38

I accept that my premise may be incorrect, but here it is.

The word alone, when used as an adjective, seems only to fit in sentences of the form:

  1. The X is alone.

and not in the form:

  1. The alone X...

I can't think of any other adjectives that have this property, that is, that are allowed in constructions like (1), but not in (2).

Question 0: Am I right in thinking that The alone X is not correct English?

Question 1: Are there any other such adjectives that have a similar behavior?

Question 2: Is there a word for such an adjective and is there any rationale for their existence?

Dancrumb
  • 5,080
  • Very interesting question. I agree that the [alone N] construction sounds funny. But here is a context where it doesn't sound so bad: "I saw two men. The first was alone. The second was with his wife. The alone man had an umbrella." – DyingIsFun Jun 29 '16 at 14:07
  • Isn't alone an adverb, there, which makes it easy to use to modify a verb, but more difficult for an noun? – Dan Bron Jun 29 '16 at 14:08
  • 2
    Good example. It still feels a little strange - I think I would probably say "the one who was alone had an umbrella". Maybe it's more of a question of idiom than correctness, but still... it strikes me as a strange aberration. – Dancrumb Jun 29 '16 at 14:09
  • I believe it is both an adjective (describing the man) and an adverb (describing an action): "My mother alone can open the jar of pickles", meaning only my mother can do this. – Dancrumb Jun 29 '16 at 14:10
  • @Dancrumb, I agree. By the way if you search "the alone" in Google books (a construction you'd expect to be very frequent if 'alone' could precede a noun) you get almost nothing. https://www.google.com/search?q=%22alone%20man%22&tbm=bks&lr=lang_en&gws_rd=ssl#q=%22the+alone%22&lr=lang_en&tbs=lr:lang_1en&tbm=bks&start=10 – DyingIsFun Jun 29 '16 at 14:10
  • 3
    You might be mixing up "alone" and "lone". "The lone X" makes more sense to me than "The alone X". However, "The X is alone" is okay and it uses "adjective" in place of noun. That's all. – NVZ Jun 29 '16 at 14:18
  • 1
    Dancrumb, I edited your post to include a short proof that 'alone' is an adjective in (1). I also re-ordered your questions (for aesthetic and pedagogical reasons). Roll-back if you don't like. – DyingIsFun Jun 29 '16 at 14:19
  • 5
    The corresponding adjective is lone. Also, lonely. – Kris Jun 29 '16 at 14:25
  • 1
    @Kris, but 'alone' is listed as both an adverb and adjective in most dictionaries. And (1) demonstrates that 'alone' can appear as an adjective. As an adjective why is it blocked from appearing in (2)? That is the second question. Is your conjecture that the existence of 'lone' is enough to block it from appearing there? – DyingIsFun Jun 29 '16 at 14:30
  • 2
    @Dancrumb Please wait a day or two before selecting an answer. You may get several more answers with other ideas or different advice. But people may not bother to write you another answer if you've already selected one! :-) – Araucaria - Him Jun 29 '16 at 15:11
  • 1
    Thanks... I pulled back my acceptance (sorry Silenus) to wait to see what comes... some great answers so far!! – Dancrumb Jun 29 '16 at 15:44
  • Partly answered at When can an adjective be postposed? and Aberrant usage of the adjective “incredulous”. – Edwin Ashworth Jun 29 '16 at 16:01
  • 1
    Possible duplicate of Is the phrase "the alive animal" grammatically correct? – Edwin Ashworth Jun 29 '16 at 16:09
  • @EdwinAshworth The question is "What is the name for X?", not "When is X used?" – Araucaria - Him Jun 29 '16 at 23:58
  • @EdwinAshworth I think it is the right thing to do to close the old question as a duplicate of this question as the answers to this question are more comprehensive and generic. –  Jun 30 '16 at 04:42
  • @Rathony I thought it was (2) site policy to combine duplicates under the original question, where there are better answers in the later one; (1) the responsibility of [potential] answerers to check (within reason) for likely duplicates, and to close-vote and answer at the original. I've asked for clarification at Is it appropriate to mark an earlier, poorer question as a duplicate of a later, better one? on Meta. – Edwin Ashworth Jun 30 '16 at 09:37
  • 1
    "The X is small, weak, and alone." "The X is small, weak, and in trouble." Unless we regard complete prepositional phrases as adjectives, the proposed proof fails. – Gary Botnovcan Dec 04 '18 at 16:00

3 Answers3

26

There are three main positions for adjectives in English:

  1. Predicative: where they occur as the Predicative Complement of a verb, such as FEEL, LOOK, SEEM or BE:

    • He felt alive.
    • The elephants looked serene.
    • She seemed discombobulated.
    • He was irate.

  2. Attributive: where they occur as the modifier of a noun:

    • You're talking utter nonsense.
    • That's a large elephant you have there.
    • I can't stand the hypocritical cant of politicians
    • The once ubiquitous sparrow is now an endangered species.

  3. Postpositive: where they occur after the noun they are modifying:

    • the quickest route possible
    • the worst conditions imaginable
    • the best deal available
    • the body politic

Notice that the adjective responsible can occur in all three positions. So we can say that responsible can be a predicative, attributive and postpositive adjective. This would basically just be saying that responsible is an adjective that can undertake all the normal adjective functions that we normally expect adjectives to be able to.

However, some adjectives can only be used in some of these positions. In other words the syntactic functions that they can carry out are restricted. For example, notice that the adjective alive in group one cannot be used attributively:

  • *I found an alive man. (ungrammatical)

Some people say therefore that alive is a predicative only adjective (this is not correct, as explained further below).

Conversely, the adjective utter in group two can only be used attributively:

  • *Their stupidity was utter. (ungrammatical)

Utter is therefore often described as an attributive only adjective.

Most adjectives that occur postpositively in set phrases only occur postpositively, such as the word politic in group three above. We usually just say that the words are postpositive adjectives.

The Cambridge Grammar of the English Language (Huddleston & Pullum, 2002) broadly categorize adjectives with restricted functions into two categories: attributive-only adjectives and never-attributive adjectives. Attributive-only adjectives are:

  • "Adjectives that do not normally occur except as (heads of) attributive modifiers [...]".

Never-attributive adjectives are:

  • "Adjectives which can occur predicatively or postpositively, but not attributively".

Note that this is a much better description of the adjective alive, for example, which can of course occur postpositively as well as predicatively:

  • Anything currently alive is salvageable.

References: The Cambridge Grammar of the English Language, Huddleston & Pullum, 2002. pp 553-562.

Araucaria - Him
  • 47,146
  • 1
    "I found a man alive" (postpositive) seems quite grammatical to me, so I"d say it's "never-attributive" rather than "predicative only". – Monty Harder Jun 29 '16 at 21:10
  • @MontyHarder Thanks for the helpful comments :) You're quite right, which is why the CaGEL description is the best one. I realise my original description is not. I didn't make it clear that it isn't a good description. I think I've fixed the wording. Do you think the post is correct now? – Araucaria - Him Jun 29 '16 at 22:51
  • I don't believe there's much value in exploring the "predicative only" dead end. Also, the CaGEL division of adjectives is obviously wrong as written here; in addition to attributive-only and never-attributive adjectives, there are those that can be used attributively but are not exclusively attributive. – Monty Harder Jun 30 '16 at 14:11
  • @MontyHarder Yes, those are called "adjectives" under the H&P framework! ;-) – Araucaria - Him Jun 30 '16 at 14:13
  • Postpositive is still "attributive"; it's just not in the typical "prepositive" position. – eques Jun 30 '16 at 14:53
  • @eques "attributive" here doesn't mean describing an attribute. It is a syntactic label, not a semantic one. It means "pre-head modifier in a nominal". The fact that attributive and postpositive functions are different syntactic functions is however evident from the different meanings available when the same adjective occurs in either attributive or postpositive functions. – Araucaria - Him Jun 30 '16 at 15:52
  • @eques It's far from obvious that postpositive position is an attributive position. It's much more likely to be a predicative position, since the following descriptive generalization holds (counterexamples welcome): If an adjective is an "attributive only" adjective (like utter, or mere) it can only appear prenominally, and never postpositively. So the postpositive and the predicative adjectives pattern together, and the prenominal attributive only pattern separately. This implies that postpositive adjectives are predicative adjectives. – Alan Munn Jun 30 '16 at 15:53
  • @Araucaria Wikipedia's article on the Adjective (https://en.wikipedia.org/wiki/Adjective) includes "postpostive" under Attributive. – eques Jun 30 '16 at 17:12
  • @AlanMunn Your proof is far from conclusive. A predicate is a defined grammatical structure (https://en.wikipedia.org/wiki/Predicate_(grammar)); thus a predicative adjective is one that forms part of the predicate by referring back to the subject. In English, postpositives are exceptions to the normal rules; that is, most adjectives are prepositive attributive and predicative adjectives. Some are only predicative or attributive. Only few can be used postpostively possibly as the only way. – eques Jun 30 '16 at 17:17
  • @eques As soon as you start talking about 'exceptions' you are missing generalizations, particularly if the 'exceptions' have a definite pattern. Definitions are only useful if they allow you to make accurate generalizations about language, so citing a definition isn't an argument against my (and Araucaria's) point of view. We've shown a generalization about postpositive adjectives and predicative adjectives, but you have no empirical counter argument whatsoever. – Alan Munn Jun 30 '16 at 17:30
  • @AlanMunn "postpostives" are an exception in that the majority of adjectives do not function postpostively in standard usages (e.g. outside of poetry). Attributive adjectives both prepositively and postpostively still form a coherent abstraction. Your assertion is that postpostives group with predicative adjectives is just that an assertion based solely on the fact that they don't precede the noun. Postpostive and prepostive attributive adjectives modify nouns to create noun phrases; a predicative adjective exists in the predicate and relies on a copula-like verb to connect – eques Jun 30 '16 at 17:46
  • @eques Predicative adjectives do no require a copula in cases of absolute constructions such as With John unhappy, we can't continue. or inside verbs like consider or make: I would consider John happy, I made John happy. No, my claim is made on the fact that adjectives that can't appear in predicative position but can appear in attributive position are unable to appear in pospositive position. This is unexpected if postpositive adjectives are attributive. – Alan Munn Jun 30 '16 at 17:56
  • Let us continue this discussion in chat. – eques Jun 30 '16 at 18:01
  • @MontyHarder Oh, I hadn't proofread part of my post carefully, so you may have had a point. I meant to say that CaGEL classify those adjectives which are restrictedin the functions they can fulfil into those two categories. – Araucaria - Him Jun 30 '16 at 20:34
  • We did learn all this in school. I thought the questions was Why? instead. – Kris Jul 01 '16 at 07:16
  • Is there a good reason to prefer classing these words as restricted adjectives rather than intransitive prepositions? – Gary Botnovcan Dec 04 '18 at 14:59
  • @GaryBotnovcan Good question. Answer's a bit long for the comments though. If you ask it as a Q, I'll try and write an answer :) – Araucaria - Him Dec 06 '18 at 16:39
  • Done. – Gary Botnovcan Dec 06 '18 at 18:32
  • awake also can be used in the same fashion.(the adj.one - predicative) – Flonne Dec 08 '18 at 16:28
22

It is true that 'alone' can occur as both an adjective and an adverb. This is attested by many dictionaries.

If you look at the Oxford definition, you see:

alone

Having no one else present; on one’s own:

[AS PREDICATIVE ADJECTIVE]: 'she was alone that evening'

[AS ADVERB]: 'he lives alone'

Notice that Oxford classifies it as a predicative adjective in its example sentence. Predicative adjectives are adjectives that occupy predicative positions, that is, come after copula verbs like 'is' and 'make'. If an adjective precedes a noun, it is said to be in attributive position. NOTE: The descriptors 'predicative' and 'attributive' are not absolute. They are used to describe an adjective relative to the adjective's grammatical context.

When 'alone' occurs in your example (1), it is indeed in predicative position. But when it occurs in (2), it is not in predicative position, which is why it is ungrammatical.

'Alone' is restricted only to predicative positions. We might call it a predicative-only adjective (following @Araucaria). 'Alone' is not the only adjective that is restricted to predicative positions. 'Afraid' is also restricted to predicative positions. Interestingly enough, other adjectives are restricted only to attributive position (for example 'main'). Here is an instructive passage:

Most adjectives can freely occur in both the attributive and the predicative positions. However, a small number of adjectives are restricted to one position only. For example, the adjective main ("the main reason") can only occur in the attributive position (predicative: *"the reason is main"). Conversely, the adjective afraid ("the child was afraid") can only occur predicatively (attributive: *"an afraid child").

Many adjectives beginning with the preffix 'a-' are restricted to predicative positions (for example 'afloat', 'awake'), as pointed out by @Rathony. The Wikipedia article on English prefixes even lists the prefix 'a-' as one that turns verbs into predicative adjectives. An interesting question is whether or not all predicative-only adjectives were formed from the 'a-' prefix. As far as I know, this is an open question.

Community
  • 1
DyingIsFun
  • 17,909
  • 1
    I see no evidence above that an adjective that cannot be used in the attributive position is classified as a 'predicative adjective'. Note that 'alone' can be used in absolute constructions (Alone with her thoughts, she sat on the garden bench for hours.) From IGE: Most adjectives can occur both before and after a noun:

    the blue sea ~ the sea is blue.

    Adjectives in the first position - before the noun - are called ATTRIBUTIVE adjectives. Those in the second position - after the noun - are called PREDICATIVE adjectives.

     – Edwin Ashworth Jun 29 '16 at 15:40
  • The Internet Grammar of English, at least, adopts the usage that 'predicative' etc refers to the usage under consideration and is not a label of a type of adjective. Thus 'blue' may be attributive or predicative (or absolute) depending on the way it is being used. This corresponds to transitive and intransitive usages of verbs, and count and non-count usages of nouns. – Edwin Ashworth Jun 29 '16 at 15:44
  • @Edwin Ashworth, I agree, most adj can occur in both positions. (1) The blue sea. (2) The sea is blue. My view is that it is the same adjective in both. It's not an attributive adj in the first sentence and a predicative adj in the second. This would mean that there are two words, 'blue' (qua attributive adj) and 'blue' (qua predicative adj). I don't like this. On my way of speaking 'blue' is simply a predicative-and-attributive adjective, and I use 'predicative adjective' to mean 'predicative-only'. Same goes for 'attributive adjective'. But I might be drifting from common usage. – DyingIsFun Jun 29 '16 at 15:46
  • You need to back up your view with an authority sanctioning this usage. Rathony uses '[Some] adjectives ... function only as a predicate adjective', which I believe to be correct. – Edwin Ashworth Jun 29 '16 at 15:48
  • 1
    @EdwinAshworth, thanks for helping me to clear this up. It's subtle which is why I have a hard time with it. I've made an edit which I hope conforms with common usage. – DyingIsFun Jun 29 '16 at 15:57
  • According to a comment in the previous question (which I believe is close enough to be considered a duplicate), CaGEL (p559) lists alive in its group of "very clearly non-attributive adjectives" which seems to validate your original usage. CaGEL is at odds with other sources (surprise, surprise). – Edwin Ashworth Jun 29 '16 at 16:13
  • @EdwinAshworth Your comment doesn't mean so much if you don't explain how CaGEL's description diverges from other writers'. It will appear to be just a bit of a Brexit comment,. – Araucaria - Him Jun 29 '16 at 23:07
  • The awake student passed the quiz; the asleep student did not. – Eric Towers Jun 30 '16 at 07:16
  • @Araucaria The previous comments show two examples of 'attributive' used to describe a particular usage rather than a subset of adjectives. CaGEL clearly uses [non-]attributive as a subset descriptor. Do try to keep abreast of the whole debate. – Edwin Ashworth Jun 30 '16 at 09:27
  • 1
    @EdwinAshworth Your being rather fast and loose with the actual facts. CaGEL call that group of adjectives "never-attributive adjectives" which shows that they are using "attributive" to describe the syntactic function and not a class. They use the term 'non-attributive' in passing in an adjectival sense, not as a name for the class, by which it is just meant that these adjectives cannot normally occur in attributive function. You're tilting at windmills. Incidentally, Bas Aarts (of IGE)'s view is that CaGEL may well be the greatest grammar of English ever written. – Araucaria - Him Jun 30 '16 at 11:16
  • Best not to use dictionaries for grammar! That usage of alone that they give is definitely not an example of alone occurring as an adverb. Consider "I went there naked" ;-) – Araucaria - Him Jun 30 '16 at 11:40
  • @Araucaria The phrase '[o]ne group of very clearly non-attributive adjectives' must be defining a set characteristic. I'd consider 'one group of count nouns' incorrect, and I consider CaGEL's terminology wrong here. They should have stuck with '[o]ne group of very clearly never-attributive adjectives' or used ''[o]ne group of adjectives which are very clearly not/never used attributively'. – Edwin Ashworth Jun 30 '16 at 22:45
  • I should fully agree with this, though the OP did indeed check a dictionary and would have known this, too (may have overlooked the predicative part for some reason). See my comment at Araucaria on this page. – Kris Jul 01 '16 at 07:18
7

Many adjectives that start with the letter a function only as a predicate adjective, e.g.:

alike, afraid, asleep, alive, awake, alone and aloof, etc.

Some of the above listed adjectives are made with "a + noun". According to Online Etymology Dictionary, aloof comes from

1530s, from a- (1) + Middle English loof "weather gage," also "windward direction," probably from Dutch loef (Middle Dutch lof) "the weather side of a ship."

Asleep is from

c. 1200, aslepe, o slæpe, from Old English on slæpe (see sleep). The parallel form on sleep continued until c. 1550.

The same dictionary explains about a in the above words as follows:

in native (derived from Old English) words, it most commonly represents Old English an "on" (see a (2)), as in alive, asleep, abroad, afoot, etc., forming adjectives and adverbs from nouns; but it also can be Middle English of, as in anew, abreast (1590s); or a reduced form of Old English past participle prefix ge-, as in aware;

Ben Kovitz
  • 1,255