English Language & Usage Stack Exchange is a question and answer site for linguists, etymologists, and serious English language enthusiasts. It only takes a minute to sign up.
Sign up to join this community
Is it possible to use un- with new words such as sit, sleep, sad? I'm currently seeing many words (in programming) which use un- in the meaning of undoing something.
For example, is it possible to use "make me unsleep" (the Chrome spell-checker underlines it) as a synonym for "wake me up"?
If yes, why is it so rare? And if not, why can't it be used?
Normally un- with a verb means to reverse—“undo”—the previously-taken action of the verb. You “unscrew” a jar lid someone previously screwed on; you “unwrap” a package someone previously wrapped.
You can even use un- with verbs signifying actions normally regarded as irreversible—create or kill, for instance—as long as you are speaking or writing in a context where those verbs have non-literal meaning, such as game design.
But it is semantically awkward to use un- with activities or statives — verbs which signify actions without particular goal or existence in a particular state.
“Sleep” is normally a stative, so it can’t be reversed. Unsleep would be acceptable only in cases where sleep was being used transitively, with the sense “put to sleep”, or inchoatively, with the sense “fall asleep” —neither of which is in my experience used.
Programmers, of course, can name variables or methods with any words in any sense they like. They are “communicating” only with compilers, which aren’t encumbered with the linguistic rules which obtain in natural languages.
And of course you can coin unsleep and use it in any sense you like. That’s how the language grows. But if you use such nonce-words in an unconventional way you will run the risk of mis- or non-understanding.
EDIT:
Since you ask about sad: You have considerably more leeway in prefixing un- to adjectives, including adjectives formed from participles, because here un- bears more general negative senses. Unsleeping, for instance, is perfectly conventional: an ‘unsleeping’ watchman is one who is not sleeping, not one who ‘unsleeps’ someone else. An ‘unlicked cub’ is a cub which has not been licked, not one which has been ‘unlicked’. Un- can also be prefixed to nouns with a simple negative sense—unperson, for instance. Verbs can even be formed by prefixing a noun with un- in the sense of ‘deprive of’ (unstate) or ‘remove from’ (unearth).
This looks like what you're asking about.
As you can see, it's a topic of great interest to linguists, and Larry Horn is pretty much the expert on negation.
To quote the abstract:
Since Whorf (1936), many linguists have tried their hand at corralling the restrictions on the formation of "reversative" un-verbs; cf. e.g. Marchand (1969), Dowty (1979), Horn (1988), Clark et al. (1995). Why can you unwrap a sandwich but not unrecognize its contents or unremember to toss it in the trash? Why can a snake uncoil while a painting can't unhang? If unfreeze is the opposite of freeze, why is unthaw a synonym of thaw?
The standard approach to the constraints on un-verb formation invokes Whorf's CRYPTOTYPE -- a covert category encompassing transitive verbs of covering and enclosing that rules out a wide range of possible bases and outputs of the relevant rule. Pullum (1999), for example, reckons that there are "about a dozen verbs" that allow un-prefixation, citing undo (a good deed) and unknow as examples of formations we know "intuitively" are impossible. Clark et al. (1995) exclude unbury and unbend, while Kemmerer & Wright (2002) rule out unboil and undecorate. Yet many of the verbs depicted in the literature as impossible, non-occurring, or -- as in Whorf's label for unsay and unmake -- "semi-archaic" are readily attested, even when the actions they denote may be physically irreversible.
While the Pullum/Whorf view may be extended to predict the pleonastic interpretation of source-oriented reversatives in Swedish, French, and English (unloosen, unthaw), it incorrectly limits the productivity of un-verb formation by conflating the SEMANTIC (aspectual) restrictions with the PRAGMATIC conditions on the way the world (normally) works; verbs like unsay, unknow, unboil, and unhappen are motivated precisely by the need to describe those (typically counterfactual) situations in which the tape of reality is reversed.
The pragmatic nature of the restrictions on un-verb formation is supported by a survey of contexts that favor the emergence of innovative un-verbs: advances in science and technology (as in the unerase and undelete commands, the unfriend or unlike verbs of social networking, or the unfuck program to reverse software protection), science fiction (as in time-travel scenarios), advertising copy (as in KFC's current unthink campaign), and the imagination of poets from Shakespeare (whose "un-king’d" Richard II is the unchallenged monarch of this realm) to pop songsters ("How can I unlove you?", "Un-break my heart"), whether the implausibility of a given reversal is conceded, mourned, or overridden.
As you say, un-, when attached to certain verbs, reverses the action expressed in it, as in uncover, undo, undress, unfasten, unleash, unload, unlock, unplug, untie, unwind, and go and go on. But many other verbs, like - for example - break, expect, seek, smile, cannot be reversed in this way. Sleep belongs to this last category, so you cannot say 'unsleep' for saying "wake me up".
Per my answer to Should it be unlike or dislike button in Facebook?, and noting OP's mention of programming, install/uninstall, register/unregister, delete/undelete, do/undo are now common in that context. And programmers wouldn't bat an eyelid at references to an unsleep function.
As that first link explains, there wasn't really any need for unlike before Facebook, because we already had dislike. But that didn't quite fit the Facebook context, so the new coinage succeeded.
Historically speaking, as this answer to a related question points out, in general, words take un- when they are of English (Germanic) origin. But it seems to me native speakers do tend to go for un- rather than other negating prefixes such as a-, an- in-, de- dis, dys, etc. when coining unashamedly facetious nonce-words, regardless of the base word's etymology.
I don't actually think un- is at all "rare" as a negating prefix for "standard" words. The reason it "can't" be used with unagree, unsit, unwatch etc., is mainly that these are not currently considered valid words (we already have the word disagree, and unsit could only mean stand, but I wouldn't be surprised to see unwatch appear in an eBay context soon).
As these few instances show, unsleep is "recoined" from time to time (often in literary/poetic contexts) but we're dealing with highly idiomatic usages here, and I would certainly advise non-native speakers to avoid forms they don't see others using frequently.
TL;DR: un- is not rare, but "acceptable" new coinages in established contexts are uncommon (few would be happy with unrare there, for example).
