Back on December 27th, Doug Harris sent me this example (crucial bit boldfaced), from that day’s Daily Beast, in the article “U.S. Health Care Is Failing My Patients: From chronic conditions to mental health, our system is failing patients and doctors alike” by Farah Khan:
(1) Substance abuse, easily one of the most widespread mental health problems in this country, has yet to be adequately addressed by the current health care system. Rehab services are far and few between for patients who are addicted to drugs and alcohol.
Formally, this looks like what’s known in the speech errors business as a word reversal (Vicky Fromkin’s preferred term), word exchange (my preferred term), word metathesis, or (more colorfully) word-level spoonerism: the conventional form of the boldfaced expression is few and far between. There’s no question that such reversals or exchanges do occur as inadvertent speech errors, but there are reasons for thinking that (1) is not in fact an inadvertent error, but is more like a classical malapropism, in which the speaker or hearer produces exactly what they intended, but their production doesn’t accord with the practices of the larger community. And there’s a third possibility: that the practices of the larger community have changed to such an extent that it can no longer be claimed that (1) is clearly not in accord with them.
I’ve posted on word reversals / exchanges before (8/12/11), with an example from gay porn, in the heat of the sexual moment:
(2) You wanna fuck your shooting load! You wanna shoot your fuckin’ load! (sexual partner to Rod Barry, who’s jacking off, in one segment (from Born To Be Bad) of The Best of Rod Barry)
As is very frequent in such cases, the speaker of (2) immediately recognized his error and repeated his exhortation with the wording he’d intended in the first place. From that earlier posting:
For some other (less racy) examples, see section P (Word Reversals) [with 35 examples] of the Appendix to Vicki Fromkin’s Speech Errors as Linguistic Evidence (available on my website, here).
Contrast these inadvertent errors with classical malapropisms (CMs), which are not inadvertent. From my “Classical Malapropisms” (Language Sciences 1.2.339-48 (1979)):
My colleague who said
like social psychology, behest with all these experiments [beset → behest]
maintained, when questioned, that behest was indeed what she intended to say, and she did not see what the problem was; and the person who exhibited the substitution
cholesterol → chlorester oil
did so throughout a discussion of diet, and repeated chlorester oil clearly in answer to a direct question about the name of the substance.
It’s characteristic of CMs that the target expressions (the ones common to the larger speech community) are infrequent and/or technical.
Side observation: the target in (1), few and far between ‘scarce, infrequent’ (literally ‘not many and rare’, with a kind of emphatic redundancy), has come up on this blog once before, on 8/25/11, on the expanded version few and far in between, which, I noted, was not in the OED, but common enough that
many people on the net have worried about which version of the idiom — few and far between or few and far in between — is correct.
I also noted that there was a truncated version, few and far — a nonce truncation, with missing material to be supplied from context. The truncated version will have a role to play in what follows.
Now back to few and far between. OED2 has it from 1668 on, and a Google search pulls up a gigantic number, which reduces to 393 (not at all shabby) when duplicates are removed.
Predicative few, standing alone, is reasonably common, and entirely standard (indeed, a bit on the formal side). Predicative far, standing alone, is rare, however; far apart, far from, and far away [from some reference point] are common enough, but predicative far between is not, though it would probably be understood. So predicative few and far between counts as an idiom — not easily decomposable.
[It does have alliteration on its side. So it was suited, truncated but in the order far and few (required by the rhyme scheme), for the refrain of Edward Lear’s nonsense verse “The Jumblies”: Far and few, far and few, / Are the lands where the Jumblies live; / Their heads are green, and their hands are blue, / And they went to sea in a sieve.]
As an idiom, it’s learned as a whole, probably without much appreciation of its parts, which means that when some of those parts are syntactically and phonologically very similar, as few and far are, and when in addition both truncated far and few and few and far occur, the way is clear for some speakers to try the non-conventional order of those parts; after all, we don’t expect idioms to make a lot of sense in their fine details, so why not? All it would take is for some speakers to produce the other order, either as an inadvertent error or by misremembering (CM-stye) the details of the idiom, or by creatively varying the order, and these speakers can then serve as the focus for the spread of far and few between. Once this version spreads, we have a core of new speakers who just think that it’s the way the idiom works, or that it’s one of two equally acceptable versions of the idiom (since they’re probably hearing both). For them, far and few between is not some kind of error.
We’d expect spread to come first in informal contexts where there is relatively little normative pressure, and then eventually to find its way to contexts where more care and more editing are likely (contexts like the one for (1)). And this is what we see: Google Ngram (up to 2000) has only tiny amounts of far and few between, hugely dwarfed by few and far between, but a Google net search turns up substantial numbers of the “reversed” variant, comparable in size to the figure for the older variant — 438 hits with duplicates removed, actually a bit bigger than the corresponding figure (398) for the older variant. So it looks like we’re on the way to coexisting variants, roughly like the current situation for have another think coming (the older variant) and have another thing coming (the innovation).