Friends of Semantic Compositions

January 2009

Sun Mon Tue Wed Thu Fri Sat
        1 2 3
4 5 6 7 8 9 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31

Site Statistics

Blog powered by Typepad

« Soft & Dry | Main | Coloratura donnas »

September 13, 2004


language hat

If you hadn't explained the "coloratura bass" thing in your next entry, I would not have had the faintest idea what you might be referring to. I wonder whether you picked up the association from someone else or made it up your own self?


Your floccinaucinihilipilificatory bass may be ... well, off-base.

My Hebrew is not of scholarly quality; my knowledge is that of a sabra's son with linguistic leanings. I take it the word in question is /lishmoa'/, the infinitive of "hear". There should be no hiatus between the /o/ and the /a/ in this word. If your colleague wants to insert a glottal stop between them, well, he's just wrong. The /oa/ sequence here has a smooth transition. And as the rounding on /o/ gives way to the open-lipped /a/, any English-speaker will perceive a [w]. By all means avoid any extra rounding in the transition, but don't put in a glottal stop.

The /a/ just before the final ayin is an interesting phenomenon. I don't know what Hebrew grammarians call it, but it's inserted by phonological rule before any of a small class of final consonants: ayin and alef (two kinds of glottal stop), /H/ (chet, but NOT /x/ chaf), and /h/. For example, the word /sameaH/ "happy", exhibits this rule: we would expect an adjective of this kind to have the form /sameH/. The feminine is /smeHa/, after all. Compare /Samen/, /Smena/ "fat". But this rule puts in an /a/, so we get /sameaH/. Similarly, /liSmoa'/ belongs to a large class of strong verbs of the general form /liXYoZ/, where X, Y, and Z are arbitrary consonants; but its membership in that class is obscured by the inserted /a/.

Semantic Compositions

Your Hebrew is definitely better than mine; my knowledge is limited to what I picked up in a few years of Sunday school and Bar Mitzvah training, and so I too readily deferred to the other choir member's knowledge (he claims to be a Hebrew speaker, but I haven't had occasion to hear him converse in it).

That /a/-insertion is interesting. In learning about autosegmental phonology, Arabic (and sometimes Hebrew) is presented as being a template-based system, where clusters of consonants fill in patterns of empty consonant slots. The /liXYoZ/ pattern you indicate would seem to indicate that the consonant cluster is just /sm/, and so based purely on the theory, I might expect the word to be /liSmoma/. But there's no second /m/ in there, which I guess might be due to a rule-ordering issue.


I have to start this clarification by admitting that I don't remember how the root for "hear" is spelled. I know it starts shin-mem (/Sm/) but I can't remember if the final consonant is an ayin or an alef. For our purposes it doesn't matter. Semantic Compositions, you either have the text before you or you could put your hands on it quickly: I would appreciate knowing if it's an ayin or an alef. But let's call it an ayin.

So the root is {Sm6} where that {6} represents an ayin. Fitting it into the /liXYoZ/ pattern gives /liSmo6/. Now /6/ is one of the phonemes that triggers final a-epenthesis, yielding /liSmoa6/. The /m/ doesn't have to be echoed, because there is already a consonant in that position. The final /6/ is probably not pronounced in most dialects -- I know my mom didn't pronounce it, but my Yemenite tutor did. So the epenthetic a is just left dangling, something like [liSmoa].

Heinlein loved the Big F-Word. It's in The Number of the Beast too, in closer to the sense I meant it: a hairsplitter, a perfectionist, a nitpicker. In fact it's possible I learned the word from Heinlein, in which case we should probably go back to the OED and make sure I'm using it right.

The comments to this entry are closed.