In a comment to Acrasial Philogramy, reader kts notes another recent change in OED labelling policies:
[…] they’re also replacing the label nonce-word with “Apparently an isolated use” — aw, that was part of the OED’s charm, but I guess it too must have been too obscure for some people (Yes, pancakewards has the new label; don’t we all have a secret soft spot for that one?)
Do they enter nonce-words anymore? Or is the policy against them now?
Don’t we all have a soft spot for nonce-words, which at their best show off language in its most vigorous and generative permutability, even if at their worst they resemble nothing more than gratuitous word torture. For every chirographosophic, decapterygious, or monoglacialist, that is, there is a chocolatical, or don’t-knowist, dotardy, growlsome, guzzledom, harassery, panfrivolium, or indeed pancakewards–the list goes on and on.
And don’t we have a soft-spot for “nonce-word“, the word JAH Murray made up to describe words people make up for single use? OED2 had defined it, autoglassatingly [won’t find that in the OED!], sv. nonce, n1:
4.4 attrib.: nonce-word, the term used in this Dictionary to describe a word which is apparently used only for the nonce (see vol. I, p. xxvii); similarly nonce-use, etc.; nonce-borrowing, -combination, -form, -formation, -meaning.
But the current definition in OED3 has:
C1. General attributive or as adj. Designating a lexical item formed for use on a specific occasion. Occasionally also in extended use.
with the further note:
C2.Nonce occurs in numerous other compounds in N.E.D. and the O.E.D. Suppl., such as nonce-abbreviation, nonce-diminutive, nonce-expression, nonce-translation, nonce-verb, etc., and some of these are occasionally found in other sources.
This does certainly imply, as kts fears, that the term has been left aside or replaced in OED3, and unless you happen to keep old versions of the dictionary hanging around, as I like to, you have no way of knowing that this was not always the case.
Indeed, until just recently, the revised [in 2003] definition of “nonce-word” was:
[…] one of a number of terms coined by James Murray especially for use in the N.E.D. a word apparently used only “for the nonce”, i.e. on one specific occasion or in one specific text or writer’s works.
This way of putting it made no claim of obsolescence regarding the usage of nonce-word, was historical, and had the further excellent property of being self-describing: “nonce-word” is a nonce-word on his definition, even if it has gained a certain restricted currency since its coinage.
So is it true that nonce-word only occurs in NED (1928) and OED Suppl (1989), and not in the OED3 revision? Certainly it was occurring until very recently, if mainly in legacy form. As late as January 2020 there were 747 occurrences of nonce in OED3, though only 7 of them occurred in new entries added since 2000 (chart shows 11, but includes 4 false positives such as “noncey”), and 235 in fully revised entries. Compare that to 3,981 instances of “isolated use”, including in 84 new entries and 1,687 fully revised entries, and it is clear enough that a policy shift away from “nonce” labels has been in effect since some time after 2017 (when “nonce” had diminished considerably, but “isolated use” remained… well, relatively isolated).
This trend has accelerated recently, however, with “spot” revisions taking place in otherwise unrevised or lightly edited entries (including all the new post-2000 entries, where the few new nonce-words that were added are now relabeled rare or isolated…). Today there are just 376 remaining senses containing a nonce label, under 347 headwords, which, checking rapidly, are all in unrevised entries (though you wouldn’t necessarily know that, since the red revision labels contain “modified version” dates in the late 20-teens and early -twenties). How they escaped spot revision so far escapes me, but I would assume they too will disappear shortly.
Call me growlsome but I don’t much like the terminology of “isolated use”. I suppose it is more accurate and descriptive, but it lacks cachet, and sounds to my ear anyway somewhat approbatory–more dismissive and less of a challenge than “nonce-word” sounds in some circumstances.
Oh well, onwards and pancakewards, I suppose, is the way of things.
Here’s a selection of words that used to be nonces but now are rare or isolated”
“until just recently” — huh? That’s still the definition of nonce-word; it has its own definition, in addition to the more general definition of nonce compounds at C1.
Also, I’m highly skeptical of the extension in that 2003 definition to “one specific text or writer’s works”. If it’s used multiple times in a text, or across texts by a specific writer, that makes it established in the context, and not nonce. Nonce to me means it was made up fresh, on the fly: the writer had never used it or even heard it, but still the reader is expected to understand it because it was created using familiar word-formation methods. A new assembly of off-the-shelf parts, you might say. I don’t think “nonce-word” was ever self-descriptive: it was intended for regular use throughout the dictionary.
Just as a guess, maybe they’re taking out “nonce-word” because it requires interpretation of the writer’s intention, which they can’t prove objectively? But the whole dictionary involves interpretation of every citation anyway, so that would be a pretty weak reason. I’m with you on “isolated use” lacking cachet, and sounding dismissive! And it fails to recognize the deliberate playfulness of a pancakewards.
Or, another guess: maybe they don’t want to stick a “nonce” label on formations that might yet catch on, or be re-invented, or even antedated? The OED2 cited Joyce for free-fly (v.) and labeled it “nonce-use”; the 2008 revision turned up a couple of much later citations from engineering contexts, and removed the label. I’m sure those engineers didn’t get the word from Ulysses, they independently re-invented it! But in general, investigating whether other uses were independent re-inventions or not might be beyond the scope of the dictionary.
I googled “nonce site:public.oed.com” and found a trace of a former “nonce”, in the glossary of grammatical terms: -erati (comb. form) was formerly defined as ‘Forming nouns (often humorous nonce-words) designating elite or prominent groups of people who are associated with what is specified by the stem word.’ Now the definition reads “humorous ad hoc formations”. Actually, I’m OK with this one. It conveys the same thing.