the valley girls

I always say [fIgj@`] (/`/ is rhoticity in XSAMPA) and I would say the vast majority of North Americans do as well. I've heard plenty of non NAE speakers omit the palatal glide /j/, however. I had a British professor last year who always said [fIg@].

It's interesting that you mark that as [@`], that is, as a normal vowel that happens to be rhoticized, rather than [r=], that is, as a syllabic consonant. I myself prefer the method of marking it as a syllabic consonant, simply to be consistent with how [l=], [n=], and [m=], all of which are used in at least in NAE, are marked, as none of those can be marked in a fashion similar to [@`], so hence to use [@`] rather than [r=] would be a bit inconsistent consequently.

Oh, I mark rhoticity there with [`] because that's generally how I see it done in conventional transcription so it's what I'm used to (it's also what I learned in phonetics class). I sometimes see syllabic [r] used for that but the explanations I've heard for using [`] are that showing a schwa that happens to be rhoticized is a more accurate description as compared to using [r=]. I think [r=] is often used in some older American transcription systems that are somewhat similar to IPA but have several differences. Anyway, it doesn't really matter too much, it doesn't bother me to see [r=], I'm just going along with convention and what I'm used to by using [@`].

It would be most surprising if a British professor, or indeed anyone from Britain or other commonwealth countries (perhaps excluding Canada) said anything other than [fIg@].

But the interest lies in why AmE includes [j] in the pronunciation of this word, more than in the rhoticity, given the general tendency of AmE speakers to omit the [j] in many words which end in "ure". It seems to be a one-off, and one which is entirely at variance with all other forms of English, even those which are slightly rhotic, such as the Irish and Scottish Englishes.

I say figure, mature, endure and lure with the [j]. There are also demure and tenure, which I've never hear an AmE speaker say without the [j].

I'm surprised to hear that most AmE speakers omit the [j] in most words ending in -ure, as the people I associate with tend to use it. Is this yet another instance of my age showing? Do young people today really consistently leave it out?

Actually, Ned, as far as I know, yod-dropping doesn't occur after [g] in North American English. Yod-dropping usually occurs after [s] [z] [l] [T] and alveolar consonants. Other consonants which don't fall into that class still get yod afterwards. So, I have things like /mjuzIk/ /fju/ /kjut/ and /fIgj@`/ for "music" "few" "cute" and "figure", while words like /nu/ /sut/ /dud/ and /tun/ for "new" "suit" "dude" and "tune" don't get yod in my speech. Since yod-dropping only applies to the consonants I listed I don't think it's too surprising that "figure" still has /j/. (I only did phonemic representation here because I didn't wanna draw attention to the fact my /u/ is really much fronter and unrounded because that's irrelevant to what I was talking about and didn't wanna create any confusion over it).

Deborah, I have [j] in only "figure", "demure", "tenure" in that list. Just to list how I pronounce them all, for me "figure" is /fIgjr=/ (--> [fI.gj@`]), "mature" is /m@tSr=/ (--> [m@.tS@`]) (if I'm being very formal, I will use /m@tSur/ (--> [m@.tSur\])), "endure" is /Endur/ (--> [E~n.dur\]), "lure" is /lur/ (--> [lur\]), "demure" is /dEmjur/ (--> [dE~.mjur\]), and "tenure" is /tEnjr=/ (--> [t_hE~.nj@`]). As for [j] being in words ending in -"ure", well, here it varies a lot, being included in some words, not included altogether in others, and in some, especially where "t", "d", or "s" is before the -"ure", being lost, but in turn palatalizing the previous consonant, as in my example of "mature".

I guess I should clarifiy a bit..."tenure" "manufacture" "annual" are [tEnj@`] [mænj@fækS@`] and [ænju@l] for me but these exceptions seem to follow a rule: all are multisyllabic and the yod isn't dropped following a stressed syllable. So, for me as long as the normally dropped yod in those consonants isn't in the stressed syllable maybe it can still show up in other places (I overlooked this because I was just thinking monosyllabic words initially, obviously with no need to clarifiy for stress).

"mature" "endure" and "lure" are [m@tS3`] [Ind3`] and [l3`] as I say them. One thing I should mention is that [u@`] is commonly leveled to [3`], at least here. My university, UC San Diego, is split up into 6 undergraduate colleges, one of them being John Muir College, so I hear the word "Muir" all the time, and the majority of students (and me) say [mj3`]. This is interesting to me because it shows not only a retention of yod in the expected condition, but it's analyzing it as the salient feature of the /ju/ cluster in such words, to the point that when /u/ is leveled, /j/ remains while /u/ disappears, putting the stress on the final vowel instead of /u/.

I believe in East Anglia it's common to drop the yod after all consonants. On the travel show "Globe Trekker" there is this guy named Ian Wright, whom I believe to be from East Anglia, who pronounces "museum" as /muzI@m/ and so on.

In the UK, they pronounce TUNE asTCHOON, DUCE as JUICE and DUDE as JUDE...

Hey, Jude :p

Did the pronunciation of U as /ju/ originate as an attempt to imitate the French /y/ sound?

I should have pointed out though that I may also pronounce "endure" as /Indur/ (--> [I~n.dur\]); in formal speech I'm more likely to favor the /Endur/ pronounciation, while in informal speech I'm likely to use the /Indur/ pronunciation. Overall, unstressed word-initial /En/ will likely become /In/ here in many cases.

Yeah, I don't have the "pin/pen" merger in my speech but unstressed word-initial /En/ is always /In/ for me. This also applies to high-frequency words like "when" in normal conversation.

Lazar, I think that's right about /ju/ being an attempt to imitate French /y/...I've never read that anywhere but that's where I always thought it came from.

I myself do not speak a "pin"-"pen"-merged dialect, yet even still, /E/ seems to be in many places reduced when unstressed to [I] rather than to [@], or if followed by a syllabicizable consonant, simply elided, with the following consonant being syllabicized in turn. However, in single-syllable words like "when", this does not appear to happen, where it remains as [E] usually; hence, for me, "when", that is, /wEn/ or /WEn/, is [wE~n] or [WE~n], not [wI~n] or [WI~n]. Mind you though that the use of /W/ rather than /w/ is not a native feature of the dialect, as /W/ does not natively exist in the dialect here, but rather just a learned feature of my own idiolect which I use inconsistently, and more often in formal speech than in informal speech.

Oh, that's interesting...I definitely use [wIn] for "when" in most normal speech. I'll say [wEn] if it's for emphasis. I never have /W/ in my speech, even in formal situations. I've tried producing that sound just for fun but can only accomplish it with a preceding /h/, so [hw], but needless to say /W/ isn't native to the dialect here either...I'd say even in the most formal of speech.