Conference thebay::joyoflex

	For phonetic funnies, there's always English. The way some words 
	should be pronunced in that wide-spread language is unpredictable.

	For example, the sound of  i  in the word "life" is totally different
	from the sound associated with the same letter in the verb "to live".

	Note that the  i  in "to live" is completely different from the one
	in  "alive".
	
	Or perhaps it's the other way round :)

	roger		

    Re .1
    
    Roger, never mind the I in 'life' and 'to live', consider the I
    sound in 'live' and 'live', i.e.
    
    I live in a house - short I
    
    This electric socket is live - long I
    
    And as George Bernard Shaw showed 'ghoti' can be pronounced 'fish':
    
    gh as in cough
    o  as in women
    ti as in nation
    
    :-)  Nick

    There is a formal description of discrepancies between written and
    spoken language, the DECtalk exception tables.

    Rather large for English and smallish for French, Italian, Spanish etc.

    DECtalk lacks certain capabilities, an integral part of local
    ( south France) "O la la " is a waving movement with the palm of right 
    hand . Without it, you really understand that you talk to somebody with
    a heavy foreign accent.

re: meaning changed by intonation

In Russian if you say, "Ya piSHOO," you are writing; if you say, "Ya PIshoo,"
you are pissing. Of course, there sometimes is no difference...

-Perry

   Um, well, the latter should be PIsayu.  There is no 'sh' sound in the
   Russian word for pissing.


   							Terza

English has words that change meaning depending on the accent.

PROcess - what you do to make cheese-whiz

proCESS - what you do at graduation

Wook

    miNUTE	tiny
    MINute	60 seconds

    	I think at my graduation I precessed
    
    "sway in such a way that its axis describes a circle" (Oxford
    Illustrated Dictionary).
    

    "minute" , is there only change in intonation ?

    Isn't "u" in 60sec a short one and in 'tiny' a long one ?


    				 Rydz Szczypaczki

    In 60 seconds, the first syllable is a short "I" and the second is
    a schwa (unstressed vowel sound). The accent is on the first syllable.
    
    In tiny, the first syllable is a long "i", and the second is a
    long "u". The accent is on the second syllable.
              

    The directory of all available vowel sounds is published in the
    Schwa Catalogue.

The Schwa Catalogue is, no doubt, edited by Perry and Vanna White. 
Great Caesar's Ghost.

Dave C.

>>The Schwa Catalogue is, no doubt, edited by Perry and Vanna White. 
>>Great Caesar's Ghost.
    
    
    Vanna sure knows how to turn a phrase...

    I think this might the right note. Moderator, please move if not.
    
    One way to resolve ambiguous phonetics is to use a phonetic Alphabet. 
    For example in English one might say on the phone:
    
    	"DEC has headquarters in MAYNARD, thats M in Mike, A in Alpha, Y as
    	in Yankee, N in November ...
    
    The whole "Amercian" phonetic alphabet that I learned in the Navy
    (there are others) was:
    
    	Alpha, Bravo, Charlie, Delta, Echo,Foxtrot, Golf, Hotel, India, Juliet
    	Kilo, Lima, Mike, November, Oscar, Papa, Quebec (KAY_BEK)Romeo, 
    	Sierra, Tango, 	Uniform, Victor, Whisky, X-ray, Yankee, Zulu.
    
    I recently ran across a German version (note the added need to
    distinguish CH and SCH as well as umlauted letters)
    
	"Buchstabier-Tafel": 
    
        Anton, �rger, Berta, C�sar, CHarlotte, Dora, Emil, Friedrich,
        Gustav, Heinrich, Ida, Julius, Kaufman,Ludwig, Martha, Nordpol,
        Otto, �konom, Paula, Quelle, Richard, Samuel, SCHule, Theodor,
        Ulrich, �bermut, Viktor, Wilhelm, Xanthippe, Ypsilon, Zacharias
    
    My question. Anybody know others? What about non-European ones for 
    Hebrew, Arabic? Is there a strategy for languages like Japanese or 
    Chinese? Is there a different one for "British"?
    
    Tango.Hotel.Alpha.November.Kilo.Sierra
    
    -Neil

    I imagine `British' means the Queen's English as it's spoken in the
    United Kingdom.
    
    There was one such alphabet (used for messages in the Royal Flying
    Corps?) that was similar to yours. I don't know it all, but it
    started Able Baker Charlie and had `Sugar' for S - which kind of
    defeats the object.
    
    When I worked (temporarily) in Vehicle Registration (which involved
    discussing license-plate numbers over the telephone) we were encouraged
    to use `the International version' - whatever that is: maybe the
    phonetic alphabet used in contexts where English is the official
    language. Any pilots know better?
    
    The version referred to in reverential tones by my superiors
    as `International' was very like yours (.-1), but for some
    reason `November' was replaced by `No-one' (I've no idea why;
    `November' does the same job just as well if not better). 
    
    I always used `November' - except when I was making up something
    whimsical. Inspired by a colleague who used a Biblical system (Abel,
    Balthazar, Colossans [did Paul write to them?], Deuteronomy, Ezechiel
    and so on) I used to coin new systems as long as they worked. This was
    OK, until I took the lateral step of thinking of license-plate numbers
    as words [there's a note about this somewhere] and gave a number as
    `ELK as in moose'. The policeman at the other end sounded as if he was
    looking blank, so I reverted to Echo Lima Kilo, and never strayed into
    the realms of whimsy and incomprehension thereafter.
    
    b

    Is there a list for the "silent phonetic alphabet" anywhere in this
    conference?  As in:
    
    			A as in boat
    			B as in doubt
    			C as in scene
    And so on...
    
    					andrew

The Alpha, Bravo, Charlie phonetic alphabet was in while I was in the army in
the early 60's.  It had recently replaced the Able, Baker, Charlie alphabet
which was in vogue at least as far back as WWII.  I think the changeover 
occurred at about the same time that combat boots went from brown to black and 
distances went from yards and milies to meters and klicks.  This was about 
1955-1958.

It seems to me that the military phoetic alphabet has changed again - I noticed 
some new letter mnemonics last year during the Gulf War.  Are there any recent
veterans or reservists out there to confirm? 


Larry

>        Anton, �rger, Berta, C�sar, CHarlotte, Dora, Emil, Friedrich,
>        Gustav, Heinrich, Ida, Julius, Kaufman,Ludwig, Martha, Nordpol,
>        Otto, �konom, Paula, Quelle, Richard, Samuel, SCHule, Theodor,
>        Ulrich, �bermut, Viktor, Wilhelm, Xanthippe, Ypsilon, Zacharias
    
	This is interesting, as it sounds mostly like first names -- and
	is thus rather different from the English.  What do the non-name
	ones mean?  Here are my guesses:

	Arger -- something about arguments?
	Kaufman -- merchant (alos a last name)
	Okonom -- something about economics?
	Quelle -- which?
	Schule -- school
	Ubermut -- ???
	Ypsilon -- the letter

	Viel dank,

		-- Cliff

    
    re .-1
    
    
    >    This is interesting, as it sounds mostly like first names --and
    >    is thus rather different from the English.  What do the non-name
    >        ones mean?  Here are my guesses:
    
    >        Arger -- something about arguments?     
    �rger = anger
    >        Kaufman -- merchant (alos a last name)
    Ja
    >        Okonom -- something about economics?
    Ja
    >        Quelle -- which?
    Quelle = fountain, source
    >        Schule -- school
    Ja
    >        Ubermut -- ???
    �bermut = presumption ????
    >        Ypsilon -- the letter
    Ja
    
      
    Nordpol = North pole      
          
    
    -Neil

    Trying to find a list of the Swedish version of this alphabet list,
    but no luck so far... But it's made up of only men's first names. Since
    I can never remember them, I just use any name that comes to mind.
    
    But in my telephone directory I found this list for "international
    calls":
    
    Alfred, Benjamin, Charles, David, Edward, Frederick, George, Harry, 
    Isaac, Jack, King, London, Mary, Nellie, Oliver, Peter, Queen, Robert,
    Samuel, Tommy, Uncle, Victor, William, Xray, Yellow, Zebra.
    
    Considering all the names, could this be something made up by Swedish
    Telecom just to make us feel at ease, not having to use strange words
    like Alpha and Bravo? :-)
    
    
    Hans

    G'day,
    
     re -.a_few.... and silent letters
    
    Do you mean like 'p in swimming'?
    
    
    
    ;-)
    
    
    djw

    A  aisle    (or "uh" as in "a short word", or "aesthetic" or...)
    B  bog
    C  cue
    D           (maybe "Doi", a Japanese family name or Filipino nickname)
    E  eh       (or "eye" or "ewe" or "earn" or "Europe" or "effort" or "extra")
    F  fee
    G  gnat
    H  heir
    I  isle
    J  jeer
    K  knight
    L  loam
    M  mnemonic
    N  nary
    O  oersted
    P  phi      (or "psycho" or "pneumonic")
    Q  quay     (or "queue" to go with "cue")
    R           ("rack" to go with "wrack"?  not good enough, is it...)
    S  see
    T
    U  urn      (to go with "earn" -- or "uh" to go with "a")
    V  vowel
    W  why      (or "who" -- or "wrack" to go with "rack")
    X  xylophone
    Y  you
    Z

    To answer a question in I think .14, I doubt that there's a Japanese
    version.  The Japanese alphabets (as opposed to the pirated ;-) Chinese
    characters) are 99% phonetic to begin with.  The exceptions are when a
    "ha" character is pronounced "wa" in one grammatical position, and a few
    similar cases.  In these cases, if a speaker is instructing a listener on
    what to write, the speaker will pronounce the written character instead
    of the spoken character.
    
    If a Kanji (Chinese character) needs to be described for writing, then
    the components are described by naming the components and/or by mentioning
    other Kanji that contain similar components.  If a word is being "spelled"
    for pronunciation or for writing in a phonetic alphabet, then each
    syllable is simply pronounced carefully and slowly.  The name of each
    alphabetic character is simply its sound.
    
    -- Norman Diamond

    I agree with .17's mention of WWII. `Q for the cinema' used to be
    `Q for rations.' I think the Urtext for the joke alphabet came from
    a BBC radio programme about this time: It's That Man Again, maybe.
    
    I know only bits of it:
    
    A for 'orses
    B for mutton
    C for thHighlanders [since the publication of Options for Change
    			this has become C for yourself]
    D for ential
    E for brick
    F for vescent
    .
    .
    .
    J for oranges
    .
    .
    .
    L for leather
    .
    .
    .
    O for the wings of a dove
    P for relief
    Q for rations
    .
    .
    .
    Y for pete's sake
    .
    .
    .
    
    Any more? [It'd be easy to coin some others - `U for nerve' suggests
    itself - but I've tried not to editorialize.
    
    b

    I for U
    
    U for me
    
    N fer Hell (translinguistic -- consider Fran�ais)
    
    T for 2
    

	I 	for an eye
	R	for Godfrey
	T	for two
	U	for me, and me for you
	X	for hundred

    I always liked the Stiller and Maera Version where they only used 
    two examples in their act.
    
    	A as in Aardvark
    	P as in Pneumonia
    
    Larry
    

    G'day,
    
     I remember doing this as a kid.....
     Lets raid teh memorybanks and see how many I can remember...
    
    
    A for 'orses
    B for mutton
    C for thHighlanders [since the publication of Options for Change
    			this has become C for yourself]
    D for ential
    E for brick
    F for vescent
    G for Whizz
    H for pains
    I for an eye
    J for oranges
    K for shoes
    L for leather
    M for to Bristol
    N for Nellie
    O for the wings of a dove
    P for relief
    Q for rations [Q for the loo]
    R for Askey
    S 
    T for two
    U for me
    V for 
    W for Double me
    Y for pete's sake
    Z for winds
     
      well that's most of em
    
    derek
    

X for breakfast

    Re: .21
    
    >Do you mean like 'p in swimming'?
    
    A small print hangs in my lavatory, depicting an Huck Finn-ish lad
    lazily causing circles to be made in a pond, with the resident fish
    swimming a hasty retreat.  The caption reads:
    
    "Ne buvais jamais d'eau!"
      (Never drink water!)
    
    /Ed ;^)

    Re .30:
>    "Ne buvais jamais d'eau!"
    That should be: "Ne buvEZ jamais d'eau!"
                           --
    		Denis.

    Speakin' of phonetic funnies (at least to the Anglophone (more
    properly, "Ameriphone"(??)) ear), what's the deal with the initial
    phoneme of the last name of the current Secretary General of the United
    Nations?
    
    When American media announcers say it, it starts with some minor
    variant of a "G".  But when I hear the beeb announcers, it seems to
    start about 2 inches below the uvula...  And I have to wipe off the
    loudspeaker grille afterwards.
    
    I won't risk telling you of my difficulties when I'm catching the news
    via earphones.
    
    PS -- And besides "James James Morrison Morrison" are there any other
    Americans or Westerners with the same two first & second names?  (imho
    JJMM doesn't count because his last name's doubled too...)

>    When American media announcers say it, it starts with some minor
>    variant of a "G".  But when I hear the beeb announcers, it seems to
>    start about 2 inches below the uvula...  And I have to wipe off the
>    loudspeaker grille afterwards.
 
My guess is that in Arabic the "gh" is a voiced velar fricative, or 
something like that.
   
>    PS -- And besides "James James Morrison Morrison" are there any other
>    Americans or Westerners with the same two first & second names?  (imho
>    JJMM doesn't count because his last name's doubled too...)

Well, there's Major Major Major of Catch-22 fame.  There was also a 
baseball player named William (Billy) Williams.  That's the best I can 
do.

>    PS -- And besides "James James Morrison Morrison" are there any other
>    Americans or Westerners with the same two first & second names?  (imho
>    JJMM doesn't count because his last name's doubled too...)

Actually his full name is James James Morrison Morrison Weatherby George
Dupree, so I'd say he does count. In case it matters.

re .34:

>    PS -- And besides "James James Morrison Morrison" are there any other
>    Americans or Westerners with the same two first & second names?

I can't parse this.  Do you mean X X, X X X, X X-Y, X-X Y-Y, or some other
combination?

    Re .34
    
    James James Morrison Morrison Wetherby George Dupree was English, wasn't
    he?
    
    Or was it an American character that A A Milne picked up for one of his
    poems?
    
    Nick

  >> Well, there's Major Major Major of Catch-22 fame.  There was also a 
  >> baseball player named William (Billy) Williams.  That's the best I can 
  >> do.

	Along these lines, and although he's not famous, I have a friend
	who knows someone named Russell Russell Russell.
	His father's name was Russell David Russell, by the way.

	Everyone's a comedian.

	Di

    Re: .27
    
    	"I always liked the Stiller and Maera Version where they only used 
    	two examples in their act.
    
    	"A as in Aardvark
    	"P as in Pneumonia"
    
    For one thing, I think this was Nichols and May, not Stiller and Maera.
    For another, the name they were spelling was "Kaplan," so they also
    had:
    
    	K as in "knife"
    
    I don't remember what they used for "L".
    
Lew Lasher

  And who can forget "N as in Newelpost"?



  Kathrin

    Dare I ask how "Newelpost" is pronounced?
    
    Coming from Massachusetts, I can deal with Worcester, Haverhill, and
    Leominster, so lay it on me.

B as in Doubt (or Debt)

U for Ria

Wook

    Currently in an add for some ponetic reading program, there is a claim
    that English only has 44 unique sounds (phonemes?). 
    
    	a) can someone verify this? is this American? British?
    
    	b) what are the equivalents for other languages? Especially
    
    	French (which I would guess has quite a few more)
    
    	German (my guess less)
    
    	Italian (my guess much less)
    
    	Spanish (also much less)
    
    	Hawaian (the least????)
    
    -Neil

    Of course I meant "ad" not "add".
    
    -N

re: .42, .43;

I assume you also mean "phonetic" as opposed to "ponetic"   :^)

Interesting you should say that, because sitting right in front of me on my 
desk is a chart showing 55.  What I've been told is that the number of 
recognized phonemes can vary, depending on whether or not a minor variation 
is treated as a unique phoneme.

Clay

    It depends on the dialect.  The number of different vowels and
    consonants which are phonemically distinct varies in English
    from one dialect to another, so the number can vary quite a bit.
    Two examples: I don't distinguish between the "o"s in "caught"
    and "cot", but many speakers do, so they have (at least) one more
    vowel than I do; some speakers of Southern American English don't
    distinguish between the vowels of "pin" and "pen", so they have one
    less than I do.
    
    Apparently Hawaiian is the record holder for the smallest number,
    and it is in the 'teens.  I was told that Karbadian from the
    Caucausus has the largest number, in the hundreds.
    
    Phoneme = sound which distinguishes meaning, so that two different
    words can differ if the only difference between them is that one
    has one phoneme and one the other.  Linguists look for minimal pairs,
    such as the following, which is a set of minimal pairs for some of
    the English vowels:
    
    	Keyed Kid OK'ed Cad Cawed Cod Code Cud Cooed
    
    i.e. "Keyed Kid" shows that the sounds spelled "eye" and "i" carry
    meaning, and are not just different ways to say the same vowel.
    
    There's more to say, but I can't take the time right now.
    
    		-John Bishop
    cad

    There's more to say, but I can't take the time right now.
    
    		-John Bishop
    cad

>    	Keyed Kid OK'ed Cad Cawed Cod Code Cud Cooed
>    		-John Bishop
>    cad
    
    "Cad"'s already on the list, but how about "could"?
    
    Meanwhile, I thought that the long "ay" sound of "OK'ed"
    was considered to be a sequence of two phonemes, the first
    "e" of "schedule" and the "eye" of "keyed" -- even though
    English spelling permits it to be written as one letter
    (or less :-)
    
    -- Norman Diamond

    I well remember using PDP-8's equipped with filter-banks & A/D
    converters to teach English diphthong phonemes to Spanish-speakers. 
    Spanish is, as I recall, almost devoid of diphthongs.  A Spanophone
    (?sp?) really mangles a minimal pair of "beet - bait."  They render the
    diphthong as some intermediate vowel pronounced with no, or very
    little, diphthongization.  They're trained, it seems, to put the tongue
    in one of the 5-ish positions for vowelization, and NOT TO MOVE IT
    during voicing.
    
    Very frustrating...  But very instructive.
    

    Actually, the `eye' sound is two phonemes, the `a' of "ah" and the `ee'
    of "beet."  In classical Latin, the � digraph (ae, for the DEC_CRT2
    impaired among us), has this sound; the word "puellae," meaning girls,
    was pronounced poo-`ell-eye.  In Medieval Latin, the digraph's sound was
    reduced to the `ay' of "way."

    By the way the product the  company was advertising is "Hooked on
    Phonics". Anyone know what this product uses for the 44 phonemes?
    
    Other questions:
    
    	Are the following considered unique phonemes or combo's:
    
    	1. Diphthongs as discussed above.
    
    	2. Palatized and asperated consonents eg t' (Russian), BH (Hindi)
    
    	3. CH = t+sh, J = d+zh, ST, TS, etc.
    
    	4. double length vowels: aa vs a (Dutch?)
    
    	5. double length consonants: ss vs s (Italian)
    
    -Neil

    re .50 (and the "keyed" discussions before):
    
    This is actually one of the nits that linguists spend
    lots of time arguing about: is the (phonetic) fact that
    there is an addition of two features sufficient reason
    for making some sound into a phonemic sequence, or should
    it be left as a single unit.
    
    In my phonetics class we had a homework assignment of
    analyzing Moro, a language of the Sudan.  We were given 
    tapes of a native speaker saying lots of minimal pairs.
    Different students came up different ways of making phonemes
    out of the sounds.  Some liked sequences (/ndzw/) others 
    liked "pre-nasalized, post-affricated rounded /d/" all 
    as one phoneme.
    
    Other than using essentially aestheic criteria (e.g. that
    the chart looks nice, with no holes, or that if everything
    can be pre-nasalized, it's nicer to separate that out as
    /n/, but if only some things can be pre-nasalized, it's
    nicer to leave things as two different phonemes, or that
    non-segmental things like pharyngealization or tone should
    be/not be turned into segments...), there's no real way to
    figure out what the "right" segmentation is--all of our
    analyses of Moro worked.
    
    In practial terms, if you wanted to produce an orthography
    for Moro, you'd pick a phonemicization which didn't require
    lots of non-Latin symbols, and might even re-use graphs in
    combinations not used by the language to represent non-Latin
    phonemes.
    
    A good example of this is English, which uses sequences like
    "sh" and "th" for what are usually treated as single phonemes
    and not sequences.  I've seen transcription systems for toned
    languages which use vowel combinations to represent tones: a
    made-up example (I can't remember a real example) would be
    writing the various tones for /i/ and /a/ as:
    
    	ii, ai  - high tone
    	ie, ae  - high-falling
        iu, au  - low
        iy, ay  - low-rising
    
    Clearly this trick doesn't work if the language actually uses
    those vowel combinations, any more than "th" would work as
    a digraph if English also had post-aspirated "t", and thus
    "th" would also be read as "t" + "h".
    
    So in answer to .50: linguists differ.  Often the consensus
    is to go along with the orthography of the language in 
    question (so Hindi "bh" is usually treated as a separate
    phoneme and not a sequence, but double consonants in Italian
    are treated as sequences).
    
    		-John Bishop

    The NSOED uses 52 symbols to 'describe' the pronunciation of words.
    
    There are:
    	8 short vowels
    	6 long vowels
    	9 diphthongs
    	6 plosives
    	11 fricatives
    	2 affricatives
    	7 liquids and nasals
    	3 semivowels
    
    And the symbols used are those of the International Phonetic Alphabet -
    my keyboard can only accomodate about half of them, even with Compose
    Character, so there's no chance to reproduce them here.
    
    Nick

    	When I was doing speech processing for a living, some 20 years ago,
    I seem to remember that the "alphabet" had somewhere around 70 symbols
    in it. Some of these (the most often quoted example was the "click"
    from some African languages) are almost impossible for most people with
    a European native language.

    Anyone have a copy of the "ARPAbet?"  'Twas a means of rendering, in
    ascii digraphs, enough of the phonemes of English to drive the crude
    speech synthesizers of 15-20 years ago.  I last saw it in use in
    "MacinTalk," an accessory for MacOS that played a significant part in
    one of the most wonderful Desk Accessories of the Mac:  The Talking
    Moose."  I probably still have, online, a bunch of quotations from
    Samuel Jonson that I transcribed into the ARPAbet and loaded in, along
    with speech intonation markings in the same argot, into the Moose. 
    Quite comical, but so out of character (hard to imagine the good Dr. J.
    with antlers donchaknow) that I soon tired of the triumph of hope over
    experience.
    
    There were the occasional opportunities for punning in the ARPAbet,
    where the orthography that produced a given utterance was actually
    antithetical thereto when read literally.  Can't remember any of 'em. 
    I wonder whether that is anything akin to punning in American Sign
    Language? 
    

    Drive a Spanish speaker nuts with words like "strength" or "twelfth",
    as foreign to them as Polish is to us.
    
    To them, lots of our "syllables" sound like two syllables, 
    owl ==> aa-oo-ll.
    
    And long and short vowels?  An outrage.  They are so grateful for the
    silent e that signals a long vowel sound, as in man v. mane, etc.
    
    The language is almost completely phonetic, as is German.
    

    re .53 "impossible"
    
    Not so.  The sound written "tut, tut" is a click, and most of
    the other clicks are easy to make.  See reply 825.12 in here.
    
    The hardest ones I know of are co-articulations, like the
    r-with-a-hachek of Czech or the "gb" of Ibo (I think it's Ibo,
    could be another West African language).  The tricky part is
    that you have to make many voice-path changes at once, or
    the co-articulation turns into a sequence, which is wrong.
    
    One measure of difficulty is how late in childhood things
    are learnt and how common a less-difficult version is as
    an alternative.  The Czech one is typically learnt very late,
    at four or five.  Any later and it'd probably cease to be
    learnt during the phase of childhood when children are open
    to language "imprinting", and thus would disappear.
    
    Try it yourself: 
    
    v
    R	is a trilled tounge-tip r, like Spanish "perro" at the
    	same time as you produce a "zh" as in "Zhivago" or "allusion";
    
    GB	is /b/ and /g/ simultaneously.  It's different from a 
    	velaric-egressive /b/ or a quick sequence [Ok, in real-life
    	simultaneity doesn't exist, but it's not just /bg/ or /gb/].
    
    Personally the ones I could never get were the Finnish "s"
    and the "Wellesely whistle" dialectical version of English
    "s", both of which are learnt easily and early by many children,
    so your mileage may vary!
    
    		-John Bishop

    Re .49
    
    >Actually, the `eye' sound is two phonemes, the `a' of "ah" and the `ee'
    of "beet."  In classical Latin, the � digraph (ae, for the DEC_CRT2
    
    The "eye" sound of the word "eye" is two phonemes, as you say.
    
    The "eye" sound of the word "keyed", which someone else had included
    in a list for comparisons and which I also compared to, is a single
    phoneme, the same as the "ee" sound of the word "beet".
    
    Now in reply to a subsequent followup, I must insist that the "eye"
    of "eye" is indeed two phonemes and not one.  After all, what is a
    phoneme?  The word "sighed" is a single syllable (by English usage
    of syllables), but it's regarded as a sequence of phonemes, right?
    By exactly the same reasoning, the vowel portion of this syllable
    is a sequence of two phonemes.  If you want to define a phoneme as
    being sometimes a minimal unit of recognizable/meaningful sound but
    sometimes longer than minimal, then I'd say the word no longer has
    any meaning.
    
    -- Norman Diamond

    re .57, vowels in "keyed", "sighed", number of phonemes
    
    I've seen phonemicizations for English vowels like this
    (using ASCII approximations to IPA):
    
       i	      u
    	I            U
         e          o
    	  E        )
    	   �  ^  a
    
    By this system, "keyed" has one vowel (/i/) and "sighed"
    has a vowel and a glide (/ay/).
    
    Another system has only the "lax" vowels from above and 
    uses glides to produce the others:
    
    	I            U
    	  E        )
    	   �  ^  a
    
    "Keyed" is /kIyd/, with /Iy/.  Similarly, /Ey/, /)w/ and /Uw/
    produce the same sounds as /e/, /o/ and /u/, but the system
    claims they are sequences of phonemes (you could even go on
    to analyze "�" as /^y/ and "a" as /^w/, but I don't recall
    ever seeing this done).
    
    Now, how can you choose between these two systems?  One way is
    to look at how they differ in their rules about diphthongs or
    glide combinations: the former has to permit the sequences
    /ay/, /)y/ and /Ew/ and no others; the latter adds /Iy/, /Ey/,
    /)w/ and /U)/ to the list.  In the latter system, you don't 
    have to rule out /oy/, /ey/ and so on individually: all you 
    need is a rule that only one glide can follow a vowel. 
    
    This more general rule seems more satisfying to many (Occam's
    Razor and all that). 
    
    On the other hand, the "generative" school tries to find deeper
    phonemic connections (like between "divine" and "divinity"),
    and winds up postulating a phonemic system for current English
    which looks a lot more like the phonetic system of Middle English
    (i.e. they postulate a "morphophoneme" [i] which in some 
    environments becomes /I/ and in others /ay/ and so on).  While
    the rules are complex, this (morpho-)phonemicization covers a
    lot more territory, and thus some people like this system more.
    
    If you find this confusing, consider (as suggested back around .50)
    the English sound we write as "j"--/d/ and /z^/ (approximation to
    "z-with-hachek") are separate English phonemes in most 
    phonemicizations of English, so we could just use /dz^/ .  But "j"
    is historically a single unit in our othography, and probably a
    single unit in peoples' heads.  Further, while we often see /dz^/
    and /ts^/ initially, we don't seen the parallel sequences /dz/ and
    /ts/ (except for loan words like "tse-tse fly"), and we do see all
    four finally.  This means that a rule about allowable syllable forms
    has to either special-case /z^/ after /d/ (if we make "j" a 
    sequence) or just say "initial sequences are
    
    	[s][stop][glide|liquid] 
    
    and /j/ and /c^/ are stops" (the actual formula is more complex,
    but you get the idea).
    
    Another analysis calls "j" the result of /dy/, which also explains
    the absence of /dz/ and the allowability of /dz^/ in initial
    position.  Unfortunately, some speakers use /dy/ as well as /dz^/
    (e.g. "dubious" in British English, forming a minimal pair with
    "jujube" ).
    
    Is this making sense, or am I failing by assuming too much context?
    
    		-John Bishop

    <<< Note 803.52 by ATYISB::HILL "Don't worry, we have a cunning plan!" >>>
    .
    .
    .
    �And the symbols used are those of the International Phonetic Alphabet -
    �my keyboard can only accomodate about half of them, even with Compose
    �Character    
    
    I wrote a short story once using an English phonemic transcription system
    based on just ASCII (taking a few short-cuts based on guesses about
    the way English/American might evolve). I bowdlerized it a bit to
    make it a bit more readable for general digestion (and in line with
    a competition rule - Nick might know the one); so I'd need to unmassage
    it for this notes file. Maybe, one day...
    
    b

    Re .42 and .50:
    Pleasl see also note 1143 regarding Hungarian alphabet! Some of the
    consonants cs, gy, ly, ny, sz, ty, zs) have double length versions
    (e.g. ccs ggy lly nny, ssz, tty, zzs) as well. Hadn't enlined in 1143
    LJ that I don't remember was it considered to be separate one or not...
    Sorry, that was some 30 years ago...
    
    Joska