| No 7, Vol. 3, 1997 |
 |
Give grammar some content
MA in Danish, French and Danish as a foreign language and a second language Institute of Nordic Philology, University of Copenhagen |
When is grammar a good means to the desired end; where does grammar teaching help, and how are we to organise teaching so that it can help pupils and course participants progress towards communicative competence? For this is surely when our main problem lies. With the main aim of demonstrating how essential it is for the teaching of grammar and systems not to take place alongside content teaching but to be an integral part of it, I intend in this article to present certain principles that would seem to dictate the way we acquire language functions and language systems - principles that can help us signpost our teaching.
One of the central conclusions I have been able to arrive at from my PhD research (Lund, in course of publication) is that the only way we can reach the point when we can predict how easy or how difficult it will be for our pupils to acquire a grammatical structure is by looking at the extent to which the grammatical structure is part of a meaning context that is transparent. If, for example, we take that well-known feature of Danish, inversion, it turns out that learners have much greater difficulty in acquiring inversion in statements than in acquiring precisely the same structure in questions? Why should that be?
If we look at inversion in the two contexts from a functional point of view, we can see that inversion has a crucial meaning-conveying function (semantic and pragmatic) in questions. Apart from a different intonation pattern, inversion is the only feature that distinguishes questions from statements:
In other words, inversion can be of vital importance if one is to convey one's message to another person.
In statements, on the other hand, inversion is without significance at first glance. It makes no difference whether one says:
The first utterance is admittedly incorrect according to Danish norms, but the missing inversion can never give rise to misunderstandings. That which apparently plays a crucial role in the course of acquisition is: In what way is the feature 'wrong'? In questions, we are dealing with a 'wrong' content. In statements, it is basically only a question of 'wrong' grammar.
This would imply that a structure - here, inversion - that is only justified in terms of grammar is difficult to acquire, whereas the learner quickly finds out how to solve this sequence principle if the same structure has a clear semantic, pragmatic function (See also Gisela Håkansson's article in this number).
Within acquisition of negation, investigations have shown that apparently inexplicable systematics can occur in the individual learner at the same stage of development. The learner places the negation correctly in connection with auxiliary/modal verbs, but incorrectly with main verbs:
How can one explain the fact that learners manage the context that is grammatically speaking the more complex - where there is both a modal and a main verb - earlier than the less complex, where there is only one verb? The fact of the matter is that learners cannot. For they have not actually solved the problem of position at all yet. As long as learners place the negation before the finite main verbal, they cannot be said to have discovered that grammatical finiteness governs position.
A much more likely explanation is that learners have formed a semantic hypothesis, according to which the negation is to be placed in front of the content-laden verb, no matter whether this verb is finite or an infinitive. In other words, the correct sequence 'Peter vil ikke spise kartofler' is right for the wrong reasons.
The same feature is found in learners' subordinate clauses, where the first correct position is at the finite main verbal. Not until later in language development is the position before a modal/auxiliary verb acquired. At the same stage of development we thus meet:
From a purely grammatical point of view it looks very strange. In main clauses the learner first learns the more complex context, but in subordinate clauses the opposite is the case! Confusion. Yes, but only if we look at things from a purely structuralist finiteness angle. Analysing from a semantic point of view we can in fact conclude that the learner in all four instances sticks to one hypothesis and one only - that the negation is to be put in front of the content-conveying main verb:
The analysis of both inversion and negation demonstrate how the early-stage learner, on the basis of which semantic and/or pragmatic contexts a given structure is part of, forms hypotheses which build on a one-to-one principle. For each unit of content one structure or one system is sought for. For questions this leads to (WH-) + Verbal + Subject; for statements to (X) + Subject + Verbal. For the position of negation to one fixed position in front of the content-conveying verb, whether it is finite or infinite.
On the basis of this systematics of acquisition, it is my overall hypothesis that the acquisition of a new language is driven forwards by the learner's seeking for the systems which make it possible to express content. The learner does not ask: When does one use inversion in Danish? No, the learner asks: How do I ask about something? How do I tell someone something? What word orders help me to what communicative aims?
The learner finds it difficult to acquire inversion in statements, and in acquiring negation some learners have trouble with these two contexts: negation after a final main verbal in main clauses, and negation before a finite modal/auxiliary verbal in subordinate clauses. Why, though, are these contexts so difficult? It is because they are exclusively based on grammatical systems. The systems in language that are not 'backed up' by semantically, pragmatically transparent functions, and which therefore have to be understood solely via a more abstract form of grammatical insight are those which are difficult to acquire. For inversion in declarative sentences it is necessary for learners to have grasped the syntax and to understand the sentence member principle. For negation, that they have understood the principle of finiteness. In the other negation contexts learners can make do with a semantic hypothesis, since here grammatically and semantically based position coincide, c.f. the above examples.
This makes it possible to conclude that the more semantically, pragamatically clear and transparent a linguistic structure is, the fewer difficulties the learner will have in acquiring it. The more semantically, pragmatically intransparent a structure is, the more difficult it will prove for the learner. The more learners have to tackle the task of solving problems solely with the aid of grammar or systematic insight, the more problems they will have. This leads to the following continuum of difficulty:
This continuum of difficulty helps to explain a great deal of the difficulties our pupils have in learning the systematic sections of a new language. It applies to all systems in the language - also the pragmatic, such as the modal adverbs, which have always seemed to defy systematisation.
Let us see what this understanding implies for one of the areas which takes up so much teaching time in both foreign languages and Danish as a second language: morphology.
It is, among other things, because the morphemes are difficult to hear - not least in Danish, where they are always unstressed and have very little coding material (-Ø, -t and -e). but why is it not so difficult to learn noun endings, where the same applies? It is not because the adjectival morphemes are redundant, people normally say - the noun already shows that we are dealing with a plural: gule cykler [= yellow cycles]. But quite often the -r on the noun is also redundant, because a word indicating amount has already indicated a plural (to, mange, flere gule cykler) [= two, many, several yellow cycles]. In spite of this, pupils do not have anything like as much difficulty in learning the plural of nouns as they do with precisely the same grammatical category of adjectives. Why ever not?
Let us look at the problem from a semantic point of view. When the language is used, the various word classes have various functions. We use nouns to talk about objects, feelings, persons; adjectives are then used to describe them. We use verbs to talk about actions and states. As far as nouns are concerned, they can refer to objects in reality or in the imagination. In other words, a morpheme such as the plural morpheme has a special affinity to nouns that is easy to comprehend. That which the adjective refers to, however, is not countable - one cannot count the colour 'yellow', nor the quality 'poisonous'. To acquire the plural morpheme of adjectives the learner therefore has, at some level or other, to have understood the purely grammatical control system that exists from the noun to the attributive or predicative adjective. Which is an intra-language, relatively abstract grammatical insight. . So it is easier to arrive at the semantic, pragmatic function that lies in the plural when it is linked to nouns and, for example, personal pronouns than when the plural is linked to adjectives and, for example, possessive pronouns.
Verbs are used to talk about time and time relations. It is therefore not nearly as difficult for learners to link meaning to tense morphemes as it would have been if we declined our verbs according to person and number - bearing the English 3rd person singular in mind. Tense morphemes have an affinity to verbs. Number and person do not.
Some people would perhaps claim that adjectives as such are a difficult word class. An understanding of affinity will perhaps predict that comparison is relatively easy to acquire, since comparing is precisely that which one does with adjectives and not with other word classes. Comparison can thus be said to have a particular affinity to adjectives. That will make it much easier for learners to link meaning to comparison morphemes than to number morphemes.
| Easy < --------------------------------------------------------------------------------- > Difficult | |||||
| number | two/many + noun (analytic) | > | noun + e/er (synthetic) | > | adjective +e (synthetic) |
| verbs | yesterday + verb (analytic) | > | verb + ede/te (synthetic) | > | verb + person/number (synthetic) |
| adjektives | more/most + adj. (analytic) | > | adj. + (e)re/(e)st (synthetic) | > | adj. + number (synthetic) |
| inversion/V2: question | > | deklaratives | |||
Naturally, there is no doubt at all that the moment are at a morphemic level, we have already climbed a few rungs of the grammatical abstraction ladder. All things being equal, it is easier to link meaning to two, many and yesterday than to link meaning to -e/er and -ede/te. But since these morphemes have an affinity to their respective word classes and therefore point outside language, pupils can, without too many abstraction costs, use an equals sign and form analogies between two and -e/er, or between yesterday and -ede/te.
Each time language points outwards rather than simply inwards, learners have problems. As soon as contact can be made with something outside language itself, things are much easier.
Pupils' language must be gradually grammaticised. No doubt about that. Systems have to be imposed and abstraction is involved. The question is whether or not we can do something to facilitate this process of grammaticalisation.
If what the learner is striving for is to find meaning in linguistic systems, that must be precisely the route we should take in our teaching. We must help pupils to assign meaning to the system-sides of language. This makes the systems concrete and understandable. And this is a basic requirement for acquisition. Not until the learner has linked a more abstract system to a known and less abstract content is he/she in a position to acquire it. Only subsequently will the learner be able to operate within the category which is, of itself, more abstract. Instead of talking about one as a learner becoming better and better at thinking abstractly, I think we ought to say that as a learner one should become better at making abstract categories concrete. That applies to all learning.
If pupils struggle to assign meaning to systems, we must naturally also do our bit to facilitate this process by always ensuring that system work takes place within a semantic, pragmatic universe of meaning.
Bolander, Maria: Nu ja hoppas inte så mycke. Om inversion och placering av negation
och adverb i svenska som andraspråk. in: K. Hyltenstam and E. Lindberg (eds.):
Första Symposiet om Svenska som Andraspråk. Stockholm; Stockholm University, 1988,
pp. 203-214.
Hyltenstam, Kenneth: Implicational patterns in interlanguage syntax variation. in:
Language Learning, Vol. 27, 1977, no. 2, pp. 383-411.
Lund, Karen: Kommunikativ kompetence - hvor står vi? in: Sprogforum, Vol. 2, 1996,
no. 4, pp. 7-19.
Lund, Karen: Rosa - en langsom lørner. Hvordan ser den langsomme lørners sprog ud
og hvorfor? in: Skovholm (ed.): Hvorfor er det så svært at lære dansk? En antologi
om langsom indlæring af dansk som andetsprog. Specialpædagogisk forlag, 1996.
Lund, Karen: Lærer alle dansk på samme måde? En længdeundersøgelse af
voksnes tilegnelse af dansk som andetsprog. Special-Pædagogisk forlag. In course of publication.
Meisel, Jürgen: Strategies of second language acquisition: More than one kind of
simplification. in: R. W. Andersen (ed.): Pidginization and Creolization as Language
Acquisition. Rowley, Mass; Newbury House, 1983, pp. 120-157.
Schmidt, Richard: The role of consciousness in second language learning. in:
Applied Linguistics, 1990, no. 11, pp. 129-158.
Sprogforum's homepage |
Contents of
this number
|
Ordering
Sprogforum
|
