Of the many words that distinguish American accents from British, ‘tune‘ serves as a particularly good test. Many British accents insert a small ‘y’ sound before the vowel–resulting in something like ‘tyoon‘–whereas for most Americans, this word is homophonous with ‘toon.’ The process of losing the ‘y’ is termed yod-dropping.
Yod-dropping seems to be an ongoing process in English. Earlier in the development of the language, words like ‘brew’ and ‘chew’ were also a part of of this category of ‘yoo’ words. In contemporary times, the distinction between ‘brewed’ and ‘brood’ is only made (in this respect) in a select number of dialects (Welsh English being the most commonly cited example).
American yod-dropping does not impact every word of the ‘yoo’ type. John C. Wells summarizes the variability here in his Accents of English (vol. 1, pg. 207):
In GenAm, and also in parts oft he south and midlands of England, /j/ is lost after alveolars, /t, d, n, l s, z/ but not after labials or velars …
So, logically, for most Americans ‘news’ becomes ‘nooz‘ and ‘due’ becomes ‘doo,’ but ‘fuse’ remains ‘fyooz‘ and ‘cute’ remains ‘kyoot.’ For the most part, the American ‘yod’ seems easily predictable.
And yet, examining my own accent, the ‘y’ before ‘oo’ in some words seems more weakened than entirely dropped. One notable example of this is the word ‘news.’ I doubt many Americans pronounce this ‘nyooz,’ but in my own accent the word doesn’t entirely rhyme with ‘booze,‘ either. For me, at least, the vowel has a very slight onglide, something a bit like IPA [nɪuz]. So in some words for which I drop the ‘yod,’ there are nevertheless slight remnants of it.
Of course, Americans are not the most robust yod-droppers. In older dialects of East Anglia, yod-dropping impacts all words of the ‘yoo’ type. Hence ‘music’ becomes ‘moozik,’ ‘cute’ becomes ‘coot,’ and ‘fuze’ becomes ‘fooz.’ Why we Americans haven’t gone down this road is quite puzzling. Only time will tell if we do.
I’m afraid I’m going to finish this post with a question that sounds a bit unseemly: do you drop your ‘yod?’
*’Yod’ is a word of Semitic origin that refers to ‘y’-type sounds.
For such a small sound, it’s surprising how much semantic work it can do. A number of times I’ve completely failed to understand a sentence because an American has pronounced “due” as “do”. Once a friend who had been to England told me how it took her a while to work out what the /tup/ was; I had no idea either. Only after she started explaining about public transport did I realise she meant /tjub/.
In Australia yods are absolutely obligatory after /t, d/ but has become optional after /l, s/. For /n/ it varies: it’s always /njuz/ and /njuklɪə/ but rarely /njud/.
Re nuclear: George Bush was quite correct to want to insert a yod, he just put it in the wrong place!
I actually had a similar thought writing about the East Anglian yod-dropping. ‘Feud’ and ‘food’ are homophones? ‘Cute’ sounds like ‘coot?’ How could people possibly understand what East Anglians are talking about? Of course, Australians and Brits are just as confused when I talk about a ‘do date!’
Once when I was in Peterborough railway station, I was asked whether I was in the [ku:]. I had to think. I had heard [ku:] for “cow” from Scots, but there was no cow anywhere. I had to think hard to realise that he was referring to the queue.
“For me, at least, the vowel has a very slight onglide, something a bit like IPA [nɪuz]. So in some words for which I drop the ‘yod,’ there are nevertheless slight remnants of it.”
Random musings:
You may also notice – in others, if not yourself – that difference between Liquid U and U-fronting can be a source of confusion to dialect students.
The yod-dropping you point out in East Anglia can be observed elsewhere. By the way, I’m not sure its use in East Anglia is limited to older dialects, as Suffolk native Ian Wright (one of the hosts of Globe Trekker), who is not terribly old, yod-drops.
Have you noticed yod-dropping in unstressed syllables? I’ve heard this a lot following m in words such as “communist” – in both American and English speakers. I’m certain that all the Americans and most of the English would preserve yod in words such as “music.”
I also think I hear more Americans dropping the yod in “figure” even in formal speech.
Thank you for mentioning Ian Wright Amy! That’s exactly who came to my mind as I was reading this post. He definitely has very robust yod-dropping. He has it in beautiful, music and museum. And you get to hear him use those words a lot during his travels. Unrelatedly, he also uses a boat load of glottal stops. His accent actually sounds quite Cockney to me outside of the yod-dropping, but I’m just a simple American so I could be missing something.
“Figure” is an interesting case. I maintain the yod in that word, but many in the Northeast seem to opt for the ‘figger’ variant. That’s almost worth a separate discussion, though, ‘-ure’-type words exhibit so much variation in American English. I’ve heard Americans treat ‘pure’ and ‘cure’ as part of the NURSE set or the FORCE set depending on the speaker. There’s also an exceptional phoneme of the /uə/ type which some Americans maintain for only a tiny selection of words (such as ‘tour’ and ‘manure’). And then there are words of the ‘-ture’ type which seemed to have undergone an earlier process of yod-coalescence and levelling with the NURSE vowel.
BTW, I love Globe Trekker!
“I also think I hear more Americans dropping the yod in “figure” even in formal speech.”
According to dictionaries and other things I’ve looked at, “figger” is the British pronunciation.
Sorry, meant to give you an example of “elsewhere” above! I have heard yod-dropping of the kind you describe in East Anglians in speakers from the Erewash Valley (border of Nottinghamshire/Derbyshire in the Midlands).
I’m American, and I’m a yod-dropper.
However, it isn’t uniform across the USA. News is definitely and only “nooz,” rhyming with booze (and shrews, lose, Flooz, goos, ruse…). No yod-dropping for music, cute, puce, mute, spew–pew and poo are NOT homophones (it’s a more fun example than mute vs. moot).
The one word I see as a big identifying split between accents is “mature.” Even when two speakers try to announce in GenAm, some will say “machure” while others say “matoor.” Would anyone know what that split is?
And what’s the term for the mashup of American yod-retaining that make “machure” or “shure”?
Out here in the San Francisco Bay Area, I usually hear a yod in “new” and “news” from people my age (late 30s) or younger. My impression is that it’s on the rise in these two words only; I wouldn’t expect to hear it in “knew”, for example.
I’m curious if that has to do with the environment: in the California Vowel Shift, /u/ is often reported as being fronter after coronals. Not quite sure how that would result in ‘yod-reinsertion,’ though!
Not only does /u/ front, but it can also gains diphthongal qualities similar to what you describe, as is the case I’m used to hearing around here. Fronted but non-diphthonged /u/’s sound odd.
After once talking to an older speaker who kept the yod in ‘news’ and perceiving it as sounding like mine, I was actually confused whether I was a yod-dropper or not, but pairs like do/dew, tune/toon, and due/do are still homophonous, but I can’t readily hear them as distinct from true yod-full versions.
Hmm, my ‘manure’ is in the NURSE set with good frequency, but for some reason I perceive it as sounding antiquated/foreign/british.
Oh, the fronted /u/ in Californian accents!
I call that the “dude” vowel. You know the way that the word “dude” only really makes sense and demonstrates the casual-compadre meaning when spoken by a Californian?
It’s not dood, it’s not dyood, it’s deuhd.
Dude totally xD
Yod-coalescence.
Thanks, Amy!
I feel the stress plays a role in when to palatalize (yod-coalescence) alveolars.
Here are some combinations of unstrssed alveolars followed by a sorta back vowel.
atu:con-‘grat-u-la-tion
adu:’grad-u-al
etu: ‘pet-u-lant
edu: ‘ed-u-‘cate
itu: ‘sit-u-a-tion
idu: ‘in-di’vid-u-al
otu:’clot-ure
odu:’mod-u-le
utu: ‘mut-u-al
udu: ‘frau-du-lent
,Oppor-‘tu-ni-ty (no palatalization, but aspirated t)
For “mature”, I think the pronunciation is related to the pure-poor split. Among people who’ve lost the CURE lexical set completely, “mature” would have to be pronounced matyore, machore, matore, matyurr, machurr, or maturr. Of these, “machurr” is the only one that sounds like I’ve ever heard it. If you haven’t lost the CURE lexical set completely, I believe this is one of the words that tends to keep the CURE vowel. The yod is either coalesced or dropped; I’ve heard it both ways.
I’m not familiar with the CURE lexical set. I say “kyoor” in formal conversation and “kyurr” in casual conversation. Similarly, I’ll say “machoor” formally and “machurr” casually. I guess I have a CURE lexical set?
Cure as kyoor would be CURE set, as kyurr, NURSE set. [ kyore would be FORCE set]
CURE words would be tour/mature/cure/pure/lure/poor/your/who’re/sure/endure/ensure/impure/insecure/unsure/mural etc.
There’s the pure-poor split where some CURE words merge with NURSE words and some with FORCE words.
The pour-poor merger where CURE words merge with FORCE words. sure= shore, poor= pour, you’re= yore, tour= tore
And the cure-fir merger where some CURE words merge with the NURSE set.
I can’t speak for your CURE set, but generally for me: cure/pure/mature/sure/endure= NURSE, tour/poor/who’re= CURE, poor≠ FORCE, so I’d say I’m cure-fir merger and no pour-poor merger, leaving me with a reduced but intact CURE set of sorts.
Regarding variation as tarwicks brings up, my your and moor is sometimes FORCE, and my lure and endure vary between CURE and NURSE.
What’s interesting is when I try to switch from a NURSE to CURE vowel for formal situations, I may hypercorrect to a FORCE vowel for some reason, so my sure becomes shore xD
Thanks, m.m. Here’s my breakdown as best I can figure:
Pure CURE: lure, tour, endure, who’re, moor, you’re, velour
Formal CURE, casual NURSE: sure, mature, pure, secure, cure
Pure NURSE: security, endurance, purity, cur, hurt, burst, figure
FORCE: poor, pour, door, lore, core, boar, your
Some of us (me) have the pure-poor split (remove the y from pure and I get purr, which doesn’t sound like poor) AND the pour-poor merger (homophones for me).
So, seems to me that “The pour-poor merger where CURE words merge with FORCE words” isn’t quite right, since at least some of us with the pour-poor merger (assuming pronouncing those words alike means we have the merger) pronounce the CURE words in multiple ways.
Or maybe these sets are just more trouble than they’re worth, in this case.
Pure-poor split means pure= NURSE and poor = FORCE
Pour-poor merger means pure= FORCE and poor= FORCE
I don’t see how you can be split AND merged, but indeed the CURE set might just be a bit of spoken disaster xD
I just realised that I DO make ‘tour’ a NURSE word, albeit with very low frequency.
@mm
It’s three different words: pure, poor, pour.
Pure has the vowel of nurse, except with a y sound before it.
Poor and pour are homophones and have the same vowel as force.
I do not, though, have a CURE set; that is, the “CURE” words don’t all have the same vowel sound for me.
That’s pure-poor split.
If you were poor-pour merged, pure without the ‘y sound’ would be merged with pure and poor.
But I also, as I said, pronounce pour and poor alike. But that’s not a poor-pour merger? Surely it is. Maybe there’s more than one, in which case, you need a more specific label if you are going to go telling me I don’t have the merger even though I pronounce the words alike!
I had an Australian instructor at grad school, and his “tune” sounded very much like “chune” with the y sound.
Yes, that’s called “yod-coalescence” and I believe it’s quite common in Australia (and England too nowadays).
I’m American and I think I’m very inconsistent here…
I do pronounce yod in:
new, few, pew, chew, spew, cure (kjir, but not kjoor), mature (debatable, but not like matoor)
I do not pronounce yod in:
due, dew, stew, lewd, Tuesday (but it’s not quite Toosday… the eu is still a dipthong), tune, Cupertino, coupon, tune, brew
Oops, that was not meant as a reply to this subthread.
What I *meant* to say ways, this reminds me of a Coldplay song called Clocks. One section sounds to me like:
…
Come back and take you home
I could not stop that CHEW [you] now know, singing
Come out upon my seas
Curse missed opporCHEWnities
…
I’m from Derry, Ireland where we seem to love the yod. I get laughed at quite a bit by my American boyfriend for saying words such as ‘car’ like ‘cyar’. Apparently I sound like a pirate!
I’ve noticed a pastor in Buffalo, NY and another from Michigan pronounce church and chapel with a y after the initial consonant sound.
Trawicks, you wrote: “Earlier in the development of the language, words like ‘brew’ and ‘chew’ were also a part of of this category of ‘yoo’ words.”
I can imagine the yod in “chew”, but for the life of me I can’t imagine it in the word ‘brew’. I don’t even know how to say it.
In Welsh English, where this distinction is (apparently) still made to some extent, it’s not a full-on /ju:/, but rather something like [ɪu]. /rju/ is admittedly pretty hard to pull off!
Yes, it’s a diphthong in Anglo-Welsh [ɪʊ] (tune, few, used). The diphthong also occurs in Welsh. Souther Anglo-Welsh varieties distinguish between ‘blew, threw’ [ɪʊ] : ‘blue, through’ [uː].
Say the word “re-use” and shorten the first vowel to nonexistence.
… and then prepend a /b/ 🙂
There are two pronunciations for coupon, one with a yod, another without it. There is no yod in the pronunciation of the city Cupertino; however, there is a yod in ‘accumulate. Both words contain stressed -cu- syllables.
Cupertino is a direct loan from Spanish.
Accumulate is a backformation from accumulation, which is itself a 15th-century direct loan from Latin accumulatio (-onis). These words follow the general practice of rendering Latin /u:/ as /iu/, which later became /ju:/ with yod-dropping appropriate to dialect. I would imagine that this results from the influence of French, where Latin /u:/ was fronted to /y/.
Coupon is a recent (19th-century) loan from French, where the first syllable is yodless /ku/. Neither etymology nor spelling provide any justification for initial /kj/ in English: this word perhaps acquired a yod (in some, I think mostly North American, varieties) by anal0gy with other words (e.g. cube) that already had it.
It is well and good to know the history of words. The history doesn’t play much of a role in “loan word phonology”. Dictionary.reference.com lists kyu version of Cupertino as an alternative pronunciation.
Ellen, you are right: it got two yods.
Two yods, actually, in “accumulate”. 🙂 (Though only one -cu- yod.)
I just realized i tend to drop the first yod in “accumulate”. strange.
I think I have yod-weakening too, or perhaps some slight fronting of the [u] in historical [ju] words. It’s hard to tell without a spectrogram.
I actually tend to agree with you in that I don’t completely drop it. It may be that it’s not as pronounced in the word news as it is in the word cute, but it’s there.
I routinely hear a yod or a dropped yod in the same words often from the same speakers, dependng on prosodic variations. Also, on an individual basis, in the New York Metro area Noo Yawk varies with Nyew York, Toosday/Tuesday, etc.
Younger RP speakers in the UK usually favour some degree of yod-coalescence when they’re [dʒuː] to take the [tʃuːb] on [ˈtʃuːzdeɪ], but not so reliably after [s] or [z], except at word boundaries, where [aɪ geʃ ʃu] might hear that and it wouldn’t [səˈpɹaɪʒ ʒu].
I think that [aɪ geʃ ʃu] is the odd one out there. I’d consider that Estuary rather than RP.
A strange development in my home area is that the town of Dewsbury is now frequently pronounced [du:zbrɪ], having always been [dju:zbrɪ] previously. I don’t hear much yod-dropping in other words in this area, so I’m unsure why the town-name’s pronunciation is changing.
[aɪ geʃ ʃu] strikes me as an instance of yod coalescence working in tandem with glottal stopping, ultimately resulting in the /t/ disappearing entirely. The process might be described as something like [get ju] –> [geʧu] –> [geʔ ʃu] –> [geʃ ʃu].
There is no t to drop.
…where [aɪ geʃ ʃu] might hear that…
That would be “I guess you”.
Whoops! Didn’t read that carefully enough. “I get you might hear that” would probably only work in America. Ignore that!
I keep my yods after /n/ and /s/, but after /t/ and /d/ I am inconsistent. I drop them after /st/, but sometimes keep them in other contexts. In fact, for me dual and duel are no longer homophones because I drop the yod in dual.
Pingback: This Week’s Language Blog Roundup | Wordnik ~ all the words
Pingback: Srimp and Jritos at the Groshery Store « Literal-Minded
Pingback: Link love: language (35) « Sentence first
I’m American, born just outside the New York City isogloss bundle some fifty years ago, and have a robust historical CURE /ju:/ lexical set, with one anomaly: Sure! as a word of agreement is NURSE, whereas all other uses such as I’m sure are CURE. I wonder if I borrowed this pronunciation from another dialect….
Historically, the words ending in unstressed -ure were yodless: the use of yod in them is a spelling pronunciation, which partly explains why it is so common in AmE.
Historically, the words ending in unstressed -ure were yodless: the use of yod in them is a spelling pronunciation, which partly explains why it is so common in AmE.
Really? How then do you explain the yod-coalescence in “measure” and “pleasure” (and “azure”)? Surely the sequence must have been:
1. /zjuːr/
2. /ʒuːr/
3. /ʒəɾ/
Similarly, “culture”, “nature”, “future” etc. must have been:
1. /tju:r/
2. /tʃu:r/
3. /tʃər/
And “pressure” etc. must have gone from /sj/ to /ʃ/.
These words make up the great bulk of English words ending in unstressed “-ure”. The only major exception I can think of is “figure”, where it’s possible that the /j/ in AmE is a spelling pronunciation — but I would love to see a source on that.
In Central Texas, and in many parts of the South, older speakers — and I’m starting to be one — retain their yods pretty well. E.g. Styoo for stew (as kids we made fun of “styoo”), styoodent (student) tyoon (tune) dyoo (due and dew). I say nyoos but generally stoo for stew. My father, b. 1936, has a robust yod.
PS My dad was born and reared in Travis County, TX.
I’m from Central Texas as well, and I agree with what you say. I was born in Austin in 1976, and grew up in Williamson and Bastrop counties. I speak GenAm, so I drop ‘yod’ where GenAm speakers tend to.
My dad (born 1947) has lived in Williamson County all his life, and retains his yods, as do the older members of that side of the family. My mother (born 1949), was raised in Austin and does NOT retain yods, despite otherwise having a central Texas accent. Her parents, from Travis and Williamson counties, DID retain yods.
I live in Austin but teach in Taylor, and the kids I teach that DO speak with Texas accents tend to yod-drop, like my mother.
I grew up in Upstate New York. Here, “new” is always “noo,” “dew” is always “doo,” “Tuesday” always starts with a “too,” etc. I wasn’t really aware of the yod pronunciation until I started to study linguistics in high school.
When I’m in the New York City area, I often hear “avenue” pronounced like “ave-nyoo.” I’m not sure if this is just a lexical exception though, because “New York” usually sounds like “noo.”
It’s easier to preserve yods across unstressed syllable boundaries. Someone whose accent disallows ny- clusters (and hence “nyoo”) could still have “aven-yoo”.
An outstanding share! I have just forwarded this onto a coworker who had been doing a little homework
on this. And he actually ordered me breakfast
because I found it for him… lol. So let me reword this.
… Thanks for the meal!! But yeah, thanx for spending some time to discuss this matter here on your web site.
This article is just toooo funny – loving the debate !!
You wrote,
“One notable example of this is the word ‘news.’ I doubt many Americans pronounce this ‘nyooz,’ but in my own accent the word doesn’t entirely rhyme with ‘booze,‘ either. ”
Well said! Robert Siegel, an “All Things Considered” presenter on NPR, regularly used to pronounce “news” as “nooze.” This always surprised me, because in his work (as a professional news-reader, you know) his diction was otherwise quite good.
I sometimes wonder if he did it on purpose for a little laugh, maybe to fight boredom.
I apparently didn’t notice this years ago when this post was new, but it’s striking me now.
In the original post: And yet, examining my own accent, the ‘y’ before ‘oo’ in some words seems more weakened than entirely dropped. One notable example of this is the word ‘news.’ I doubt many Americans pronounce this ‘nyooz,’ but in my own accent the word doesn’t entirely rhyme with ‘booze,‘ either. For me, at least, the vowel has a very slight onglide, something a bit like IPA [nɪuz]. So in some words for which I drop the ‘yod,’ there are nevertheless slight remnants of it.
I find that curious, that you would see them as not rhyming. For me, muse, and news (one with yod, one without) I would consider to be rhyming words. So it’s interesting that you consider the yod to affect rhyming. For me, it doesn’t, for the same reason that I say “a union” not “an union”. Or does the fact that it’s not an actual yod make a difference?
It might be that Mr. Trawick-Smith uses a strict interpretation of the verb “to rhyme”, taking it to imply, for monosyllabic words, that the entire pronounced vowel sequences must be identical. Then “use” would rhyme with “views” but not with “snooze”. I do not think there are any definite rules about how much similarity is required to declare that two words rhyme. Some poets rhyme words whose ending-similarity is incomplete.
For me (lower Midwest, US), “muse” doesn’t have a yod. Instead, it has a diphthong like [ɪu]. Ben apparently uses a similar sound in “news”, but I don’t. “News” is just [nu:z] for me. “Union” and “use”, however, do both have a yod at the beginning. Apparently I can only have a yod at the beginning of a syllable. I’m not sure if I would say that “use” and “views” rhyme. However, I would say that “use” and “news” rhyme. Not that I expect anyone to care about any of this. I’m just bored 🙂
It’s funny with me. Both my parents are British(one with a weakened Scouse/Liverpool accent, one with a Welsh accent) but I grew up completely in America.
And I can’t say whether I drop my yod or not- I’m pretty sure I do both.
For me, sometimes I’m a /njus/(nyoos), sometimes I’m a /nus/(noos). There must be a pattern as to the situations in which I drop my yod, but I’ve never thought to collect data on it. Maybe I will.
It’s very trouble-free to find out any matter on web as compared
to textbooks, as I found this article at this web site.