Re: E and e (was: A break in the evils of English (or, Sturnan is beautiful))

At 10:23 pm +1000 30/4/02, Tristan wrote:
>On Tue, 2002-04-30 at 05:07, Raymond Brown wrote:
>> In English there is contoversy over whether we have:
>> /ij/ ~ /i/
>> /ej/ ~ /e/
>> _or_
>> /i/ ~ /I/
>> /e/ ~ /E/
>>
>> But versions that give /eI/ ~ /E/ are, to my mind, perverse and confounding
>> phonemic transcription with phonetic notation [eI] ~ [E] in which, [E]
>> means "a sound close to, but slightly retracted from, cardinal vowel [E]".
>
>Why? Aren't phonemic representations related to how the people hear the
>sound?
Partly - and most (all?) non-liguist anglophones would surely consider the
vowels in 'gate' and 'get' as two different sounds, i.e. the /get/ ~ /gEt/
phonematizations.

>In which case, using /ei/ ~ /E/ is perfectly normal and to be
>expected: the two vowels are completely unrelated except for a few
>historical orthographical oddities.
Quite so, which is why your 'average anglophone' is more likely to think of
[eI] as 'some sort of _a_' sound, rather than connect it with [E].

But the phonemic theory (and _theory_ it is) is primarily concerned with
contrasts and distribution of sounds.  To have a _phoneme_ /e/ which occurs
only in one position, i.e. the diphthong /eI/ is very suspect.  Indeed, it
must imply a contrast with /EI/ which, I'm sure you'll admit, is wrong.
Those English dialects that use [EI], do not use [eI].  As I understand the
phonemic theory, one would say that [e:], [E:], [e@], [I@], [eI], [EI] (and
probably one or two others) are all dialect allophones of the phoneme /ej/
(or /e/, depending upon your phonemic analysis).


I don't get this lax/tense version
>of the other sound argument, it just sounds like linguists have just
>pulled something out of their collective arses
I am not a professional linguist (and certainly not a "university
linguist); this type of argument is not helpful.

The terms lax/tens(d) or retracted/advanced are used to denote features
actually found in some languages and to reflect what the tingue is actually
doing.  If you do not subscribe to this idea, then how do you exlain the
type vowel-harmony found. e.g. in Igbo, Efik and some other African
languages?

>(why is French pardoned here?).
It's not pardoned.  The plain fact is tho, e.g. French _peine_ and English
_pen_ are likely to be both transcribed as /pEn/, the vowels in the two
languages are not the same.  The English sound is both higher and more
retracted than the French.

>A relaxed (which is what 'lax' suggests to me) version of /ei/
>yields either /ei/ or /@/.
This is confusing phonetic and phonemic notation.  A 'relaxed' version of a
phoneme is still the same phoneme, unless it leads to loss of phonemic
distinction as when unstressed vowels often become just [@] in English,
except that unstressed /i/, /I/, /e/ and /E/ tend toward to a sound closer
to [I] than to [@].

[snip]
>
>I must say, I especially hate anything that doesn't show the length
>difference between /i:/ and /I/. Redundancy is a good thing, especially
>when writing a dictionary. (Goes out and curses the Macquarie...)
Maybe so.  But phonemically /i:/ ~ /I/ surely must suggest two components
of contrast: hight of tongue & length, i.e. it implies four phonemes /i/ ~
/i:/ ~ /I/ ~ /I:/, which is not correct.  While phonetically the difference
may be [i:] ~ [I] (tho some English dialects tend to diphthongize the first
as [ij] or [Ij]), phonemically we have, as I see it, to choose between /ij/
~ /i/, /i/ ~ /I/ or even, possibly, /i:/ ~ /i/.

Of course, it may be that the phonemic theory is incorrect - but that's
another story  ;)

Ray.



======================
   XRICTOC ANECTH
======================

Replies