Recency
« previous post | next post »
Language Log reader Jukka Kohonen has written to me about the Recency Illusion, the (often inaccurate) belief that a usage you have recently noticed is in fact a recent development in the language. Kohonen wondered whether anyone had studied its causes (and effects) systematically, and he had a specific instance in mind. I had to admit to a profound ignorance on the subject, and to considerable worries about how the topic could be studied systematically.
Some background notes.
First: some usages are genuinely recent, so that the judgment that they're recent is accurate. There was a time, not long ago, when blog was brand-new (the OED entry has a first cite in 1999, and a first cite for weblog in 1993). People who remarked that they hadn't heard it until recently were right.
Second: after some period of time, usages that were once genuinely innovative become commonplace. and the perception that they are recent becomes less and less accurate. Blog is shading off into this territory. We're entitled to raise an eyebrow at someone who complains about blog because of its recency. (Note: if you find blog unpleasant on phonological grounds, then certainly you shouldn't use it; perhaps you'll use online diary instead. In any case, you're entitled to your tastes. Maybe you feel that blog is phonologically too close to blob — though the innocent block is equally close phonologically — but I'm not here to dispute your tastes, only to note that many other people don't share them, and it's unseemly to be petulant about that.)
Third, looking at the other end of the time scale: "words" (identified by spelling or pronunciation) sometimes can be tracked back for quite some time, but these early occurrences turn out to be entirely irrelevant, because they are not instances of these "words" with the meanings and uses we're talking about nowadays. Old occurrences of truthy and truthiness in texts (the OED has several 19th-century cites) have nothing to do with recent uses of these words — because these older occurrences are of different lexemes, meaning 'truthful' and 'truthfulness'. The (genuinely) recent usage involves the derivational suffix -y of "(disparaging) approximation" (as I put it here).
Fourth: it's not uncommon for a word to be innovated with the same, or very similar, meaning again and again, by different people at different times in different places (similarly for syntactic constructions) — but without "catching on" in a larger community of users. The significant event for these larger communities is when such usages start to "take off", and that's when people start to notice them.
The thing is, oftentimes, people who notice them long after they've spread think they're novel. That's the Recency Illusion, often accompanied by other opinions about the usage (for instance, that it's very frequent).
(Further note: the Recency Illusion arises, of course, for all sorts of social and cultural practices, not just language use. It's easy enough to observe specific cases, but (again) hard to understand what's going on in general.)
Now back to Jukka Kohonen, who wrote me first about a particular case in English — N times more/larger/… than (which I'll call the Times-er construction), rather than N times as many/large/… as — and, later, about a Finnish counterpart, N kertaa suurempi (about which he wrote, he reports, a short piece in the Finnish linguistics journal Virittäjä in 2006 (110:409-16), "Havaintoja kertaa-komparatiivista").
Kohonen notes that MWDEU has a discussion of the English construction (in its entry for times), with examples from 1551 on; Kohonen has corresponding Finnish and Latin examples from the 17th century. Nevertheless, he writes
I keep encountering a persistent belief that it has only recently appeared (or that its meaning has changed) as some kind of modern corruption
and hopes
to better understand what exactly makes people feel so certain that such constructions are recent.
My guess is that we're looking at a piece of folk-linguistic belief, the idea that non-standard (or merely informal or merely spoken) variants (or variants that are merely perceived to be one of these) are either recent or local — because otherwise they would have been eliminated by pressures for correctness. This places a touching faith in the engines of correctness. In fact, these engines have little effect on how most people speak and write.
Indeed, many non-standard variants are both venerable and extraordinarily widespread, clustering together in what you might think of as General Non-Standard variety (see my Local Color Illusion posting, on the "curious grammar of Ohio"). Most people don't appreciate the age and geographical/social spread of linguistic variants, and that's not the result of some ideology they hold to: it's just what follows from the fact that they have no way to get a synoptic view of the historical, geographical, and social distribution of the variants. All they have to go on is their experience, which is a tiny and biased (and not always easily interpretable) sampling of these things.
The upshot of all this is that people in general are going to be inclined towards the Recency Illusion with respect to non-standard, informal, and spoken variants, and people with considerable education (with their faith in the engines of correctness, instilled by that very education) might well be especially subject to the illusion.
All of this is dauntingly hard to research. It would be easy enough to ask people what words or constructions they think have come into the language recently. But then I'm stumped. You certainly can't ask people WHY they think this, at least not if you expect to get a useful answer; in general, people simply can't access the sources, in their personal histories, of their beliefs.
To study the phenomenon systematically, you'd need to look at a number of individual people, separately, somehow tapping their personal experiences with variants and somehow taking into account their attitudes about variation and its sources.
On a question that's a lot easier to research — what people say about the Times-er construction. Korhonen reports the following reactions (from different people, obviously):
– recency (it's a recent development);
– antiquity (it's obsolete);
– in-group character (for Finnish Times-er, "Finns saying that it's a Finnish oddity");
– out-group character (e.g., "Finns angry about the invasion of such an anglicism");
– colloquiality ("the phrase is only used in 'sloppy everyday talk' (while in scientific works it is either absent or has a different meaning)").
That last reaction incorporates one criticism of Times-er, namely that it is "illogical" or "irrational": X times more than Y MUST MEAN 'Y plus X-times-Y (that is, 'X+1 times Y'), not 'X times as many/great as Y' (that is, 'X times Y'). (In the most common variant of this reaction, X times more than Y is disparaged because it is said to be ambiguous, with both the 'X times Y' and 'X+1 times Y' interpretations.)
The appeal here is to the idea that ordinary-language expressions are simply realizations of logical (or arithmetical) formulas. This is just backwards. The formulas are there to represent the meanings of expressions; they are not the prior reality, merely cloaked in (those devilishly vague) words of actual languages.
There's nothing illogical about "We don't need no steenkin' badges". If you're asked to translate this sentence in Logic 101, the only right answer is something equivalent to standard English "We don't need any [supply epithet] badges".
Similarly, English and, or, and if have any number of uses that that don't just translate into logical connectives. And English assertions with bare plurals ("Linguists love to eat") don't translate directly into universal quantification in logic, since they aren't falsified by some number of exceptions to the generalization. Not only are the uses of such expressions much more complex than their simple "translations into logic" would suggest, but English (like every language) has any number of idioms and special constructions, each with their own interpretations.
So we swing back to Times-er. The than here is the default marker of comparison, so it's no surprise it's gotten used here.
A further complexity: in addition to the N times more/larger than usage, there is also a N times less/lower than '1/Nth as much as' usage, which has come under criticism separately from the more than usage. The MWDEU times entry treats them both, but leads with the less than usage, with a quotation from Swift:
… but now I am resolved to drink ten times less than before –Jonathan Swift, Journal to Stella, 30 July 1711
(and then Gladstone (1879), the Chicago Sunday Tribune (1946), and more). MWDEU remarks, a propos of recency, that times
has now been used in such constructions for about 300 years, and there is no evidence to suggest that it has ever been misunderstood.
But the critics roll on. Here's lexicographer Laurence Urdang on ADS-L on 9/4/07:
From a letter in today's Daily Telegraph: ". . . the murder rate in London is five times lower than some cities in the United States . . ." I am not interested in the source or in the sense but in this typical use of times that has sprung up in the past couple of decades (according to my observation) with the meaning 'one nth': in the present instance, in my dialect (!) I should have said, ". . . is one fifth (of) that in some cities." I cannot conceive how or why times, which is an indication of multiplication, not division, has come to mean its opposite. Am I the only English speaker on earth who has noticed this or is bothered by it? I have never seen another comment on it. A typical context would be, "The average temperature at the Antarctic is five times lower than [that] at the Arctic." [Forget about the truth of the statement, for grammar and truth are unrelated.] In other words, instead of using the appropriate fraction or percentage indicated, 'one quarter of' becomes "four times lower than," 'one third of' becomes "three times less than," etc.
A particularly lovely instance of the Recency Illusion (along with a shocking failure to consult scholarly resources). Among the follow-ups was this one from James Harbeck, an editor and writer, also on 9/4/07:
… in the style guide I created and enforce where I work, "times less than" and "times lower than" are explicitly anathematized as being without clear meaning or mathematical value. (The style sheet also recommends avoiding "times more" and "times higher than" because they're ambiguous; we aim for percentage figures when possible. But it's an uphill battle sometimes.)
A personal note: I know the disparagement of Times-er from long ago, from my grade school years, I think; I was taught that two times more than X really means 'three times as many as X'. Since authority figures insisted on this interpretation, I avoided the construction entirely (as, as far as I know, I still do). Yet I've never stopped asking, "Why don't you understand the clear meaning of what people are saying?"