LINGUIST List 6.29

Sat 14 Jan 1995

Sum: Semiotics, Syllable appendectomies

Editor for this issue: <>


Directory

  1. , Semiotics Summary
  2. , Summary re Syllable appendectomies

Message 1: Semiotics Summary

Date: Mon, 09 Jan 1995 15:57:51 Semiotics Summary
From: <D.W.Saltderby.ac.uk>
Subject: Semiotics Summary


Semiotics and Linguistics

Late last year I posted a query on whether anyone had a view on the
relationship betweeen semiotics and linguistics and whether some kind
of semiotic analysis can be used in parsing.Thanks to the select few
who responded - either its a boring question or we are on to
something new here!

Chris Nelson gave a very clear summary of the basics of semiotics as
follows:
)The original statement on the relationship between semiology (the
)study of signs which was initiated in the modern era on the
)continent) and linguistics was made by the founder of
)semiology--Ferdinand de Saussure.
)See Chapter 3, section 3 (Place of language in human facts:
)semiology) in his _Course in General Linguistics_ in which he places
)linguisitics within the realm of semiology. Though Suassure saw
)semiology as encompassing linguistics because language was only one
)kind of social sign system and semiotics studies all social sign
)systems, later semiologists expanded the scope (or rather,
)rediscovered the ancient scope) of semiology by recognizing that
)there are natural, non-arbitrarily meaningful signs (like smoke,
)which is taken as a natural sign of fire--"where there's
)smoke ..."). Thus, semiologists after Saussure have come to see
)linguistics as a science dealing with but one kind of
)non-natural/conventional/arbitrarily meaningful sign system, while
)semiology is a general science of sign systems, natural and
)non-natural. Roughly at the time Saussure was re-creating
)semiology on the continent, Charles Sanders Pierce (pronounced
)"purse") was arguing for the recognition of a science of signs he
)termed semiotics. From the start, Pierce's notion of a science
)of signs encompassed both natural and non-natural sign systems, of
)which language is clearly only one kind of non-natural system.
)However, I don't know of any place off-hand where he
)spoke specifically about the relation between semiotics and
)linguistics as sciences. Given that semiotics/ semiology is the
)science of signs, and covers the study of all sign systems, and that
)language is but one sign system, I would be surprised if anyone
)suggested semiotics/semiology was either entirely separate from or
)subordinate to linguistics!

Jacques Steyn gave some similar information and the following:
)In the States, however, with the popularity of behaviorism and then
)generativism, language received an autonomous status, and the focus
)of how language was defined became all the more narrower. For
)example, the focus first on morphology and phonology and then on
)syntax, decontextualised language to such an extent that anything of
)semiotic interest was ignored.
)the answer to your question will depend on which school linguists
)who will reply belong to. If they decontextualise language and
)regard it as an autonomous system, they are very likely going
)to ignore anything related to semiotics. If, however, language (and
)here your definition of language is important) is considered as one
)of the tools of communication it will probably be regarded as a
)subsystem of semiotics. Some may even regard any mode of
)communication as language (as we can see eg in computerese), and
)thereby making these terms synonymous.

On a related topic Margaret Winters told me:

)My husband Geoffrey Nathan and I have a paper in the festschrift for
)J.D. McCawley on the relationship between linguistics and philology.
)One of our discoveries in surveying over 100 linguists on the
)subject was that the field was divided. We got in approximately
)equal numbers the (usually strongly voiced) opinions that
)a)linguistics subsumes philology, b) philology subsumes linguistics,
)c) they are the same thing,and d) they are entirely different.

Karen Ward expressed interest in an expanded discussion as she has
come up against semiotics in relation to computing. I also have come
across the use of semiotics in computing although to
my mind the computing references I have seen are not very rigorously
researched or necessarily convincing. I am still continuing with my
research into semiotics and if anyone else is still interested I
would like to hear from them. In particular I am looking at language
used in negotiations and I intend to try a semiotic analysis in
tandem with syntactic/semantic analysis.

The references I have been given are as follows:-
R Antilla - "Historical and Comparitive Linguistics"
W.U.Dressler - "Leitmotifs in Natural Morphology"
S Shaumanyan - "A Semiotic Theory of Language" Indiana U. Press 1987
T Sebeok - "A Sign is Just a Sign" Indiana U Press 1991
F de Saussure "Course in Linguistics"

Many thanks to Peter Coughlan, Jeff Allen, aromi, Chris Nelson,
Jacques Steyn, MargaretWinters, Karen Ward

Sam Salt
University of Derby
(d.w.saltderby.ac.uk)
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue

Message 2: Summary re Syllable appendectomies

Date: Wed, 11 Jan 1995 17:50 -05Summary re Syllable appendectomies
From: <Mike_Maxwellsil.org>
Subject: Summary re Syllable appendectomies

Several weeks ago I posted a query to LINGUIST re the notion of syllable
appendices. This is the summary of the replies.

The original query concerned s+stop clusters that occur stem-initially in
English ("stop", "spit" etc.). Under the widespread assumptions that (1)
sonority should monotonically increase toward the syllable nucleus, (2)
these s+stop are clusters, (3) both members of the cluster belong to the
same syllable, and (4) fricatives are more sonorant than stops, there is an
evident problem.

One solution that has been proposed in similar cases is that, on a
language-specific basis, certain consonants at the margin of a word may be
exempted from the requirement that all "segments" in a word must be
syllabified. Such extra-syllabic consonants are sometimes said to belong
to an "appendix" of the word.

My specific query concerned a different approach, which I had read of in
the book "English phonology: An introduction", by Heinz J. Giegerich.
(Giegerich credits the idea to O. Fujimura 1979 "An analysis of English
syllables as cores and affixes", Zeitschrift fur Phonetik,
Sprachwissenschaft und Kommunikationsfurschung 32:471-6. I have not seen
this reference.) Under this approach, a syllable (rather than a word) may
have an appendix, in which consonants specified on a language-specific
basis may appear. As I had not seen this idea before, and since it seemed
to me to be more ad hoc than the word appendix theory, I decided to ask how
accepted the approach was.

Thanks to the respondents: Anthony Shore (anthcruzio.com), Geoffrey S.
Nathan (geoffnsaluki-mail.siu.edu), And Rosta (ucleaarucl.ac.uk),
Wolfgang Lipp (i52023sakura.kudpc.kyoto-u.ac.jp), Markus Walther
(walthersapir.ling.uni-duesseldorf.de), John Harris
(johnhlinguistics.ucl.ac.uk), Ellen Broselow
(BROSELOWDatalab2.sbs.sunysb.edu), Don Churma (00dgchurmabsuvc.bsu.edu),
and Todd Bailey (toddturtle.psych.umn.edu).

Several respondents agreed that the notion of appendices to syllables was
somehow ad hoc. Some more specific comments:

Todd Bailey suggested that the word-internal facts of syllabification are
far from clear, and provided a reference disputing the syllabification of
e.g. "constraint" as con.straint. If correct, this might be taken as
evidence that word-initial s+stop clusters are not tautosyllabic either,
with the /s/ in this case appearing in a word-level appendix (rather than
in the seemingly more ad hoc syllable-level appendix). The reference:

 Treiman, R., J. Gross, & A. Cwikiel-Glavin. 1992. "The syllabification
 of /s/ clusters in English." Journal of Phonetics, 20:383-402.

Don Churma questioned whether word appendices solved anything either, or if
they just swept the dirt under a different rug.

And Rosta reminded me that in Government Phonology, a consonant that cannot
be syllabified in the usual way is taken to be the onset, or (presumably in
the word- initial case) the coda of a syllable with an empty nucleus. This
nucleus remains empty (unpronounced) in English by a language-specific
stipulation, while in other languages (such as Spanish), it is filled by
the language's default vowel. (Hence Spanish provides an overt
manifestation of the "extra" syllable, which in English remains covert.)

Wolfgang Lipp suggested that syllabicity might be a "squish" (Ross's term,
not Lipp's), and that the "s" in the English word "squish" (to take an
example) might be a little syllabic (as originally suggested by Hooper,
although she seemed to regard syllabicity as an either/or sort of thing).

Anthony Shore had independently developed a similar notion to Giegerich's
and Fujimura's, in his thesis:

 Shore, Anthony (1990) _Latin and Moraic Theory_. University of
 California at Santa Cruz, Santa Cruz, California.

Shore was trying to account for syllabification of codas (e.g. for the
Latin word reks 'king'), but the problem is similar to that of onsets. His
account relies on a distinction between a coda consonant that associates
with the syllable by way of a mora node, and a coda consonant that
associates directly with the syllable node.

Markus Walther suggested that fricatives and stops do not, in fact, differ
in sonority, in which case /ski/ would not violate a requirement that
sonority increase *or remain level* as you approach the nucleus. A 25 page
summary of his German thesis can be found under

 ftp://ftp.cogsci.ed.ac.uk/pub/phonology/papers/wp-8-4-walther.ps.Z.

John Harris gives an appendix-free treatment of sC clusters
and final coronals in ch 2 of his book "English Sound Structure" (1994,
Blackwell). I haven't had a chance to look at this yet.

Ellen Broselow sent a copy of her unpublished paper "The Structure of
Fricative-Stop Onsets", in which she suggests that s+stop onsets can
sometimes be reanalyzed as adjunction structures, like the following (I'm
simplifying a bit):

 Root
 / |
 Root k
 |
 s

("Root" is, of course, a Root node in a segmental feature structure; it
would in turn be dominated by the syllable's nodes.)

Thanks for all the responses. It reminded me of how difficult syllables
can be to work with. Maybe Chomsky and Halle were right after all!
Mail to author|Respond to list|Read more issues|LINGUIST home page|Top of issue