lexical/phrasal stress in alphabetisms

[Martin Haspelmath]

Larry,

Yes, the connection to other lists such as phone numbers seems 
plausible. On the other hand, phone numbers are just lists, whereas 
alphabetisms are words. And the fact that in Swiss German, alphabetisms 
are often not stressed on the last letter, seems to support my 
conjecture (French borrowings also get initial stress in Swiss German). 
Also, the fact that contrastive stress can be lexicalized (as in German 
'PKW, which is stressed on the first syllable also when it does not 
occur next to 'LKW) seems to show that we're not dealing with phrasal 
expressions synchronically. (Or maybe there are also phrases with 
lexicalized contrastive stress?)

Anyway, we clearly need more data from more languages, though it will be 
hard to find languages whose alphabetisms are not influenced by French 
or English. (Are there alphabetisms in Georgian, Armenian, or the 
languages of south and Southeast Asia that use a non-Latin script?)

Martin

Larry M. Hyman wrote:
> Martin - If your conjecture re French is correct, it would presumably 
> be PHRASAL stress. On the other hand, if each letter is a phonological 
> word, then this just the "nuclear stress rule" of SPE, no? In other 
> words, this is not "lexical stress". Larry
>
> ----
>
> David - Sounds interesting. Wish I had data from other languages for 
> you. It's actually not on the last syllable, but rather on the 
> stressed syllable of the last "letter": cf. WWW (which is harder to 
> say that World Wide Web), which has antepenultimate stress on the last W.
>
> I've tried to think of counterexamples, but your rule works, e.g. 
> LSMFT ("Lucky Strike Means Fine Tobacco" ... a TV ad in the US from 
> the 1950s), or E-G-B-D-F*, which was not originally an acronym but has 
> been made into one ("Every good bird does fly") etc. The same works 
> for other "lists". E.g. numbers: My zip code here at Berkeley is 
> 9-4-7-2-0*, for instance (I'd also stress the last number of my social 
> security number). The same happens when you mix letters and numbers: 
> E.g. my father's amateur radio call was W-6-V-B-D (which he would 
> repeat as "W-6-Victor-Baker-Dog"*, with phrasal stress on Dog*).  In 
> any such "listing" of elements each one is treated as a separate 
> prosodic word, so it would be natural to assign  English 
> "pitch-accent" to the last prosodic word (letter, number etc.).
>
> Larry
>
>
>
>
>
>
>
>
>> Dear all,
>>> I'm interested in patterns of lexical stress in a specific type of 
>>> acronym -- let's call them letter-by-letter acronyms -- whose 
>>> pronunciation consists of each letter bearing its own individual 
>>> name, eg. English US [yu:es], LFG [elefji:], etc.  (Not all 
>>> languages have letter-by-letter acronyms, for example Hebrew does not.)
>>>
>>> In two languages that I'm familiar with, English and Papuan Malay, 
>>> word stress is commonly or predominantly penultimate; however, 
>>> letter-by-letter acronyms invariably place the stress on the last 
>>> syllable, eg. [yu:ES], [elefJI:].  Is this a coincidence, or is 
>>> there a general principle at play here?  (One might perhaps wish to 
>>> argue that the final stress is phrasal rather than lexical, but in 
>>> other respects these acronyms behave like single words.)
>> This is a very interesting observation that had never occurred to me. 
>> In two other languages that I'm familiar with, German and Russian, 
>> the same holds true: Lexical stress in alphabetisms is always on the 
>> last syllable, although other words may have stress elsewhere (and 
>> prefer to have stress elsewhere, especially in German). I disagree 
>> with Wolfgang Schulze's claim that stress may also be initial in 
>> German. I cannot say 'FDP (Freie Demokratische Partei), except 
>> contrastively (e.g. die 'SPD und die 'FDP). (But it is true that the 
>> exact conditions for contrastive use are unclear. These two 
>> abbreviations also contrast with respect to their last letter, though 
>> less saliently. And one can also say, in coordination,  'CDU und 
>> 'CSU, although these two do not contrast in theeir first letter at all.)
>>
>> My guess would be that this stress pattern represents a borrowing 
>> from French, which was a dominant language when this kind of 
>> abbreviation became widespread.
>>
>> Incidentally, the term "alphabetism" is relatively well-established 
>> for what David has in mind, and there is no need for the ad-hoc term 
>> "letter-by-letter acronym" (see 
>> http://urts120.uni-trier.de/glottopedia/index.php/Alphabetism).
>>
>> Martin
>>
>> P.S. On the topic of "pseudo-partitive" (a cup of tea) and "true 
>> partitive" (a cup of the tea): Doesn't someone have a better pair of 
>> terms for these? "Pseudo-partitives" are of course much more common, 
>> so it's really strange to have only a "pseudo-" term for them.
>>
>> -- 
>> Martin Haspelmath (haspelmath at eva.mpg.de)
>> Max-Planck-Institut fuer evolutionaere Anthropologie, Deutscher Platz 6
>> D-04103 Leipzig     Tel. (MPI) +49-341-3550 307, (priv.) +49-341-980 
>> 1616
>>
>> Glottopedia - the free encyclopedia of linguistics
>> (http://www.glottopedia.org)
>
>


-- 
Martin Haspelmath (haspelmath at eva.mpg.de)
Max-Planck-Institut fuer evolutionaere Anthropologie, Deutscher Platz 6	
D-04103 Leipzig      
Tel. (MPI) +49-341-3550 307, (priv.) +49-341-980 1616

Glottopedia - the free encyclopedia of linguistics
(http://www.glottopedia.org)