I've been reviewing my grammar lately and I feel a bit stuck in adjectives. Yeah, I know it's not a big deal of a topic, but I think I need some help here because I'm in the language teacher role. I do understand how attributive and predicative adjectives work, yet I can't find an easier way for learners to differentiate between both types - especially if such differentiation doesn't exist in their mother tongues. And yeah, I know I can just explain the whole thing to them, but what if they come across other adjectives they don't know how to use? I just feel a good dictionary might do the trick, that is, with a tagging system that tells adjectives apart according to that syntactic feature: pointing out if they're attributive, predicative or both. The Oxford Learner's Dictionary online sometimes does this -click here and here for examples-, but it seems to me that a more specialised source might be the answer to this dilemma. Well, I'd like to know what your take on this issue is. Suggestions are welcome. Thanks!
I do understand how attributive and predicative adjectives work, yet I can't find an easier way for learners to differentiate between both types - especially if such differentiation doesn't exist in their mother tongues.
I'd like to suggest that maybe a reference grammar might be able to help you here. For instance, the 2002 reference grammar by Huddleston and Pullum et al., The Cambridge Grammar of the English Language (CGEL), discusses this very issue: "Chapter 6 Adjectives and adverbs", part "4 Restricted function for adjectives", on pages 553-562.
CGEL, in "4.1 Attributive-only adjectives",
- Adjectives that do not normally occur except as (heads of) attributive modifiers include:
which includes an example list of 30 adjectives (e.g. drunken, future, mere, putative, umpteenth). In this part, they also discuss "Meaning differences between attributive and non-attributive uses".
They also discuss "Potential differences between attributive-only and ordinary attributive adjectives"; where they discuss the four properties that ascriptive attributive adjectives characteristically have, but most attributive-only adjectives lack one or more of them. The properties are: entailment, subset, modifiability, pro-form.
Then they go on and discuss "Some types of attributive-only adjectives". In this part, they say:
- Attributive-only adjectives are too numerous and semantically heterogeneous to permit a simple and exhaustive classification. Instead we will here illustrate and comment on some of the most important semantic types.
Some of those types are: degree and quantifying attributives, temporal and locational attributives, associative attributives, process-oriented attributives, modal attributives, particularizing attributives, expressive attributives, hypallage: transferred attributives.
CGEL, in "4.2 Never-attributive adjectives", discusses "Adjectives which can occur predicatively or postpositively, but not attributively". These include "Adjectives formed with the 'a' prefix", and provide an example list of 24 adjectives (e.g. afloat, aglitter, alike, averse).
Also, includes "Adjectives with complements". And also includes "A small set of other adjectives".
And then includes "Postpositive-only adjectives", which are described as:
- A handful of adjectives are restricted to postpositive function:
and they provide an example list of 6 adjectives (e.g. flowers galore, the President elect).
Okayee, my fingers are tired. Hopefully I haven't made too many typos. It seems to me that CGEL has a lot of good info in those pages that might help you out (pages 553-562).
In an unsearchable and potentially ephemeral comment to the original posting, Professor Lawler kindly presented the following answer:
Practically any adjective can be used either as an attributive or as a predicate. It’s dependent on the sentence, not the adjective, so a dictionary won’t help.
A predicate adjective is essentially the verb of the clause, only it can’t get inflected for tense, so it has to go around with an auxiliary be that holds the tense: She is/was/may have been tired.
An attributive adjective modifies a noun: the tired woman.
There are a very few adjectives that can’t function as both attributive and predicative — gala is one; you can say:
- It was a gala party
- *The party was gala.
I’ve marked this posting Community Wiki because it is John’s answer not my own, and so I deserve no reputation from it.
I've found the Collins Cobuild English Grammar to be very helpful hereabouts. It's not totally comprehensive (what is?), but contrasts three of the four positionings of adjectives that I'm aware of pretty well.
One thing it makes clear is that different positioning sometimes controls semantic difference:
The aeroplane had to land on Runway 7, as the proper runway, Runway 1, was damaged.
We veered off the runway proper onto the verge.
The concerned parents / the parents concerned
(Participial adjectives always need watching.)
However, some adjectives modify nouns they're not obviously attached to:
I'd love a quiet pint.
It was a proud day for her parents.
(These are 'transferred epithets' where the noun being modified has to be recovered from the context: quiet occasion / surroundings; the girl's parents.)
Even more peripheral non-semantically-predicative usages are:
A former president / a future king
A mere youth / fake diamonds
where what are traditionally labelled adjectives as they pre-not-really-modify 'their' nouns refer to former or future states, or reference whole groups to which the nouns' referents belong (or not). Most of these adjectives are never used predicatively (These diamonds are fake // *The president is now former / *I'm having a pint that is quiet / *Those youths are mere / *This day is proud for her parents).
[In] the preceding sections [I] have argued for the value of a grammatical distinction between functional adjectives and lexical adjectives. To my knowledge, this exact distinction has not previously been made in the literature. However, it is not a new observation that some adjectives lack the full range of prototypical adjectival behaviour ... The list of “defective adjectives” [is shown in] the rough classiﬁcation suggested here; it is for descriptive convenience only—the development of a grammatically and semantically adequate classiﬁcation requires further work.
(8) a. Identity adjectives: same, different, other
b. Ordinal adjectives: next, last, previous, subsequent, preceding, further
c. Degree adjectives: utter, sheer, outright
d. Signiﬁcance & particularity adjectives (predicate-evaluating): main, chief, principal, mere, common
e. Extent adjectives: entire, whole
f. Temporal adjectives: future, then [former, past, now? once?] (termed modal II by Coppock)
I'll risk adding, though there is often semantic content in addition to a non-canonical syntactic role, and predicative use may occur:
g. Frequency etc adjectives: occasional, rare, common, frequent, sporadic, intermittent ...
g’ Other adjectives referring to how assembled referents co-relate: various, assorted, diverse, disparate, diversiform, similar,
h. Adjectives of potential / lack of (etc) ...: potential, budding, would-be, wannabe, manqué, frustrated [barred sense] ...
i. Modal and veridical (including privative) adjectives: true, suspected, alleged; (false, fake – can be used attributively)
j. Adjectives of selection / confirmation / rebuttal (very, exact, correct, wrong)
k. Event-manner adjectives: eg beautiful in a beautiful dancer, heavy in a heavy smoker
l. Transferred epithets, usually psychological experience adjectives: eg proud in a proud day, disabled in disabled toilet
Those listed in h are certainly 'peripheral' (not grading or normally appearing predicatively). Those in g semantically are non-central; they specify the type of occurrence etc of the noun's referent rather than a truly inherent property - a background is implied with dots marking occurrences. With temporal adjectives, lines against a time background (marking eg the past) are implied.
Coppock has a later publication on The Predictability of Predicativity in which she enunciates the rule: 'An adjective is syntactically predicative if and only if it is semantically predicative'.
Attributive adjectives are used with a noun, mostly before a noun, sometimes after a noun: a little child, an old man, a large forest
Predicative adjectives are used after to be (and similar verbs): The man is old.
The Latin grammar term attribute is a term for anything you can attach to a noun: articiles and article words (they go into the same position as the articles: possessive adjectives, demonstrative adjectives), adjectives, participles, and relative clauses)
"Predicative use" of an adjective means the adjective is part of the predicate.
Most dictionaries indicate if the use of an adjective is limited to predicative use: The entry of alive of OALD says: adjective, not before a noun. Most adjectives of the type alive, beginning with the prefix a- can only be used as predicative adjectives.
I agree with the identification of the two types of adjective by position, but can we consider this idea, it's been working for me.
Considering that using an adjective with/for a noun is like passing a judgement, I identify an adjective as predictive when there is an antecedent or experience that bring about the conviction for your judgement. Consider these:
grass-fed beef tastes better than others.
The examination was unbelievably easy.
The judgement in the above sentences were based on the conviction drawn from experiences of tasting the different types of beef and writing a particular exam. On the other hand, I consider an adjective to be attributive when the judgement is based on instant perception drawn from visible and perceivable attributes/characteristics of a noun. Let's consider these:
he lives in a two storey building.
He goes in a blue Landrover jeep.