Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.2 9/18/84; site spar.UUCP Path: utzoo!linus!decvax!decwrl!spar!ellis From: ellis@spar.UUCP (Michael Ellis) Newsgroups: net.nlang Subject: Re: Easy languages (number of vowels: English vs. Esperanto) Message-ID: <730@spar.UUCP> Date: Tue, 31-Dec-85 05:29:44 EST Article-I.D.: spar.730 Posted: Tue Dec 31 05:29:44 1985 Date-Received: Wed, 1-Jan-86 04:28:11 EST References: <1791@uwmacc.UUCP> <839@h-sc1.UUCP> <1809@uwmacc.UUCP> Reply-To: ellis@spar.UUCP (Michael Ellis) Organization: Schlumberger Palo Alto Research, CA Lines: 64 > = Neal McBurnett >> English has 7 vowels and 7 diphthongs > >There are certainly many ways of counting, but I think it is closer >to 12 vowels and 8 diphthongs, at least with my accent: > bean, bin, Ben, ban, bun, barn, Bonn, born, burn, boon, good, banana > say, no, fly, how, fear, care, poor. There are indeed many ways of counting. I'd arrange those words according to their vowels as below: pure +y +w +r i bin bean - *beer e Ben say - care ae ban - how - aa bun - - (burn) ao +bond fly - barn o +pawn *boy no born u good - boon poor Notes: (+) `Bonn', as a German name, is pronounced in a variety of ways, sometimes more like `pawn', other times more like `bond'. These are clearly two different phonemes in most English speech I hear. Some minimal pairs: tot/taut, bottom/bought'em, are/or, doll/tall (*) `boy' and `beer' seem to be omitted in your list. (1) Due to ascii lossage, the digraphs /ae,aa,ao/ are employed here to represent `pure' vocalic sounds in most english I have heard. (2) /ur/ (usually spelled `oor') has dubious ontological status, and is often (usually?) indistinguishable from /or/. This confusion could be brain damage on my part. (3) The sounds /iy,uw,aar/ (bean,boon,burn) are arguably `pure long' vowels. There is no question that, if they are considered to be compound sounds, the preceding vowel is modified considerably from its pure sound. In most american english dialects (especially that heard on TV), postvocalic /r/ (as in barn, care..) is a retroflex glide that is quite distinct from the preceding vowel. Should I count vowel + /r/ as a diphthong? If so, why stop there? Why not also count vowel + /l/ as well? Not to mention vowel + /n,m,z,k.../? Anyway, if I exclude the r-compunds, I count a total of 7 vowels and 7 diphthongs, excluding unaccented schwa(s). If /iy,uw/ are pure vowels, the total is 9+5; adding /aar/ as pure and 5 r-compounds, the total is 10+10, a sum which equals your 20. British english and certain eastern american dialects differ from the above scheme somewhat, since the final-r's either form schwa-diphthongs, or result in compensatory vowel lengthening (I have heard that this is a later development that occurred after the colonial period). There is also considerable divergence in the vowels of (pawn,pond,pan/path) from the dialects I am familiar with. I suspect that my /ar,or/ => /a:,o:/ in your speech (whereas /ir,ur,er/ => /i0,u0,e0/ (where /0/ represents `a' in sofa)). This would transfer two more r-compounds into the pure category, thus redistributing 10+10 => 12+8, in perfect agreement with your count. Do you have a british or east-coast american accent? -michael