule to attend to the Industry of Japan question by stating: "when mass nouns and plural nouns are postmodified, especially by an of-phrase, the definite article normally has to be present". This gives us:

Japanese industry the industry of Japan

With concrete mass nouns and plurals, the definite article is optional:

tropical birds (the) birds of the tropics

Let us now test the rule with the phrases: industrial benefits and the benefits of industry.

We find here that the transformation of the phrase has altered its meaning. An additional confusion comes into play when we realise that the phrase industrial benefits also carries a secondary meaning that relating to ideas such as health insurance schemes, a signification that does not transfer to the of-phrase.

The concept of the ‘backward reference’ is often cited as a criterion signifying specificity. Textbooks explain this concept in various ways, but if we are looking for a complete explanation of the rule and go to a standard work of reference we find the following:

I bought a radio and The radio is Japanese.
(Quirk et al., 1972:154)

The problem for the course book author and the teacher is to make these generalisations transferable to help the learner internalise, by means of examples and exercises, patterns will be that comprise the great variety of conditions in the framework of article usage. Let us refer to the same source in order to enlighten a learner who has written the phrase *The Constitution of the United States is founded upon liberty’s love:

A certain amount of sympathy may be afforded to the teacher trying to turn this into a response that is quickly and easily understood by a 16-year-old in a provincial town in Poland or Thailand, especially if the teacher is not a native speaker.

The first step in finding a solution to this problem is to make the learners aware of it. The Japanese learner will probably have been taught in secondary school by another native speaker of Japanese. That teacher’s English will contain a certain number of these errors, which pass on to the learners. There is also the phenomenon of the learner who simply hears what s/he expects to hear. If the mother tongue does not contain articles it is very easy not to register their use in another language, especially when the rules are as complex as we have found them to be in English. Those who have taught Japanese students will have noted their fear of making the ‘imperfect statement’. In spoken English they have a tendency to repeat phrases until a suitable version has been arrived at. Writing is usually done in pencil so that perceived imperfections may be erased before a draft is handed in. Given this tendency, which is ingrained in secondary school, it is not surprising that the Japanese learner will shy away from jumping into the bed of nettles that is the English article system.

The second step is to point out that, when in doubt, statistically speaking it is probably going to be better to include the article rather than omit it. As we see in Table 1, 81% of the errors made by this sample were in omitting either the definite or the indefinite article.

Thirdly, once the existence of the problem has been faced, there comes an active phase in which the learner is helped to become self-sufficient. We may subdivide this process into two parts: learning from oneself and learning from elsewhere. The learning from oneself comes from a system I developed with the student groups described above. Errors were collected and classified from the essays students wrote during the academic year. In the case of the article problems, learners began to amass a corpus of phrases that they had used and to which they could refer in subsequent essays. In theory at least, students who had been made aware of article problems would recognise constructions which had previously needed correction (such as *Industry of Japan) and check a similar construction in new writing against their error correction list.

The learning from elsewhere consisted partly of a subconscious process of developing monitoring skills from the large amount of reading and listening to native speaker lecturers which the course involved. It came also from using a computer-based concordancing programme. Unfortunately the latter method was not used as much as one would have wished because of hardware problems, but there was a noticeable decrease in the frequency of article errors as the academic year progressed. In courses where such long-term exposure to native speakers is unavailable, students will benefit from English-language videotapes for individual viewing and comprehension exercises, followed by in-class discussion.

In one sense the omission of the English article is not a problem because there are very few occasions in which communication is seriously affected by it. On the other hand, learners who wish to produce English that approximates closely to native-speaker level are cheated if their instructors do not equip them to master this particularly difficult element of the language. The root of the problem lies in the contrast between the two languages involved. The difficulty is compounded by the fact that the rules that native speakers have imposed upon article usage contain so many exceptions as to be almost useless as a basis of explanation to non-native speakers. The teacher needs to develop methods of transferring patterns of usage to the learner which complement the monitoring function of indicators such as Quirk and Greenbaum’s grid and the algorithm Brender gives at the beginning of his book

It is significant that in spoken English the students in these samples tended to make relatively fewer errors in this area. This indicates perhaps that they have acquired (rather than learned) an ability to use the articles reasonably well, but can only demonstrate this ability when they do not think too hard about correctness, which they do when writing. This need for correctness comes from the Japanese cultural and educational background and while it is very commendable in building motor cars and hi-fi systems, it does inhibit linguistic fluency.

Brender, Alan S. 1989. Three Little Words A, An, The. Tokyo, McGraw-Hill (Japan).

Leech, G. & 1975. A Communicative Grammar of English. Harlow, Svartvik, J. Longman,.
Quirk, R. & 1973. A University Grammar of English. London, Longman.
Greenbaum, S.

Quirk, R. et al. 1972. A Grammar of Contemporary English. London, Longman