tetrahedron Posted March 7, 2013 Report Posted March 7, 2013 Hi folks. This is my first foray into the forum- usually I spend my time torturing the physics and mathematics crowd with my easily provable Pascal Triangle math theory of electronic and nuclear (magic) numbers. But I've spent decades studying hundreds of languages with regard to phonosemantic iconicity (often erroneously termed 'sound symbolism'). Saussurean 'arbitrariness of the sign' has often been taken uncritically by linguists and other students of language as received wisdom. While there is much that has arbitrary character within language, not all of it is. Iconicity is underrated by too many. In fact, there is plenty of evidence that the degree of any language's complement of iconicity, symbolicity, and indexicality, from a Peircean perspective is highly variable. With regard to phonological iconicity at the word level, nothing beats ideophones. Some language (such as Cherokee or Inuktitut) have almost none. At the other extreme can be languages like Gbaya in Central Africa, where everyone knows thousands yet inventories change radically from person to person, village to village, so that the 'language' may have hundreds of thousands to millions overall, as compared to nearly identical inventories of normal words. I discovered some time back that there is a correlation with morphosyntactic type- the less a language uses derivational or grammatical morphology (either productive or fused), the more ideophones it will tend to have. Thus polysynthetic languages have the fewest, and isolating or agglutinative langauges have the most. Classical 'inflective' languages are intermediate. Attitudes against use of ideophones have been effective in killing them in many westernized societies (esp. among educated elites and city dwellers- this has happened to Zulu, for ex.). Why should numbers of ideophones vary by language type? My guess has been that it has mostly to do with the deployment of resources in the language. The languages with the most tend to have normal word roots that are rather poor at detailing manner of action, patterns of movement and other specialized semantic types, are kind of 'matter of fact'. Richly morphologized languages handle this with the morphology. This includes those that aren't like that on the surface (many isolating and/or analytical languages have fused 'roots' which were ancient compounds or derivations which carried manner specification). It also turns out that if you examine the properties of ideophones they stand in contrast with those of normal 'grams', on many fronts, as if they were the evil opposite 'twins' of morphology. And there is some evidence that they involve right hemisphere processing- another opposition. But where things get really interesting is their mapping of form to meaning, and that has bearing on both the origins and evolution of language. I've looked at, as mentioned above, hundreds of languages (grammars and dictionaries) to get a sense of the variation inherent in ideophones and other types of word-level iconicity, over a period now of 30 years. What I've found is that there are only a handful of types of mappings that examined living languages have, as well as universals. The strongest universal is that the articulatory and acoustic features form a coherent system of mappings between form and meaning. For example labial phonemes in initial position in roots have a strong tendency, when the labial is etymologically original, to connote swelling and release (pressure, fluids, gasses, solid inclusions, etc.), whereas in final position the opposite state is developed, involving capture and holding. WIth initial velars we have the opposite- capture is the norm, while final velars involve release and clearance. Both labials and velars, in keeping with their 'grave' feature, possess the notion of breadth versus narrowness, container-ness. Another difference is that labials tend to connote voluptuousness, while velars have a shell-like, dry quality to them. Initials in the dental/alveolar region (have to be careful here because sometimes one or other articulation is actually derived) often connote the notion of narrow focused impact- in final position focused stretching and withdrawing. Some of you may already have noticed a certain Freudian character to this kind of system. And palatals (vs. derived palatized) phonemes in initial position already have that pulling/stretching notion, while in final position, yes you guessed it, deal more with narrow focused impact. Curiously, for these 'acute' featured phonemes the whole voluptuous/dry opposition is mapped the other way round- that is the palatals are softer while the dentals are harder. Why should this sort of mapping be nearly universal, if words are arbitrarily assigned meaning? This is a kind of diagrammatical iconicity, where the organization of interrelationships between meanings parallels the organization of interrelationships between phonological forms (down to the featural level). I've thought about this for many years and the only conclusion I can come up with is because this mapping was already present in the original, now exapted, oral function- that is mastication and deglutition. Velars are articulated next to the molars, which handle hard, often dry foodstuffs. Lips are next to the incisors, which in primates anyway tend to deal with removing pips, stems, etc. from otherwise ripe and soft foods requiring minimum chewing (at least those portions of the food). In other apes the space between the lips and teeth is used to separate, by suction, liquids from wads of chewed materials, the fibrous part then expelled by spitting or letting it fall from the mouth. Thus the original oral functions have some parallel in the articulatory parts of meaning in iconic usage. For the acute phonemes we are instead dealing with the canines and bicuspid teeth. At least for the canines there is no problem in seeing how they associate with piercing/punching holes, again parallel to the iconic meaning. Also note the tip of the tongue and similar functions there. Less clear is the parallelism for the palatal articulation zone with regard to original masticatory/deglutition function, though the surface of the tongue is often a kind of separation board for sorting materials by texture and unit size- maybe the notion of 'stretching' belongs here on some level? The articulation of syllables and the neurology associated with them has already been shown by professionals to utilize the same machinery as that of chewing and swallowing, so the idea that meanings might have been sucked into language by association can't be too far fetched, can it? Nobody is claiming that ALL of language is iconic, but it is also known that iconicity tends to be the true precursor for symbolicity. Ideophones are in their ideal, prototypical state rather shy about syntactic entanglements- being quotable only. Thus something very similar to them may have preceded the evolution of language-as-(we think)-we-know-it. The morphosyntactic cycle, taking thousands of years, may show that normal lexemes, and grams even more so, are unstable in some way to formal change and loss/replacement for a variety of reasons. Yet at the same time the fact that ideophone root structure and semantics share a universal core that gives the forms meanings gives us a way to renew language structure when the other parts get a bit shaky. The original stem cells, as it were, are not lost, but are available when needed after the worn out overlayer is sloughed off. But this is just a stab, hopefully there may be some interest in general discussion here. Jess Tauber cal 1 Quote
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.