Some strings of words are importantly different from other strings. For example, they might be used way more often than synonymous strings which would seem to be equally good choices.

Wikipedia defines collocation as

a sequence of words or terms that co-occur more often than would be expected by chance.

This is a pretty good definition, although "the" and "man" certainly co-occur more often than they would if language were a genuinely chance grabbing of words from a bag. By the above definition, "the man" counts as a collocation. I think this is undesirable.

Despite the imprecision of the definition, we all have an intuitive idea of what a collocation is. It's a common phrase. Some people use set phrase, fixed expression, and even idiom (sense 1.1, here) in this way, meaning "common phrase".

On this understanding, all these terms mean something like "a form of expression that comes more naturally to, or is more popular among, a large subset of speakers than a synonymous expression". On this definition, "tall, dark, and handsome" counts as a collocation/set phrase/fixed expression/idiom.

I am not interested in formulating a precise definition of this concept, nor am I looking for precise criteria to differentiate, say, a collocation from a non-collocation. I am fine stopping at intuition, and boundaries, like peaches, are sometimes fuzzy.

But is there a term for the opposite of a collocation or common phrase? That is, is there a concise or canned way to describe a string like "birds fly overhead", a mere output of the normal combinatorics of language? Is there a word which describes, as @Drew puts it, "common words put together in an ordinary way"?

The reason I ask is because this site sometimes gets questions like

  • Who coined the phrase "the dog"?

  • What is the origin of the phrase "fly like a bird"? (cf. here)

  • What is the first occurrence of the phrase "in the back of the house"?

I always struggle in expressing the idea that these are not set phrases, that they were not "coined", that they were likely invented simultaneously by a hundred people, and re-invented by a billion more.

What would you call such strings?

Compositional doesn't quite work since phrases like "tall, dark, and handsome" are both compositional and set phrases. Plain language doesn't quite work for the same reason, "tall, dark, and handsome" is both plain as well as a set phrase.

I am looking for nouns, adjectives, or whole phrases which can be used to describe such strings, so I am flexible. That said, here are some example sentences:

  • "In the back of the house" is not a collocation, it's a ___.
  • "In the back of the house" is not a set phrase, it's ___.
  • "In the back of the house" is not a common phrase, it's ___.
  • Yes. Utterances that are non-idiomatic, but rather normal, are called Compositional. I.e, their meaning can be understood by composing the meanings of their constituents -- thus Bill kicked the ball is compositional, while Bill kicked the bucket is not. Aug 23, 2016 at 17:47
  • @JohnLawler, In my question I was using idiom in its looser sense of "a form of expression natural to a language, person, or group of people" (ODO). Whereas "tall, dark, and handsome" is idiomatic, "tall, mysterious, and good-looking" is __. Both are compositional. Maybe non-idiomatic is the closest word.
    – DyingIsFun
    Aug 23, 2016 at 17:53
  • Plain language? : Plain language is writing designed to ensure the reader understands as quickly, easily, and completely as possible. Plain language strives to be easy to read, understand, and use.It avoids verbose, convoluted language and jargon.
    – user66974
    Aug 23, 2016 at 18:05
  • @Josh61, interesting suggestion, and it works for some cases. But "tall, dark, and handsome" is still plain language, despite being a fixed expression. So plain language and collocation/set phrase/fixed expression are not mutually exclusive.
    – DyingIsFun
    Aug 23, 2016 at 18:10
  • 1
    Instead of co-occur more often than would be expected by chance (a lousy description, IMO), I would say co-occur more often than not, or more precisely, co-occur at least as often as at least one of the words occurs otherwise. *Motley crew is a collocation because it occurs at least as often as motley occurs without crew.
    – Drew
    Aug 23, 2016 at 20:15

1 Answer 1


The Wikipedia article on collocation makes salient points about how collocations are regarded:

The processing of collocations involves a number of parameters, the most important of which is the measure of association, which evaluates whether the co-occurrence is purely by chance or statistically significant. Due to the non-random nature of language, most collocations are classed as significant, and the association scores are simply used to rank the results. Commonly used measures of association include mutual information, t scores, and log-likelihood. Rather than select a single definition, Gledhill proposes that collocation involves at least three different perspectives:

(i) cooccurrence, a statistical view, which sees collocation as the recurrent appearance in a text of a node and its collocates,

(ii) construction, which sees collocation either as a correlation between a lexeme and a lexical-grammatical pattern, or as a relation between a base and its collocative partners and

(iii) expression, a pragmatic view of collocation as a conventional unit of expression, regardless of form.

It should be pointed out here that these different perspectives contrast with the usual way of presenting collocation in phraseological studies. Traditionally speaking, collocation is explained in terms of all three perspectives at once, in a continuum:

'Free Combination' ↔ 'Bound Collocation' ↔ 'Frozen Idiom'

[bolding mine]

So, at least traditionally. 'free combination' is the term for a (very) loose association within a string.

  • That passage suggests a conceptual link between (i), (ii), and (iii) and "free combination, bound collocation, and frozen idiom", respectively, which I fail to see. Still, free combination seems to work as a (fairly transparent, but somewhat opaque) descriptor of what I want. +1! Accidental combination and incidental combination could also work, I guess.
    – DyingIsFun
    Aug 23, 2016 at 19:17
  • 1
    You may like to look through the Collins Cobuild article on collocation. They do not include colligations (eg rely on, wait for, crowd of, cant help + -ing) as collocations, and doubtless this would rule out 'the man' etc (as a 'determiner phrase' according to DP-analysis). They also rule out what they call 'lexical bundles', but include 'fixed phrases' (though I can't see much difference). Aug 23, 2016 at 20:44
  • Thanks Edwin, I'll take a look! Although, as I pointed out, I'm not too concerned with finding a precise definition of collocation which excludes "the man" and answers to all our intuitions regarding it. You could probably block such "colligations" by adding a clause to the effect that function (or syncategorematic) words (like 'the') are to be discounted.
    – DyingIsFun
    Aug 23, 2016 at 20:49

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.