Re: Optimum number of symbols

On Tue, 21 May 2002 04:11:44 -0400, Nik Taylor <fortytwo@...> wrote:

>"Mike S." wrote:
>> It's inaccurate to say that English "h" is "not a very productive"
>> phonetic marker;  it is not a phonetic marker at all.
>
>I admit I phrased it poorly.  What I meant was that it's not completely
>arbitrary.  It's used as an (unpredictable) marker of similar sounds.
>You can't predict that {th} will be /T/, but you can be pretty sure it
>won't be /x/.
> You're right about its usage in English not being a true
>phonetic marker.  Altho, historically, at least, h-digraphs had just
>three significances in English: fricative (th, ph, gh - which originally
>indicated /x/) or alveopalatal (sh, ch) or voicelessness (wh).  Phonetic
>changes, of course, have ruined the significance of gh and wh.
>
>In some languages, of course, it's more reliable.  Latin used it to
>indicate aspiration in Greek borrowings, while Irish Gaelic uses it to
>indicate a set of phonetic changes.
The digraphs are much more irregular than regular.  Let's take
a slightly closer look at the English h:

1. In <ch> of <chin>, h can be said to mark +postalveolar, +affricative.

2. In <gh> of <nigh>, h (originally) marked +fricative, -voice.

3. In <ph> of <photo>, h can be said to mark +labiodental, +fricative.

4. In <sh> of <shin>, h can be said to mark +postalveolar.

5. In <th> of <thin>, h can be said to mark +dental, +fricative.

6. In <th> of <this>, h can be said to mark +dental, +fricative, +voice.

7. In <wh> of <when>, h can be said to mark -voice.

In no two digraphs does <h> mark the same combination of features
twice.  The digraphs are sufficiently unpredictable enough, both
now and historically, that the orthography would be better served
by separate graphemes.  Digraphs are an eyesore.

The best thing you can say is that the first character roughly
marks the place or articulation (all 7).  (What else would you expect
though?) One could make a feeble argument that <h> vaguely suggests
+fricative at least (4 out of 7); considering that Latin was not
rich in fricatives to begin with, it is not surprising that this
is where the patches are usually needed. But notice for other
features, it could mean virtually opposite things, e.g. + or -voice.

The history behind <h> is that it appeared in Greek transliterations
into Latin in the forms ph, th and ch.  From there the idea caught
on to use it in a digraph whenever a patch was needed.  It's too
bad that more letters were adopted instead; at least this resulted
in the Western alphabets being somewhat alike, graphically.

Regards

Reply