Re: / / vs [ ]

On 4 Jan 02, at 13:05, Clint Jackson Baker wrote:

> I think I've been confusing when to use // vs. [].
AFAIK // is for phonemic rendition (where you indicate phonemes) and []
is for phonetic rendition (where you render phones). I've also seen <>
used to indicate orthographic conventions.

Basically, [] is for more or less strict rendition of the sounds
produced, while // is only concerned with the phonemes of the given
language (which sounds actually are considered different by speakers of
the language).

Say, for example, that you have a language where a certain consonsant
is [g] between vowels (intervocalically?) and [k] elsewhere, and which
has [g] only between vowels and only with this consonant. Then you
might say that your language has a phoneme /k/ which is realised as [k]
or [g] depending on the environment, but there is no /g/, since this
language has no phoneme /g/ that is separate from /k/ (for example,
there are no minimal pairs and the two phonemes are in complementary
distribution[1]).

So if you want to talk about a certain word, you might write it
phonemically as /taka/ or phonetically as [taga]. The information that
the phoneme /k/ is realised as [g] in this environment is present only
in the phonetic rendition (in square brackets), but not in the phonemic
rendition (in slashes) because it's not a separate phoneme.

Similarly, in English, /k/ can be either [k] or [k'] (that is,
aspirated), but they're the same phoneme. You can write /kin/ and
/skin/ even when many people pronounce [k'in] and [skin]. (Then there's
also "broad" and "narrow" phonetic representation, depending on how
much detail you go into in phonetic representation.) Or the /l/ which
is different in <lick> and <bell> ("dark" and "light" <l>, but I'm not
sure which is which). Most English speakers would probably not consider
those two k's or those two l's as being different, which makes sense,
since they're the same phoneme.

Hope that clears things up a bit... I'm not sure how well I explained
things.

Cheers,
Philip

[1] However, the fact that two phonemes are in complementary
distribution does not necessarily mean that they are identical; for
example, in English, /N/ and /h/ are in complementary distribution --
one can only occur syllable-finally and one only syllable-initially --
but I don't think anyone considers them allophones of the same phoneme!
--
Philip Newton <Philip.Newton@...>