[Lexicog] Corpus Conundrum #2 Stress and Intonation

[bolstar1]

Corpus Conundrum #2  (Stress and Intonation) 

       Stress and intonation – these are two aspects of phonetics 
that are given short shrift in dictionaries (other than for single-
word entries). Again, the space-efficiency problem rears its ugly 
head, for what lexicographer doesn't always fight with space-
efficiency and publishers' requirements?  
       In phrasal analysis (particularly as it relates to semantics) 
stress and intonation play a much larger role than single-word 
pronunciation. This is true for two reasons.  1) words generally 
don't lose their semantic values in their derivatives/inflections 
(though they do change in stress or intonation). Otherwise they 
become distinct lexemes, warranting separate head words or sub-
entries under their root lexemes.  2) In isolation, words require a 
knowledge of stress and intonation. However, in phrasal and sentence 
use, much more stress (no pun  intended) is placed on one or two 
individual words, not parts of words, and little or no stress on 
multisyllabic words. Of course, in English semantics derives also 
from word placement (less on inflections than other languages), which 
(perhaps (?) puts more stress-emphasis/intonation-emphasis on 
particular words, too. I would be happy to get a response on this 
point in the post.  
       But phrases do change in meaning according to targeted word 
stress and intonation. Here I'm talking about created phrases -- not 
idiomatic,  proverbial, fixed phrases and expressions. Even by virtue 
of what created-phrases are -- two or more uniquely-created, 
collocated elements – their meaning include more than one semantic 
value. Combined with stress and intonation they have a potential for 
great semantic variation. With fixed-phrases this is much less so, as 
in the expression/lexeme mentioned here lately "kick the bucket." 
Three words in fixed collocation – meaning "die." Certainly the 
object of the preposition gets the stress. This is so regardless of 
it being uttered in isolation (just as in single words), or in 
sentence pattern. A synonym phrase for "kick the bucket" – also 
involving an object – "buy the farm" also stresses the object, in 
this case "farm." But stress doesn't change in either phrase. And if 
their was a change, it would be simply poetic licence. 
       This space problem speaks to the value of digital 
lexicography, especially in the forms of online reference works, 
portable machinery, and electronic paper. With those tools, one can 
then simply drill deeper and deeper into main entries to get more 
searcher-pertinent information – akin to target-marketing in the 
business world. 
     The current-era emphases in linguistics (including lexicography) 
are phraseology & corpus analysis, and because of this, more 
attention to the areas of phonetics (acoustic phonetics) & phrasal 
semantics will probably ensue (undoubtedly is ensuing). I don't speak 
from experience as to the latest activities and findings here (which 
is part of what I'm looking for in responses to this post).   
       
       
       An example of this point, regarding created-phrase forms, not 
fixed, is the simple sentence "I did not say you took her science 
book." This nine-word sentence is quite a straightforward example – 
there are no punctuation marks. Tense -- simple past tense. Voice – 
active voice. Mood – indicative mood. With convoluted, punctuation-
rich, idiomatic/non-literal/allusional prose, the story changes. But 
the semanticvariations are obvious, depending on which word is 
stressed. 

I did not say you took her science book. (someone else said it)
I DID not say you took her science book. (denial)
I did NOT say you took her science book. (strong denial of saying it)
I did not SAY you took her science book. (I (may have) implied it, 
not 
                                           said it.)
I did not say YOU took her science book. (I wasn't talking about you.)
I did not say you TOOK her science book. (You obtained it by some 
other 
                                           means. || You did 
something 
                                           else with it.)
I did not say you took HER science book. (You took someone else's   
                                           book.)
I did not say you took her SCIENCE book. (You took another kind of 
                                           book.)
I did not say you took her science BOOK. (You took another science 
                                           object.)

One more example that includes both stress and intonation variance. 
The basic structure is even simpler than the last one, with three 
words (two, if you count the abbreviated form as a word). But this 
simpler grammar example becomes perhaps a more complex token of 
vernacular speech – at least for transcription-based corpora. The 
example is "That's fine." 

#1 THAT'S fine. -- stress on "THAT'S" || flat intonation on "fine" = 
i.e. That's okay./Go right ahead./Don't mention it./No problem. (used 
when someone excuses himself for slight bothersome action, or 
questioning someone to see if preceding action/choice is acceptable)

#2 THAT'S FINE! -- stress on THAT'S/possible descending tone 
on "that's" || stress on "fine"/intensified descending intonation 
on "fine" = (strong sarcasm & resignation) i.e. "I strongly dislike 
that!" or "This situation is seriously bothersome to me!" 

#3 That's FINE! Stress on "fine" || high intonation, with clear 
descending intonation also on "fine") = i.e. "Your work/news/product 
is good quality." or "I'm happy about what you did/what just 
happened."

#4 That's FINE! – low stress/but high initial intonation on "That's" 
|| high stress/intensified descending  || = (strong sarcasm & non-
resignation to the fact) i.e. Do it; I'm leaving!" or "If you choose 
that, I will do something to counter that or get back at you for that 
(leaving, taking other action) 
  
       It is true that variations will occur in these examples, 
depending on personal stylistics, region, degree of 
sarcasm/intent/etc., but there are patterns that can be detected that 
indicate nuances in meaning. Intonation would seem to be a more 
refined, nuanced inflector of meaning than the more simply-measured 
stress points. 
       Perhaps only line-by-line identifying/tagging/contextualizing 
of transcribed audio records (including context) are needed. Perhaps 
also with written text. Machines alone don't seem sophisticated 
enough. The state of the current art relating to stress/intonation 
and semantics is another question I had for anyone with experience in 
dealing with it. 

(Addendum: Rudy, it feels right subsuming "infix" under the general 
type "affix" in relation to those "stuck-in lexemes." And fun example 
of "a whole `nother." Sorry to hear about McMillan.)

Scott Nelson







 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/lexicographylist/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/lexicographylist/join
    (Yahoo! ID required)

<*> To change settings via email:
    mailto:lexicographylist-digest at yahoogroups.com 
    mailto:lexicographylist-fullfeatured at yahoogroups.com

<*> To unsubscribe from this group, send an email to:
    lexicographylist-unsubscribe at yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/