R A Brown, On 28/07/2011 06:53:
> On 27/07/2011 20:45, And Rosta wrote:
>> R A Brown, On 27/07/2011 08:14:
>>> On 26/07/2011 20:17, Peter Cyrus wrote:
>>>> The idea of a phoneme is an abstraction of phone :
>>> A _set_ of phones, surely?
>>> David Crystal writes: "The notion of a phoneme allowed
>>> linguists to group together sets of phonetically similar
>>> phones as VARIANTS, or 'members', of the same underlying
>> (I) Some people see phonemes as a set whose members are its
>> (II) Some people see phonemes as
>> abstractions/generalizations/genericizations of the
>> allophones (so the allophone is a subtype or instance of the
>> (III) Some people see phonemes as a kind of phonetic
>> blueprint (whose level of abstractness or specificity would
>> be subject to debate).
>> (IV)Other people (including me), see phonemes as completely
>> abstract units of lexical contrast, associated with separate
>> phonetic blueprints by realization/interpretation rules
>> (analogous to the rules that relate words to their meanings).
>> I suspect that many conlangers (such as Peters Cyrus and
>> Bleakley) incline to View (II).
> I don't see (I) to (IV) as being mutually exclusive.
It's true that as formulated, (I) sounds like a more specific variant of (IV) (-- from a Type (IV) perspective, the specificity of the (I) variant would be needless/vacuous). But I had in mind another distinction: when the phono structure of a word is specified in the lexicogrammar, does the structure consist of members of phonemes or of phonemes themselves. In specifying the phono form of _cat_, does it begin with phoneme /k/, or with a member of phoneme /k/? If it begins with phoneme /k/, then it would count as what I meant (IV) to cover. If it begins with a member of phoneme /k/, then it would count as what I meant (I) to cover.