The class of adjectives is one of the more problematic ones in English grammar. In order to see what kind of problems and questions we are dealing with we need to address the following two questions: what adjectives are, and how we can recognize them. The most common answer to this question, provided by most grammars, is ‘adjectives modify nouns’ (Pence, Hodges 51, Johnson 31). This answer (or definition) may seem obvious enough but, as Roberts notes, this definition will lead to some serious difficulties.
Nouns form their plurals with -s, adjectives modify nouns, but to what class does stone belong in the stone wall? Clearly it is an adjective, it modifies or qualifies wall, but stone also has a plural in stones. Roberts says, “If we define some parts of speech on the basis of form [nouns forming a plural with -s, MA] and others on the basis of syntax [adjectives modify nouns, MA], we will necessarily find many words which fit two definitions at once. ” Not a very satisfying conclusion if you like your parts of speech in neat boxes.
Fries offers a different definition. He gets rid of the word adjective and adopts Class 3 words (Class 1 words are (what we call) nouns, Class 2 words are (what we call) verbs). His definition is that a word belongs to Class 3 if it fit into the following formula: The [Class 3 word] [noun] is [Class 3 word]. Good is a Class 3 word because the good student is good. Fries continues by giving a set of three ‘important formal contrasts’ for Class 3 words. Neither his definition nor his formal contrasts however offer a solution to the problem of the stone wall.
Fitting stone wall into his formula gives the stone wall is stone, which is, if not questionable, at least not very obvious. And what about the wooden chair? Is that chair wooden, or wood? His ‘formal contrasts’ don’t provide us with an answer either. Interestingly enough, Palmer doesn’t offer us a definition, he merely says, “another major class [he is talking about Some Traditional Concepts, MA] is the adjective, with two main functions, attributive and predicative, as illustrated by the little boy and the boy is little(p. 9). ”
But what about the difference between the first Frenchman and the Frenchman was first? Another ‘what are they’ problem, not addressed by every author under investigation, is what to do with words like my, any, you(r), the, a, each. They seem to be able to make some claim to the status of adjective (as modifying a noun, or identifying it) but seem to have different characteristics and uses. Fries and Johnson don’t treat these words at all when talking about adjectives.
Pence and Roberts label them as definitive adjectives and limiting adjectives respectively, and Palmer thinks it’s best to regard these words as belonging to a different class, mainly on the ground that these words are (almost) never used predicatively. These words (articles, possessive pronouns, demonstratives and words like all, some, neither) “are treated today as ‘determinatives’ or ‘determiners’ (p. 59),” and he makes a similar case for ordinals. Unfortunately he gives us no answer for the difference in meaning between for instance the right girl and the girl was right.
For Palmer it is therefore no longer possible to say that adjectives, and only adjectives, modify nouns. As stated before, some authors don’t mention these words at all and in doing so ‘disqualify’ their definition of adjectives as noun-modifiers (esp. Hodges and Johnson), for it is clear that they have some modifying function. Giving a definiton has obvious difficulties, as we have seen. Maybe we can figure out what adjectives are by a formal way: what they look like. Most authors seem to agree on some of the most important features of the form of adjectives.
These are usually: a. forming of comparative and superlative with -er and -est (or more and most); b. the numerous derivational suffixes (-ish, -ly, -ine, -ic, -al, -able, -y, -if, -ary, to name but a few); and c. the ability to pair with adverbs like very and extremely (Fries, Hodges, Johnson, Roberts). The comparative and superlative forms however are not used in adjectives alone: adverbs can have similar forms (Palmer p. 63). This means that the formal characteristic of the existence of a comparative and superlative form does not mean that a certain word is an adjective or not.
A similar point can be made about (some of the) suffixes: both Hodges (p. 51) and Johnson (p. 32) note that the ending -ly after an adjective creates an adverb, only Hodges notes that the same ending added to a noun creates an adjective. This means of course that a word ending with suffix -y is not necessarily an adjective. Also, not every word that is considered an adjective can be modified by very (and similar adverbs). The boy is asleep is correct, but a boy cannot be very asleep (”except perhaps in joking form” (Palmer, p. 62)). There seems to be no formal characteristic that can decide whether a word is an adjective or not.
Taking into account all these considerations we may conclude that there is no real answer to the questions raised. Is that a bad thing? Maybe a slightly different approach can help us out here. The question ‘Which words belong to the class of adjectives and which not? ’ may be very interesting, but we have seen that both a definitional approach and a ‘formal’ approach raise more problems than they solve. Maybe a derivational approach can be more fruitful in some cases (if less fulfilling for those who like to think in terms of strict classes).
Palmer (p. 4) suggests that ‘adjectives that are nouns too’ may be derived from the noun, and although nouns like steel, cotton and stone have no difference in form when used as an adjective, some nouns like wood and wool do, with their derived adjectives wooden and woollen. This does still not give us ‘instant recognition’ though, and the nouns that have the same form as their derived adjectives still leave us with an awkward feeling when Fries’s formula is appplied to them. Thinking about adjectives in terms of ‘function’ is just as unfulfilling for ‘class- thinkers’, but does make some sense.
Regarding the problems concerning the use of ‘nouns as adjectives’ Palmer remarks that “adjectival function…is a function of all nouns (p. 64),” and leaves it at that. The same can be said about his class of ‘determiners’. The very term ‘determiners’ suggests that this class-decision is made according to function. Words that ‘determine’ have an adjectival function, although their use is quite different from those that can be regarded as ‘core-adjectives’. What’s wrong with that? Well, it seems a little confusing to add another class-distinction, namely ‘use’, alongside with ‘function’. What can we say in conclusion? Not a lot, I’m afraid.
There are words that modify nouns (function), a lot of them display certain characteristics of form, and a lot of them can be used predicatively and attributively (use), but not all words we call or consider to be adjectives have the same function, form or use. That’s very unfortunate, but that’s the way it is. It seems strange that in speaking and writing we don’t seem to have that many problems with using and forming adjectives, but perhaps this is just another example of the flexibility of human beings in dealing with a system that, when described in terms of rules and classes, appears to be made up of inconsistencies and arbitrary rules.