Let's talk for a moment about how attitudes shape understanding.Chris Servanti wrote:I have seen this in buth's work especially with the upsilon. The only problem I have with that is that there is no way to know this kind of authentic pronunciation for even a few words, let alone the entire language
The type of attitude that produces statements like "maximum differentiation between letters helps students", also produces the idea that every word must / should have its own meaning (λόγος/ῥῆμα), every extant grammatical form should have a differentiable class or function (-θην/-όμην), that things spoken can not go beyond what is stated (ἡμῶν/ὑμῶν).
But does it really help students to think in those terms? Evidently not.
On the basis that in those cases of the working of the language, where it is not helpful for students to make maximum differentiation, I question whether applying the same maximalistic attitude to pronunciation really helps students either.
Chris Servanti wrote:authentic pronunciation
In Buth's terms, the way to know how vowels were confused in spelling (and by implication in pronunciation) for "even a few words", is by looking at the evidence from the papyri and inscriptions, which are autographs containing mistakes to a larger or lesser degree. Some of the evidence that has led him to those conclusions is set out in the preface to his edition of a small selection of Epictetus (ISBN 965-7352-07-X). There is nothing in what he says that has not been noticed by others, but what he has done is to spell out the system of pronunciation that must needs be existed for the errors to have been made (systematically). Besides those spelling errors, adopting that system doesn't seem to interfere with understanding.
Let me say, (from myself), that there is a big difference between listening to a text in which all the words, forms and grammatical (morpho-syntactic) structures are familiar, and listening to a text in a completely foreign language. The more familiar one is with a language, the more comprehensible a listening text becomes to understand. Beginners start out learning a language with no skill in the grammar of that language. If they want to express themselves, they unsyntactically string a few words together. When they listen, they may recognise a familiar word or two and assume a meaning (basically by contextually guessing).
Continuing with my thoughts, listening is usually not dictation, it is comprehension. Dictation is itself a useful application of listening skills, but it involves recognising what has been said and writing it down - listening and productive use of the exact same language. Most people don't walk around with notebooks, writing down what others say, then reading it to be able to make sense of it. We write down what we hear after we have understood it, so that we can remember it better or more accurately.
Further, as listening develops in a language, more of what is said is recognised. If that is forced into the reproduction model in the early stages, it seems that a lot of progress is being made, but later there is an inability to differentiate between the most important things that have been heard and those of lesser importance. Comprehension breaks down when input is too quick. If listening develops as ever more accurate (grammatically directed) guessing, then at the higher levels, ranking the importance of the various elements is a lot more natural.
Another point is to ask whether the text comes before the spoken or the spoken before the text. Spelling is a convention for what is spoken, capturing some of the speech act (spoken or thought) and recording it. Having a pronunciation system that is exactly the same as the written, might obfuscate the directionality of presidence.