I've done a little more research. The German phrase Carlos Piera cites translates to "The Law of Lengthening Limbs." This is the tendency to order list-items in ascending order by syllable length:
and the American way
Roman Jakobson mentions this briefly in "Closing Statement: Linguistics and Poetics." He says that you are more likely to say "Joan and Margery" and "Margery and Joan," associating this tendency with the "poetic function" of language. He doesn't say why, just that it sounds better. Piera is another one, citing the German guy who coined the phrase, so there are at least four or five people who have noticed this, including me.
It is probably weaker than a law. It is more of a tendency. Let's say it will happen more often than not, in the weakest formulation. The implications are not profound, maybe, but I find this kind of thing fascinating. As far as I know, there has been little discussion of word length in linguistic or literary prosody.
The nuclear stress rule (Chomsky and Halle) says that you stress the right-most stressable element in a phrase. We know that important elements tend to go last, and that ascending order is preferred in the rhetorical figure of gradation. Curiously, most dictionary definitions don't talk about word-length in relation to figures like "climax" or "gradation."
One of the striking features of "In Memory of Major Robert Gregory" is the repeated effect of an anticlimactic stumble into a short ending (perfectly apt, of course). If Behaghel is right, that's indeed distinctive.
For once, English Wikipedia is more useful.
Thanks! I could have found that I guess. It's more useful than the German especially since I don't know German very well. Also known as the law of Pannini, so it is an actual "thing."
In the Linguistics Department library where I studied as an undergrad, there was a framed black-and-white photo of a mustachioed man captioned "Panini".
It's dawning on me that since I work at the company that can do this, I'm actually in a good position to do some corpus linguistics. Panini was a great pioneer of language study, but I'm from Missouri on this law.
Excellent. I did one for bait and tackle vs. tackle and bait. I think the skepticism of the Missouri state motto is justified. By now we know that it is a thing, attested by RJ, JM, Panini, Piera and Behaghel (by intuition at least) and that nobody has claimed a preference in the other direction. So the question is how strong it is, whether it beats random chance, and whether it holds for fixed expressions and titles or for lists generally.
And Cicero said: quare aut paria esse debent posteriora superioribus et extrema primis aut, quod etiam est melius et iuc…
Vide Ford/Chevrolet. As with salt/pepper, it does seem that preferences are clear, but I don't see that they're so predictable as these laws would suggest.
Find me one that goes in the opposite direction. The only one that occurs to me is peaches and cream.
Watson and Crick is clearly preferred (but that's the order of their names on the paper).
"Beatles and Stones" dominates the alternative, and that's not due to any fixed model like the author list.
wealth and happiness
bacon and eggs
There may be a preference in some cases for that trochee + iamb pattern of
bacon and eggs, beatles and stones, peaches and cream, Starsky and Hutch
At least that pattern in not unknown. I saw coffee and tea though ngram prefers the opposite.
I say coffee and tea.
I'm sure I've seen this somewhere in a style manual or usage book, but I can't remember where (but not Strunk and White, says the Pullum disciple).
And I've definitely used the idea of Behaghel's "Law of Increasing Terms" to comment on student writing. — With the caveat that breaking the law can create a rhetorical effect, of course.
Post a Comment