Deborah, the "gure" ending is admittedly unusual, and I didn't address such words as "tenure", as I was thinking more of "mature" "endure" etc. Would you say your use of [j] in "mature" and endure" is typical of AmE speakers?
But back to "figure". What I would like to learn is why AmE speakers insert the [j] in this particular word, while I know of no other variety of English where this done for this particular word.
i.e. did it originate with someone, somewhere, sometime, deciding that it sounded "educated" to do so because the Brits (other perhaps than Ian Wright and his fellow-Cockneys) use it in almost all other words ending in "ure" or which include "ur" so it must be correct to do so?
As I said, it's an unusual word, and it has received most unusual - and different - treatment in both streams of English. But why?
Well, sometimes when I'm using really strong emphasis, I've ended up pronouncing /W/ as something more like [xW] or [xw] than [W], even though I do that less than I used to. But then, I do speak English, for the dialect here, sort of weirdly at times, considering I probably use the subjunctive more often than most here, often preserve [E] in word-initial position or right after a consonant in word-initial position, where it would often be replaced by [i], [I], or [@], depending on the given case, and so on.
However, though, a lot of those things are for me still formal register things, even though I use them more than most, and are things I tend to not use a whole lot in informal speech, where I often go in the opposite direction more than many and adopt more progressive forms, as opposed to the more conservative forms I'm likely to use in formal speech. At times I often am not sure "how" I pronounce a single word, as I often have multiple different manners in which I will pronounce individual words, and will use different versions in different contexts, with more reduced or phonologically progressive forms being preferred in informal speech, and more conservative forms being preferred in formal speech.
I definitely understand that. Sometimes I also don't know "how" I pronounce a word, because it can vary depending on the context. I have noticed, some things, however, such as that my vowels can change quite a bit (towards the most phonologically progressive and innovative California Vowel Shift forms) when I'm with friends and people my age, while my vowels tend to be more conservative in other situations (even tho they would still be considered informal).
Back to "figure," I don't know the historical pronunciation of the word but as a "-ure" word I don't think it's surprising that a /j/ would be present in its historical pronunciation. Oxford and Webster list both the /j/ and /j/-less pronunciation. I can't think of any other "-gure" words for comparison, but its unvoiced counterpart "-cure" has plenty of examples, and is always /kj/ in North AmE. Looking at the unvoiced counterpart may be useful in the absence of other "-gure" forms because on the list of consonants that do precede yod-dropping, both the voiced and voiceless sounds at a particular place of articulation go together. Thus, /s/ /z/ and /d/ /t/ precede yod-dropping. I would assume that given that precedence, since /k/ does not precede yod-dropping, then it would be expected that /g/ wouldn't either.
Explanations of yod-dropping tend to explicitly list the consonants which go along with yod-dropping (and /g/ is never one of them), implying that other consonants retain the yod. For this reason and the fact that the same sound, just unvoiced, /k/, doesn't go along with yod-dropping I think it's reasonable to say that if there were many "-gure" words, [gj] would be the expected form. So, I could be wrong but my best guess would be that the historical pronunciation of "figure" had /gj/ (as is still the norm in North AmE).
Ned, I'd say my pronunciation of "mature" and "endure" is typical of the people I grew up with. I'm in my fifties, from California.
I figure (with a [j]) that pronouncing figure with a [j] is an attempt to reproduce the French pronunciation.
Well, though, the thing is that in many cases one must remember that the /j/ has not been simply dropped, but has rather been absorbed into a preceding consonant through exerting a palatalizing effect in it, which has then been fixed in the phonemes which are in use, rather than being solely a phonological-level effect. "Mature" here is a good example of such a word, as where some people have just dropped it, resulting in /m@tur/, which I hear at times, in the dialect here the /t/ has been replaced with /tS/, as rather than the /j/ being simply dropped, it has left the /t/ patalalized as /tS/ phonemically. It is quite likely that most words in which an alveopalatal fricative or affricate is followed by /ur/ or /r=/ which happens to be written as "ure" is likely the result of this kind of phonemically frozen palatalization.
As for different versions of words, at least in my case, while there are words that just vary phonologically to some degree or another in different versions, for example "sister" being /sIstr=/ (--> [sI.st@`]) or /sIStSr=/ (--> [sI.StS@`]), there are some words in which the word exists in formal/informal pairs where the informal version is more drastically modified, for example "forget" being /forgEt/ (--> [for\.gE?]) or /fr@gEt/ (--> [fr\@.gE?]), "probably" being /prAb@bli/ (--> [pr\_0A.b@.bli]) or /prAli/ (--> [pr\_0A.li]), "recognize" being /rEk@gnaIz/ (--> [r\E.k@g.naIz]) or /rEk@naIz/ (--> [r\E.k@.naIz]), and so on. These kinds of pairs indicate that individual words have more than one version being maintained for them in parallel, corresponding to the same word but in different registers, rather than multiple word versions simply being a consequence of phonological differences between registers.
Slight correction, but [r\E.k@.naIz] should be [r\E.k@~.naIz].
Thank you, Deborah. Good point, there is a slight [j] in the French pronunciation, and it may have been present in standard English towards the end of the eighteenth century (unless it came directly into English from the Latin "figura", which in classical Latin pronunciation at least would not have involved a [j]). Much obliged.
Kirk, for info, my older OED (1959) doesn't mention the /j/ pronunciation, which says a bit.
That's interesting...I wonder if anyone knows anything more about the historical development of the pronunciation of "figure" and why [j] is common in NAE but not in other varieties.
In an interesting parallel to the /W/-/w/ merger...
In New York City it is common to drop the /h/ from the initial /hj/ in words like "human" and "huge".
Donald Trump is an example. One of the most parodied features of his speech is how he says "The winning team will receive a yuge prize!"
Another example is Larry David (originally from New York). He made reference to this in an episode of Curb Your Enthusiasm. There was a character named Hugh. Larry pronounced his name /ju/ but Hugh kept telling him it was /hju/. Finally, they had a fight. Hugh said "F*ck you!", to which Larry David responded "No, fuck Hugh!" with an overemphasized h.
I've heard New Yorkers pronounce "Manhattan" without the /h/, sort of like "mnattn."
The first time I remember hearing the h-drop from /hj/ was in high school while we were watching some otherwise boring video, but the announcer kept on saying things like [judZ] and [jum@n] for "huge" and "human"...it kinda took me aback as I had never heard that before (or had not been aware of it) but it was really noticeable and I ended up paying attention more to his pronunciations than to what he was actually saying...are there any other places besides New York City that do this? I still don't think I've ever met anyone in person who does that but I've never been to New York City and don't know that many people from there.
Considering that I've never heard of a North American English dialect dropping the /h/ in initial /hj/ before this, I would say that I know of no dialects outside of New York City which do this, as a whole. On an unrelated note, as long as we're speaking of dropping things in word-initial consonant clusters, do any of you know how widespread the dropping of /w/ in initial /kw/ clusters, which seems rather common here in Wisconsin, really is, perchance?
I wasn't aware any dialects that dropped /w/ in initial /kw/, that's the first time I've heard about it. As far as I can remember I don't think I've met anyone who did that...do you consistently do that, Travis? The only word I can think of where there's variation in that respect may be in "quarter," which was usually pronounced /w/-less a couple of centuries ago but with the influence of spelling has returned in some people's speech. Otherwise, I've never heard anything but /kw/ in any other such words, like "quiet" "quote" "queen" etc. It'd be interesting to see if any other places besides Wisconsin do this.
Well, it's only in some words, such as "quarter", "coordinate" (and derived words), "quart"; it's not in others such as "quiet", "quote", or "queen". I should have clarified that it seems to be limited to initial /kwo/ or maybe even more specifically initial /kwor/, not just all initial /kw/ clusters, and even then it doesn't apply to "quote", that is, /kwot/.
Oh, ok. That makes more sense then...yeah I'm pretty sure variation in /kwor/ words has to do with the fact that several centuries ago those words were pronounced /kor/ but spelling pronunciations have prevailed in some people's speech (I think it's stronger depending on the region, as well) so /kwor/ isn't uncommon now.