Sunday, 13 March 2011

Plato's archetypes (Ideas) and the notion of the "phoneme" in linguistics

There is in linguistics a notion of a "psychologically ideal" sound segment (the phoneme) and its variations in speech called, allophones.

The idea of "phonemes" and "allophones" is somewhat technical in nature but totally necessary in linguistic analysis and in the construction of workable writing systems (orthography). As we have seen in the "standardization" discourse in Nunavut, most Inuit communities want the writing system to best reflect their own unique dialects and claim that there are sounds (phones, in linguistics speak) unique to their dialect that do not exist in the writing systems (whether syllabics or roman).

Years ago, when I worked at the Language Bureau in Yellowknife, NWT, as an Inuktitut linguist trainee I cut my linguistics teeth analysing and writing a short paper on an allophonic variation in North Baffin on the [l], which changes to English type [r] in certain contexts, much like the stereotypical oriental tendency to change r's for l's and vice versa.

For eg, [iqaluit] becomes [iqaruit]; [ulu] becomes [uru]; [aalasi] becomes [aarasi]; but [ili] remains [ili]; [illu] remains [illu].

Before I continue explaining the linguistics analysis above, I want to quote the best piece of writing on Plato's archetypal forms that I've come across so far. It was written by Richard Tarnas in his book, The Passion of the Western Mind: understanding the ideas that have shaped our world view:

(according to Plato): "Nothing in this world is, because everything is always in a state of becoming. But one thing does enjoy real being, as distinguished from merely becoming, and this is the Idea... ...Any particular thing in the world is actually a complexly determined appearance. [...] Plato's world, therefore, is dynamic only in that all phenomenal reality is in a state of constant becoming and perishing, a movement governed by the shifting participation of Ideas."

As appearances is to Ideas in Plato's world, so are allophones to phonemes. What I mean is that the North Baffin variation of [l] lays behind the verbal relationship to the "psychologically real" phoneme, the [l] in this case, to the variations {l} and {r}.

The context in which the variation occurs is governed by the quality of the preceding vowel we call "backness" (as in opposition to "frontness") in its place of articulation: [a] and [u] are back vowels, and [i] is not a back vowel.

But the change in the segment [l] can only occur after [a] or [u] if and only if  it is a single [l], because everywhere else, it is expressed as [l] unchanged: [alla] remains [alla]; [ullu] remains [ullu]; [illiq] remains [illiq].

But since the variation is allophonic, we (the native speakers) do not normally perceive the change from [l] to [r], and in our script this superficial variation is not reflexed as a real and distinct segment where the [l] in all contexts remains [l] regardless of where it occurs and regardless of whether its syllabics or roman.

The linguistic equation for the allophonic variation above is very concise and elegant and something I'm proud to have seen as a phonological fact of the Inuit Language. But I cannot reproduce the equation with the formatting constraints of this blog.


No comments:

Post a Comment