Received: by alpheratz.cpm.aca.mmu.ac.uk id WAA14981 (8.6.9/5.3[ref pg@gmsl.co.uk] for cpm.aca.mmu.ac.uk from fmb-majordomo@mmu.ac.uk); Sun, 14 Jan 2001 22:28:36 GMT From: <Zylogy@aol.com> Message-ID: <2d.60e04d2.27938162@aol.com> Date: Sun, 14 Jan 2001 17:25:38 EST Subject: Re: Sound symbolism and language To: memetics@mmu.ac.uk CC: Zylogy@aol.com Content-Type: text/plain; charset="US-ASCII" Content-Transfer-Encoding: 7bit X-Mailer: AOL 5.0 for Windows sub 129 Sender: fmb-majordomo@mmu.ac.uk Precedence: bulk Reply-To: memetics@mmu.ac.uk
To Mark Mills: Motivation is the long-used umbrella term for phonosemantic
iconicity as well as many other phenomena in language deemed nonarbitrary.
And nomenclatural issues continue to rear their ugly heads in this area all
the time. Peircean usage, for instance, is still a bit wobbly in linguistic
parlance- is it phonosemantic iconicity, or sound symbolism (which of course
also leads us to questions of levels- surfacy phonetics (sounds) versus
underlying system (phonology)).
What I meant about parallel organizational principles is that the parsing of
terms into hierarchically salient chunks sees both the signals and their
content semantically running side by side. A fully inflected sentence is a
whole and has parts, and the parts are wholes and have parts, and on down.
Some of the meaning can be construed from the parts. Some from the wholes.
And some can only be derived from the effect of the larger structures in
relation to the smaller ones (sentences in context, for instance). Of course
there's also meaning NOT explicitly marked but implied from the nonlinguistic
context, speaker assumptions about listener knowledge, etc. But at the level
of the lexical "root" such parallel organization is considered by current
linguistic theory to halt dead in the water.
A root *?am- meaning "eat", for instance (and similar forms attested in
numerous languages worldwide (and one of the "etyma" often cited as evidence
for genetic relationship)), will retain the sense of eat/ingest, etc. on up
through derivation, inflection, and higher levels of phrasal, clausal,
textual organization. Metaphorical extension may apply, and historical change
may transform the etymological root into something of quite different
appearance and glommed together with other elements, and perhaps also quite
shifted meaning. Still, the etymological root is strongly associated with the
meaning.
Ok, then, what about a root such as *kam, meaning "hold", also attested from
numbers of languages, etc. Can we ascribe a meaning to the partial **-am, or
to the vowel or final C2 **-m? (* here means reconstructed root, either
synchronically, from language internal data, or diachronically from a number
of related languages).
In many languages root final -m is associated quite strongly with notions of
containment, coverage, apprehension, and possession. The connection is, from
a low-resolution perspective which simply lumps all terms with final -m
together (and only in roots, thus no -m from grammatical affixes etc.),
statistical only. You can increase the resolution by starting to segregate
terms by type (nouns from verbs- as mentioned in the previous post verbs tend
to be more iconically organized than nouns, and the reason for this is that
in many languages nouns are historically more derivative than verbs (often
coming FROM verbs), so old, often historically hidden derivational morphology
obscures the original root much more often than with verbs).
I'll dig out some of my sets of examples.
As for Zipf's law, I have noted something worth mentioning here: The words
used most frequently in the vocabulary of any language tend to be the most
historically stable- and thus also tend to be the ones most thoroughly
changed by a long history. Heavy usage apparently makes up for low internal
iconic transparency, and by and large the less changed terms appear to be of
relatively low frequency. But there is a split in the system between terms
which linguists call heads (modified) and dependents (modifiers). Headlike
terms (such as verbs versus nouns) have a tendency to grammaticalize- its
long been known that all grammatical affixational morphology derives from
freestanding full words- not just any set of words, mind you, but those with
a relatively broad (and thus underspecified) meaning capable of being used in
the widest variety of contexts. One of the reasons heads need to be
accompanied by modifiers (dependents).
Words which enter chains of grammatization (and any language has forms in
different stages of semantic and phonetic reduction as well as loss of
independence (and thus attachment to other forms) tend to turn over much more
rapidly, as a set, than those with modifier potential, which tend to
accumulate in the historically derived lexicon (often considered by linguists
to be the "trash bin" of language history, containing all the unpredictable
stuff).
Interestingly, its the LOW specificity dependent/modifier words and the HIGH
specificity head/modified words which tend to be the iconically transparent
ones. There are theoretical hierarchies linguists have established to explain
such oppositions, often used in typological work (there being very different
types of language with different associated structural tendencies). Hopefully
something can be gleaned from these which will explain also the phonosemantic
iconic trends.
In the meantime, here are some of the findings of the work I'm doing: In
phonological systems p, t, ch, k tend to be the most common stop consonants.
Roman Jakobson (one of the leading linguistic lights of the 20th century)
describes the set of phonological distinctive features needed to produce
these (defined in acoustic spectral terms as well as articulatory) as grave
versus acute- p and k are grave, t and ch acute- in as one polar opposition,
and then compact versus diffuse - k and ch are compact, p and t diffuse. Note
that compactness/diffuseness associates with front or back of the oral cavity
and gravity/acuteness with height in the oral cavity.
Phonosemantically gravity associates, in verb roots in many languages in the
C1 position, with broad/decentralizing distributions of force or mass, while
acute phones associate, in the same context, narrow/centralizing
distributions of same. Compact phonemes associate with retraction,
ingathering (towards agent away from patient) while diffuse phonemes
associate with extention, out-releasing of same, in the same context. In C2
some of the semantic features are reversed, and there are languages which
reverse the entire system of position of meanings within roots, along either
or both axes. In these latter language this appears to be a very specialized
occurrance, one requiring historical explanation. More later.
Jess Tauber
zylogy@aol.com
===============================================================
This was distributed via the memetics list associated with the
Journal of Memetics - Evolutionary Models of Information Transmission
For information about the journal and the list (e.g. unsubscribing)
see: http://www.cpm.mmu.ac.uk/jom-emit
This archive was generated by hypermail 2b29 : Sun Jan 14 2001 - 22:30:08 GMT