I've recently been banging out a fair few new limericks for the OEDILF and in the process been both reading a lot of the existing ones and using Rhymezone.

It's reminded me of how far apart US and UK pronunciations can be. A significant proportion of the eighty-thousand-ish limericks in the database simply do not and cannot rhyme for me.

To use some examples from Rhymezone rather than from the limericks.
(Remember I'm talking about how these things sound to me - I accept that they may be different for you.)

hot certainly rhymes with blot, dot and got but it most definitely doesn't rhyme with fraught, taut, or overwrought.

carry and tarry both rhyme with marry but how on earth can they also rhyme with either merry or Mary as all three words have distinctly different vowel sounds?

hurry and blurry? Really? They rhyme?

One limerick (which the author assures me rhymes for him) has buy-in/dye in/cyan.

Do all these examples rhyme for you?
Are there any others - things that rhyme for you but wouldn't for me?
What about the other way round - things that rhyme for me but not you?

I had an editor query my rhyming of here/disappear/tear.
Also queried were sort and thought, and one and none.

Do any of these rhyme for you?

I find it almost impossible to read many of the US pronunciation limericks and I am sure the same must be true in the other direction.

I'm not saying one is right and the other wrong (I know better than that. ) but I do wonder just how many of these differences there are.

This message has been edited. Last edited by: BobHale,

Despite being American, I've no quibbles with your assertions, Bob. I guess we just don't all hear the same sounds, but I do hear the rhymes that you do in your examples. Maybe it's my English name that makes me hear things your way?

Bob. There are many difference between UK and US English, as well as differences within those regions between dialects. In the States, the cot-caught distinction is being lost in many regions: this is so noteworthy that it has a name in phonology, i.e., the cot-caught merger. (I still pronounce the vowel in cot differently from caught.) Then there's the wine-whine merger. (I do not pronounce these differently. Then there's the Mary-marry-merry merger. For some all three are different, mainly in the NE of the USA, but for others all three have merged.) (I distinguish between Mary-merry on one hand and marry on the other.) You might want to look at some Wikipedia articles on English phonology (link).

This reminds me of how Early Modern English critics thought Chaucer was a bad poet because he oftentimes got the rhyme wrong and sometimes his meter was out of whack. Well, they simply did not realize that Middle English was different from Early Modern English.

Also, here, disappear, and tear (the noun) all rhyme for me, but tear (the verb) has a different vowel. So, to answer your questions, Bob, yes, there are a whole bunch of systematic differences between US and UK English.

Oh, and I nearly forgot: there's also a pin-pen merger (in the Southern States). I pronounce them differently but many in the US don't. And, the latest development in the States, the Northern Cities Vowel Shift (link).

[Fixed the text.]

This message has been edited. Last edited by: zmježd,

The prblem that was raised with here/tear(noun)/disappear seemed to be that there was a rhotic r in tear and disappear but NOT in here.
tea-r and appea-r but he-ah
Wine/whine hasn't come up. I'm half way through that merger. I sometimes prononce them differently but more often, in fast or casual speech prononce them the same.

I'm familiar with the cot-caught and marry/merry/Mary mergers already but it's only when you start trying to rhyme those vowels for a limerick (or any other type of rhyming poetry) that you really notice it.

The prblem that was raised with here/tear(noun)/disappear seemed to be that there was a rhotic r in tear and disappear but NOT in here. tea-r and appea-r but he-ah Wine/whine hasn't come up. I'm half way through that merger. I sometimes prononce them differently but more often, in fast or casual speech prononce them the same.

Well, I hadn't heard about that. Is this a US dialect or UK one? (My speech is totally rhotic.

I'm familiar with the cot-caught and marry/merry/Mary mergers already but it's only when you start trying to rhyme those vowels for a limerick (or any other type of rhyming poetry) that you really notice it.

Well, that what you get when you try to write limericks in more than one 'lect. If people weren't so anal (a nice word for it) about getting the rhyme right, you might have a chance. Does the Oedlif have published rules about a standard dialect to write limericks in? Maybe they could have separate but equal sections of the website dedicated to different national standards and such.

The rule is that every contributer writes in his own dialect and that if they say it's right then it's right. You can ask the question but should respect the answer.
I wasn't criticising the question, more just interested by the varieties.

Hell I have a series of limericks written entirely in my dialect (Black Country) which is a pretty far throw from any standard form.

Here, for example, is one that I haven't submitted yet. (The submitted version will include a translation.)

Got the ague, ah was ditherin' wi' code.
Wore sure o' the way tha' ah nowd.
Un wi' mah gammy foot
Well, ah wore 'alf 'ard put
Ter mek me way wum dahn the road.

quote:

If people weren't so anal (a nice word for it) about getting the rhyme right, you might have a chance.

Bob apparently disagrees, but I think people are beyond anal over there. And recently they apparently have a major "grammar nazi" whom I would love to come in contact with. I'd send him a few Language Log and Wordcraft entries! He sounds like a Strunk and White type to me. Unfortunately, a bit of that "anality" has come to WC in our limerick game.

We've talked about some of these rhymes before, but I don't say "merry," "Mary" and "marry" differently. Perhaps I should, but I don't. To me, "cot" and "caught" are hugely different. On the other hand "tear" (noun), "disappear" and "here" rhyme perfectly, but I bet it's different for the East coasters. "Hurry" and "blurry" are perfect for me. "Buy in," "cyan," and "dye in" don't work. "Sort" and "thought" are hugely different for me, but "one" and "none" works. Then of course there's the whole "how many syllables" question. For example, Bob, you questioned my saying "doctorate" with 2 syllables. Normally, however, I've found it's the British who say words with fewer syllables.

That's why OEDILF is so hard!

quote:

Originally posted by Kalleh:

quote:

If people weren't so anal (a nice word for it) about getting the rhyme right, you might have a chance.

Bob apparently disagrees, but I think people are beyond anal over there. And recently they apparently have a major "grammar nazi" whom I would love to come in contact with. I'd send him a few Language Log and Wordcraft entries! He sounds like a Strunk and White type to me. Unfortunately, a bit of that "anality" has come to WC in our limerick game.

We've talked about some of these rhymes before, but I don't say "merry," "Mary" and "marry" differently. Perhaps I should, but I don't. To me, "cot" and "caught" are hugely different. On the other hand "tear" (noun), "disappear" and "here" rhyme perfectly, but I bet it's different for the East coasters. "Hurry" and "blurry" are perfect for me. "Buy in," "cyan," and "dye in" don't work. "Sort" and "thought" are hugely different for me, but "one" and "none" works. Then of course there's the whole "how many syllables" question. For example, Bob, you questioned my saying "doctorate" with 2 syllables. Normally, however, I've found it's the British who say words with fewer syllables.

That's why OEDILF is so hard!

The way I workshop at OEDILF (when I do any at all) is to ask the question and acceptate the answer.

For example I asked about "doctorate". You said it has two syllabels and I just RFAd it.

Others may take a harsher stance and I see that at least one recent member has been driven away by what was really a fairly mild criticism (especially by OEDILF standards.)

I find, and have always found, that if you stand your ground, for every WE who won't RFA there is another one who will. Are some over-particular? Yes. Are they all over-particular? No.

Anyway, re the rhymes question.
I'd love to hear what vowel sound marry, merry and Mary have for you. Could you pick it out on an IPA chart for me?

Ditto with blurry and hurry.

For me the vowels are, respectively,

marry/merry/Mary æ/e/eə
hurry/blurry ʌ/ɜ:

Sounds can be heard here.

http://www.antimoon.com/how/pronunc-soundsipa.htm

Well, if I understand the IPA chart correctly, with merry/Mary/marry, the sound is: eəʳ

With Blurry/hurry it's ɜ:ʳ

As for OEDILF, I hadn't been there in awhile, in all fairness. However, I have recently returned because of CJ's plea, and I found that one of my six-year-old limericks was being picked apart (of course it had already been RFA'd and accepted) because of the pronunciation of baumannii in Acinetobacter baumannii, which is a gram-negative bacillus. I wrote the limerick with a microbiologist who no longer posts. He was responsible for the pronunciation as I have never heard it in speech before (just read about it). The workshopper wanted a reference, so I provided two; one supported our pronuncation and the other supported the workshopper's. Well, of course the workshopper didn't believe the one supporting ours and would only accept his. He wanted it to sound like radii. So, with my co-writer far gone, I gave in and allowed the change...which also meant the whole limerick had to change because of the added vowel sound. (ee vs. ee-eye) It was just annoying and time consuming, that's all.

The other situation was that new workshopper who is a raving prescriptivist, I hear. I received a complaint about him from a limerick poster and asked CJ about him. It was CJ's opinion that this new workshopper is ensuring quality of the limericks.

Both pronunciation (particularly of medical words) and style (I hesitate to say "grammar") are quite flexible and not as black and white as many seem to think there, such as the "whichs" versus "thats" or the ending a sentence with the preposition or so many more. I am sure, for example, that many pronounce baumannii as we originally had it.

Well, I guess my vowels are closer to the British pronunciations than I'd realized.

For me, "marry" sounds like "tarry" (ae)
"merry" has the short e from "bed"
"Mary" has a sound halfway between the long a of 'eight' and the short 'e' of 'bed' (not quite as wide as a New Englander's May-ree)

"hurry" has the same short u as "cup" (upside-down little v); "blurry" and "turn" have the same vowel sound for me, but not as closed & R-sounding as the audio for ɜ:ʳ

"cot" is like "rock" and "caught" is similar to "four" (can't do the symbols)

My accent is not precisely regional; tho raised in upstate NY (where the Northern Cities vowel shift makes Syracusans sound similar to Chicagoans), I am not many generational yrs distant from the marriage of a Nova Scotian to a North Carolingian. Their vowels were quite similar in the early 1900's; most of the clan born before 1950 has what you might call a "clipped drawl". The accent drops off among my siblings; the youngest (born in 1965) has the typical Northern Cities accent-- despite many corrections from Grandma regarding the marry/merry/Mary differential.

Back in the '50s, there was a radio show where a linguist talked to audience members and then, with a great degree of accuracy, described where they had lived (not where they were born). In a discussion about his ability, he mentioned "merry, marry, Mary" as an indicator of soneone living above the Mason-Dixon line. Northerners made three different sounds while Southerners pron them all the same way.

I think that person would have a more difficult task today since TV and radio works to homogenize local accents.

I think that person would have a more difficult task today since TV and radio works to homogenize local accents.

One often hears this, but without any evidence being offered.

I've lived in New England for my entire life and there was a great diversity of language accents when I was young. In recent years, I have noticed the Down East accent is less prominent in Maine than it used to be. Certainly not the best of evidence, perhaps, but it is my personal experience.

I like z am somewhat skeptical about TV's influence on accents; it's not uniform, & even watching a few hrs a day,one is not immersed in it as in family, school, work. Proof I also remember a variety of accents in the upstate town where I spent my 1st 21 yrs-- not sure if this prevails, but my impression on extended visits is that the accent is more uniform & can be found in many upstate places.

I would say that the town in the early 50's (& before, judging from elder relatives) was subject to a high degree of insulation, even from one rural area & town to the next.

My mother's clan may have developed & hung onto a family accent partly because of their experiences in the Depression; the entire family on her father's side lived together in 1 big house (divided into littleareas for each unit) for her formative yrs. When both her parents were on the road for the WPA, she spent months at a time in the other grandparents' home, which was similarly arranged.

Perhaps a sort of regional blending took place once the work force became mobile & nuclear living the norm. Starting & the late 50's & accelerating thereafter, the large businesses in town became nationalized, transferred workers here & there, closed/ re-opened etc. By the mid-60's I knew many kids who had arrived from somewhere else for this reason, which was rare among my childhood acquaintances.

quote:

Originally posted by zmježd:
I think that person would have a more difficult task today since TV and radio works to homogenize local accents.

One often hears this, but without any evidence being offered.

At a guess (again without any evidence) I'd say that the relative ease of transport nowadays contributes to the decline of local accents and dialects. The rise of the railway and the car means that it's much easier and more common to move out of one's birth area.

Before trying to work out why there is an homogenisation of local accents, I'd want to see some evidence that there has been such an homogenisation. I can't speak for the US but I can tell you that from my front door in the UK a ten mile journey in any direction and I start to notice the changes in accent. It doesn't, and I realise this is just as subjective a statement, sound very homogenised to me.

quote:

Northerners made three different sounds while Southerners pron them all the same way.

Not this Northerner.
"Evidence" of homogenization" would be hard to come by unless there were large samples. One person here and another there hardly does it. How do linguists collect this sort of evidence? It would be interesting research, I'd think.

What some see as homogenization might be simple change in a local regional accent. The Northern City Vowel Shift that I mentioned in my first reply to Bob is evidence of a new regional accent or accents developing. If folks on the TV are speaking one way and others around you a different way, I'd say you might talk to the TV in the standard TV version of English, but to your friends in the local regional dialect.

I have a feeling that "evidence" to linguists is different from "evidence" in my field. It would have to be a well-designed study before we'd call it "evidence." Makes sense, though, that they'd be different.

quote:

Originally posted by Kalleh:
I have a feeling that "evidence" to linguists is different from "evidence" in my field. It would have to be a well-designed study before we'd call it "evidence." Makes sense, though, that they'd be different.

I don't think so. By evidence I also mean a well-designed study.
I'd want to see samples from various regions acroos various time-periods with an analysis of regional variations and some statistical indicators that there are fewer regional variations now than previously.

I have a feeling that "evidence" to linguists is different from "evidence" in my field. It would have to be a well-designed study before we'd call it "evidence."

Well, Professor Labov is a famous sociolinguist. I have read some of his papers and seen him interviewed. I've only heard him talking about the Northern Cities Vowel Shift, and giving a summary of the vowels affected. We also know how people in the area affected spoke 50 years ago (even to the point of having recordings) and we can compare that with how they are speaking today. There's a difference between the two. Would you consider that to be evidence? Note also that the changes in the area affected is not a convergence with the phonology of General American English, but away from it and the local regional accent(s). The phonology of General American English is well understood and documented.

So, if Labov's findings are not evidence in the scientific sense (which I would think would include medical evidence), where did he go wrong? Is the Northern Cities Vowel Shift not happening? Is there some linguistic conspiracy afoot to mislead the media? What would be evidence that you would except?

If you check out the references at the bottom of the Wikipedia article, you might find some with evidence you might consider to be evidence. In fact, this handout (link) has some extra references in it, too.

While I am at it, why has the UK preserved its Standard and regional dialects? After 50 years of being bombarded by US media (TV and movies), they are still pronouncing things differently (from how they used to pronounce things 50 years ago, but also how we pronounve things today).

quote:

Originally posted by Kalleh:
I have a feeling that "evidence" to linguists is different from "evidence" in my field. It would have to be a well-designed study before we'd call it "evidence." Makes sense, though, that they'd be different.

I think sociolinguists do extensive studies and make dialect maps showing pronunciation differences.

I don't think so. By evidence I also mean a well-designed study.

As I said in my previous post, Bob. You have the beginnings of a small bibliography on the phenomenon. I'd say you could start there. I admit, since I've discussed the NCVS with several linguists (informally) that the consensus amongst linguists is that this vowel shift is currently happening and may not be finished. I admit, as I am not an academic linguist that I do not have time to or money to investigate the areas affected on my own. At this point, until I am given actual conflicting evidence, I'll say that the NCVS is happening.

I'd want to see samples from various regions acroos various time-periods with an analysis of regional variations and some statistical indicators that there are fewer regional variations now than previously.

In the 19th century, many Romantic-Nationalist notions of disappearing dialects (of the same sort, but from different causes) were bandied about. The conclusion of dialectologists was that dialects were alive and being spoken, but that they, too, like the standard languages that surrounded them were changing. The dialects-are-disappearing folks then switched course, and said, yes, we still have dialects, but they are not the "pure" ones of yore. They have become "corrupt" by mingling more the the national standards and dialects from other regions.

The quickest study I could imagine that did not involve travel to the areas would be to look at some phonetic evidence as recorded in language atlases. We have been collecting these data for at least 100 years and publishing the resulting maps in atlases. You can access them in most large university libraries.

[I note Goofy's post got posted while I was editing my last massive bit of drivel.]

I don't doubt for a moment that language change is happening and that dialect change is part of that. What I have yet to see is that language homogenisation is happening. If it is then the dialects should all be converging with the differences being gradually eroded. The changes in dialects should all be moving in the same direction.
Before I start worrying about theories as to why it's happening I'd like to know that it IS happening.

I can give a very easy example from my own dialect.

The negative form of "did" that is commonly heard in my region is "day" but from people of my father's generation "dissn't" is also often heard. This form is virtually never heard by anybody under about seventy years old and will, in the next thirty years likely disappear altogether. But that doesn't mean the dialect is losing its distinct identity - simply that its distinct identity is also distinct from what it was fifty years ago.

First of all, please let me assure everyone that I am not doubting the scholarliness of linguistics. However, I am finding the differences between evidence in linguistics and in medicine to be fascinating. For example, in medicine you would have to publish your results in a peer-reviewed journal to have any acceptance at all of your evidence. You never could cite a "talk" or the like as evidence.

Given that, I did check out the references that z asked me to; they are very interesting. On the Wiki page there were four, with one being a talk, two being books and the fourth being from a journal (Language Variation and Change). In the second link there were seven references with three being talks, one being a poster, one being a book, one being a "working paper" (?), and one being from a journal. One book must be seminal as it is cited all over the place, and that is the Labov et al. (2006) Atlas of North American English. I also found Goofy's link to linguistic maps to be intriguing. I cannot imagine the amount of work it must have taken to develop those 122 maps.

Even the citings are different. I was intrigued by one citation in z's link that said, "Labov, William (to appear)." I wonder what style guide linguists use. I have never seen that type of citation. Apparently the manuscript is to appear in Language. I did find that article to include much evidence, and it had 5 full pages of references.

So I agree with you that linguistics has tons of evidence. The way it's collected and reported, however, is quite different than the way other sciences collect and report data. For example, there must be a lot of observational data collection with linguistics. It just makes sense. There are also excellent observational studies in nursing, but often they are not given the respect they deserve. While randomized controlled trials are given the most respect, many of our questions don't allow for that type of study. For example, if you study the effects of cigarette smoking, you cannot assign subjects randomly to smoke or not to smoke and then collect the health outcomes data years lately. Would that we could! Instead, we must do retrospective, quasi-experimental, etc., studies, which are not as strong. But that's life with practice research. The same is true for linguistics. You cannot control the variables with dialect speaking (such as moving to different areas and coming back, personal differences in dialect uptake, etc.).

First of all, please let me assure everyone that I am not doubting the scholarliness of linguistics. However, I am finding the differences between evidence in linguistics and in medicine to be fascinating. For example, in medicine you would have to publish your results in a peer-reviewed journal to have any acceptance at all of your evidence. You never could cite a "talk" or the like as evidence.

Please don't judge linguistics by my citing some references from a Wikipedia article and some other online documents. If I wanted to research the Northern Cities Vowel Shift (NCVS) in-depth and systematically, I might track down the references pointed to online by looking at papers in refereed journals, then also take a look at bibliographies in such reference works as Linguistic Bibliography (link). Linguists, depending on their specialization, tend to use either sociological styles guides or anthropological ones.

Given that, I did check out the references that z asked me to; they are very interesting. On the Wiki page there were four, with one being a talk, two being books and the fourth being from a journal (Language Variation and Change). In the second link there were seven references with three being talks, one being a poster, one being a book, one being a "working paper" (?), and one being from a journal. One book must be seminal as it is cited all over the place, and that is the Labov et al. (2006) Atlas of North American English. I also found Goofy's link to linguistic maps to be intriguing. I cannot imagine the amount of work it must have taken to develop those 122 maps.

I'd say Labov's Atlas of North American English would be a good place to start, too. Linguistic atlases are about a 100 years old (as I said above) and well-understood by their intended audience.

Even the citings are different. I was intrigued by one citation in z's link that said, "Labov, William (to appear)." I wonder what style guide linguists use. I have never seen that type of citation. Apparently the manuscript is to appear in Language. I did find that article to include much evidence, and it had 5 full pages of references.

Yes, linguists tend to use different style guides than medical professionals. Working papers are early versions (pre-print) of articles that are circulated for feedback. some times they are cited if well-known and distributed. I have a book by Chomsky that was circulated as samizdat for a couple of decades before it finally saw print. Language is one of the premiere academic journals of Linguistics. It has been in print since the '20s of the last century and is published by the LSA (Linguistic Society of America).

So I agree with you that linguistics has tons of evidence. The way it's collected and reported, however, is quite different than the way other sciences collect and report data. For example, there must be a lot of observational data collection with linguistics. It just makes sense. There are also excellent observational studies in nursing, but often they are not given the respect they deserve. While randomized controlled trials are given the most respect, many of our questions don't allow for that type of study. For example, if you study the effects of cigarette smoking, you cannot assign subjects randomly to smoke or not to smoke and then collect the health outcomes data years lately. Would that we could! Instead, we must do retrospective, quasi-experimental, etc., studies, which are not as strong. But that's life with practice research. The same is true for linguistics. You cannot control the variables with dialect speaking (such as moving to different areas and coming back, personal differences in dialect uptake, etc.).

Collecting linguistic data is also pretty well understood. One of the classes I took all those years ago was a field methods class, where we were introduced to the ins and outs of collecting such data. Yes, the background of the informants, such as educational background, the linguistic area the person was raised in, etc., are important. Depending on the money and resources available, the evidence presented in the atlases can be replicated or not. One doesn't just come up to people in the mall and ask: how do you say dog? One of Labov's more famous studies was based on data collected at a couple of different department stores in NYC where speakers from different sociolect groups were asked where the patron could find a certain item. The item was not the interesting bit; it was the location of the items: the fourth floor. Labov was interested in rhotic and non-rhotic sociolects, and how people change their pronunciation when the person they are speaking to may or may not be in the informants group. He was also looking into errors of hypercorrection.

I've taken part in psycholinguistic studies and experiments and read papers on enurolinguistics in which things get a little bit more like what you may be talking about, Kalleh. Linguists are interested in aphasia and do some testing of brain surgery patients who are bilingual and such.

This article that I posted about in another thread is very much like any medical research study, even with NIH ROI funding. So I am convinced. The evidence, as it is in most disciplines, is just quite varied.