[ul]
[li]So Person A perceives Person B as having an accent.[/li][li]This perception arises because of the specific way that Person B produces vowel sounds.[/li][li]Those specific differences in pronouncing vowels are minimized (often to the point of being impossible to detect) by the act of singing itself, because the act of singing (drawing out the vowels) interferes with pronouncing vowels in that “different” way. (And that goes for all “different” ways: Brits, Aussies, Bostonians, Texans, New Yorkers, South Africans.)[/li][li]It doesnt’ matter which “different” way a person produces those sounds to start with, if you * minimize* it they end up sounding more alike than different and all of them sound like they are pronouncing things the same as they are pronounced in SAE.[/li][/ul]
Therefore it seems self-evident that SAE is a way of speaking English that *inherently minimizes differences.
*
And differences are understood to be what defines accents.
So SAE is therefore free of accent, or the closest to being free of accent of all the ways that English speakers speak.
In other words, neutral: no particular kind or characteristics, no extremes.
As for whether a highly trained ear can detect that a change in the way SAE speakers sound when singing, it doesn’t change the fact that if everyone whose regional differences are minimized when singing end up sounding pretty similar to one another in their pronounciation, and that “regional differences minimized pronounciation” also sounds very much like the pronounciation of SAE speakers, there doesn’t seem to be anything inaccurate in saying that SAE is the closest thing to a neutral (i.e. free of vowel producing particularities) of all ways of pronouncing English.
And country music? Every singer seems to default to the same country accent when singing country songs. Does that prove that the country accent is the other unaccented accent?
The convention in Australian music has generally been to sing with an affected American accent, but in recent years there have been more popular artists emerging who sing in their own accent. Some examples: Frente! (hi hawthorn - long time no see!), ThirstyMerc, SarahBlasko, LisaMitchell. Aussie Country artists like JohnWilliamson and SlimDusty have always sang in their true accents.
This is an excellent point. My family moved to Wisconsin from the South when I was a teenager, and I encountered a number of people who told me that people from Wisconsin have “no accent”, that they spoke “standard English”, and that the Midwestern accent is how newscasters on the national news speak. Many of the people who said these things to me had, to my ears, very obvious non-standard accents that didn’t sound like anyone on television, unless you count broadcasts of Fargo.
I was also surprised to learn that my own accent, which was considered pretty neutral in the South, did not allow me to blend in with Midwesterners but instead made them want to ask me where I was from.
(Bolding mine.) The bolded portion does not follow from your cite. It says that people who are singing tend to pronounce vowels in the same way as other people who are singing, not that they pronounce vowels in the same way as someone speaking with a particular American accent. It is your perception that these sung vowels sounds like SAE, but there’s more to an accent than just vowel sounds and I’m not convinced that sung vowel sounds are really the same as in SAE.
When I lived in Japan I noticed that when they had a musical act performing on television the song lyrics would often be captioned karaoke style at the bottom of the screen. On a number of occasions I was surprised to learn from the captions that some of the lyrics in a song were in English, because I sure wasn’t hearing anything that sounded like English. There was one song that repeated the phrase “hit parade”, but it sounded more like “heat puh lay” to me. So the Japanese at least are quite capable of singing English words in a way that was largely unintelligible to me as a native speaker of American English.
This is nonsense. If there exist accents that are different from SAE then SAE must be different from those other accents as well.
More examples of not-SAE singing: Lacuna Coil (Italian band, singing in English), The Shoes (French band, singing in English).
Lamia, I’m from Wisconsin but have lived in Illinois for nearly a decade; native Illinoisians can tell I’m from there within minutes of me starting to speak, even though I thought I had eliminated most of the more obvious speech markers from there. Meanwhile many friends from the East Coast can’t tell the difference between my husband (native Chicago suburbanite) and I in our accents.
By what mechanism do you think accented English evolved to become unaccented in our country? Natural selection or something? It should strike you as mighty improbable that such as “mutation” could have occurred spontaneously. It’s about as improbable as water flowing upstream.
If everyone on the planet spoke the same language, with the same accent, then we’d all sound “neutral” to each other. It’s only when you have other accents out there, that different accents become perceptible. So by definition, SAE is accented. Can’t Brits recognize Americans by the way they sound? Don’t we sound foreign to their ears? Using SAE as the reference point for determining how “extreme” other accents are is completely arbitrary.
Aha! This is the first time you’ve actually proposed what this neutral accent might sound like. Unfortunately, it’s an own goal. Standard American English, like most varieties of English, is not characterized by an abundance of monophthongal vowel sounds. For example, if a SAE speaker says the words bate, boat, boot, *none *of those will be monophthongs, although boot will come closest. If you really were to pronounce all those vowels strictly as monophthongs, you would definitely not sound like you were speaking SAE. (To my ears, it sounds like an Irish accent, if anything.)
You see, you’re confusing this quote’s (somewhat odd) use of “pure” to mean “monophthongal” with an imagined “neutral” language variety that is “pure” because it isn’t a dialect and has no accent – even though there is no such variety.
It’s important to understand why. You seem to have the idea (and it’s no shame – it took me, oh, four years’ worth of a linguistics B.A. to learn this stuff) that there is a “neutral” variety of English (and presumably other languages), and that when someone speaks it, they realize English phonemes in a “neutral” or “pure” way or a way that is in some way how they’re “supposed” to be realized.
Or, if that person is speaking with an “accent,” they will accrete features on top of that idealized realization, which will obscure those “pure/neutral” phonemes and diverge from the “neutral” variety. Or as you say:
This is linguistically meaningless. Speakers of a language have a more or less common inventory of the language’s phonemes – but phonemes are not sounds. They’re abstract mental representations. So there’s no way of saying that a particular pronounced sound (a phone) is a more or less “accurate” representation of that phoneme. Every variety (or “accent”) contains a set of ways that the language’s phonemic inventory can be realized. SAE contains one such set; other accents consist of other sets. You are always using some set or other.
So it’s meaningless to say that SAE is a “neutral” accent, and even more meaningless to speak of someone being “free from” accent. There is no concept of a “neutral” accent in linguistics.
Although in some ways the accent of rock ‘n’ roll is (or was) an American drawl there are, as many others have cited, plenty of rock and pop artists that don’t affect an American accent when they’re singing.
But whether genuine educated linguists agree that there can be no truly “neutral” accent or not wasn’t really my point; I pretty much knew that going in- if they were in agreement with my view it wouldn’t be an argument.
My point was more about actual perception by all the English-speaking humans themselves; this seeming agreement that it’s just crazy talk to suggest that any one way of speaking might be less accented than another, much less that any one in particular might be the least accented of all! is driven more by a kind of political correctness than the practical reality of the way people hear each other.
You just skipped over the statement from the (presumably educated and knowledgable) folks at the National Center for Voice and Speech who wrote:
Which I demonstrated led to my conclusion. Instead you jumped to the conclusion and called it nonsense, because you are a linguist.
But what about the plain logic?
I should first ask if you accept the statement…does singing, because it draws out the vowel sounds, minimize (perceptible) differences in the way people pronounce words, making them sound less accented? If you disagree that singing does this, then I guess the rest would be nonsense and there’s nothing else to discuss.
But if you agree that it does minimize differences, then I assume you would also agree that such minimization of difference would certainly explain the widely held perception that many singers seem to “lose their accents” when they sing.
So if you’re still with me, we’ve established that the very act of singing itself can dilute the strength and perceptibility of accents of all kinds, haven’t we?
And you stated somewhere in here I believe that this holds true for people who speak SAE as well, and I’m sure you’re right. But what I’m also sure of is that any such loss in SAE speakers goes almost entirely unnoticed by pretty much everyone, where the loss/change of accent in non SAE-speaking individuals is widely noted, questioned, and hypothesized about.
And in fact, the idea of “losing” one’s accent and “singing like an American” are used interchangably by the vast majority of people considering the issue. (Check that search again…I didn’t put the word “American” in it at all)
You contend that that my conclusion is flawed; that rather than “American” (by which I mean SAE) being (relatively…and that means something, you know) free of an accent, it reflects only the wide acceptance of SAE as a norm, which is a good argument and an idea that needs to be taken seriously. (I absolutely accept and agree that there is a strong element of truth in that, I disagree mostly about why it’s so.)
So back to the idea that the very act of singing interferes with one’s accent. If there is no reasonable way to agree that some (approximate) “neutral” pronounciation exists, wouldn’t that mean that the act of singing, in interfering with one’s natural accent, would make people sing all kinds of different ways rather than one particular way?
But since we know it doesn’t, how would you characterize the fact that the “minimization” of difference leads everyone with those differences towards sounding like they are singing with an SAE accent?
So again: if the act of singing English itself minimizes pronounciation differences, and the singers whose accent differences are thus minimized collectively find themselves using the** same alternative pronounciation** to replace their widely varying natural accents, and there is an existing spoken accent that sounds very much like the way all those singers sound when their differences are minimized, then the existing spoken accent must, by definition, have very little in it that most untrained ears would perceive as an accent, because accents are defined by the differences, and if taking away differences makes everyone sound like an American speaking SAE, then SAE must be free of much in the way of detectable differences.
It seems to me that the only way this is not true is to reject the starting premise that the very act of singing itself has the power to alter someone’s accent in any manner that isn’t an active choice. But there isn’t a lot of evidence for that, and the evidence I did find for the topic went the other way: that it is actually difficult to consciously retain one’s accent when singing, which I ran into quite a bit, as when a New Zealand student took up the issue and concluded:
He seems to be guessing that everyone ends up sounding like Americans as a result of listening to Americans, but as a linguist yourself, you don’t buy that I’m sure: that somehow people lose control of their voices as a result of listening to the radio AND that they all develop the ability to discard their accents when many couldn’t do it to save their own lives when speaking? That’s pretty far-fetched.
No, the universality of the phenomenon of accent loss when singing suggests that there is a built-in baseline to English and that whatever accent SAE is, it’s a pretty pathetic excuse for one, lacking in any identifiable characteristics whatsoever, making it functionally no accent at all.
As I said, American accents are/were the accent of rock. If you get people from any english speaking country to try to sing say an Irish folk song, many will, maybe even unconsciously, adopt an Irish accent, as this to them is the sound of folk singing.
Anyway I would say the “rock accent” insofar as there is one is not the same as SAE anyway. It’s more rooted in Southern American accents and AAVE.
I’ve recently been listening to various recordings of Les Miserables, and it’s quite easy to tell which were recorded from Broadway productions, and which were recorded from London productions. Same music. Same words. Different accents. Check it out and you’ll see.
(My contribution to non-American singing: The Beatles.)
It’s your second bullet point here that is wrong. The perception arises because of the specific way both A and B produce vowel sounds when speaking. Notice that in this situation, B probably also perceives A as having an accent, except when they sing.
What happens when they sing, is that A and B sing alike and speak differently. And of course, the singing is not like either of their speech, since it’s singing. Of course, each perceive singing as closer to their own speech (see post 33, for example), since they both perceive it as clear.
I don’t know whether it does or doesn’t. However, your quote doesn’t say what differences are minimized, and mentions only vowel length. It doesn’t refer to how the vowel is articulated, what phone it is realized as, whether it’s realized as [ʌu] or [aʊ] or what. When people perceive my speech as being Canadian by the way I say “house,” the latter is the difference they’re referring to.
However, let us assume arguendo that something about singing tends to flatten the differences between vowel sounds. This still doesn’t account for the fact that hordes of singers, as has been demonstrated, sing with perceptible accents.
But the question remains: is this perception accurate? If it’s accurate, to what extent and with reference to what phonetic features? In what way does it do so? Do the sounds actually converge, or do they only seem to do so? (For example, do they actually form a range that simply seems to be more uniform than it really is?) And how does this account for the fact that plenty of people do sing “with accents?” And how does it behave cross-linguistically?
But people do sing in all kinds of different ways. As has been pointed out. And I don’t think it’s been demonstrated that people really do converge towards a fixed point – only that people think they do, which relies on perceptions that are easy to fool.
And here’s the crux: that isn’t in evidence. What’s in evidence is that people think it does. For example, as I pointed out above, if singing really is characterized by monophthongization, this causes pronunciation to become less like SAE, not more.
As Lamia suggests, if vowel sounds really do converge in song, perhaps it’s the case that you simply associate that with another convergence (the sociologically motivated convergence in American English) that exists, especially since pop music is culturally linked with the US, and you think that people sound American when in fact, phonetically speaking, they don’t; and there’s also Jamaika a jamaikaiaké’s suggestion as well (convergence in singing could cause you to think the person now sounds more like you, when they don’t).
In addition, if pop singers converge on a particular pronunciation, and the most famous and influential pop singers come from America, we’ll start to think of that pronunciation as “American” regardless of how many phonetic features it actually shares with SAE.
But I think that, to the extent pop singers end up actually pronouncing things like Americans, the reason is here:
Why not? Singing, especially at that level, is in many respects a histrionic behaviour that requires significant and forced manipulation of the vocal apparatus over a short period of time, conditions that don’t exist while speaking. So there’s another possible explanation: perhaps if you’ve learned to perform that detailed vocal operation by listening to Americans (or people in turn imitating Americans), you may well end up (objectively) using American phonetics. (Or phonetics associated with Americans, as above, whether or not they actually sound like SAE.) By the same token, American opera singers don’t sound like Americans when they sing in English, they sound like opera singers, because of the particular way opera singers are taught to pronounce things. Or as cazzle, hajario, and hogarth point out, country music, reggae, and hip-hop have their own pronunciation. If what you propose were the case, none of these phenomena would occur.
At any rate, it stands to reason that if you learned to sing by listening to Americans, and as a consequence sing like an American, it might well take effort to stop doing so.
SAE has exactly as many phonetic features as every other lect of English. Each and every English phoneme has a way it is realized in SAE, just as it has a way it is realized in each other lect. What could it possibly mean for American English to have “no characteristics”?
Stoid your problem is that you are failing to see the effect of your own bias upon your perception of a neutral thing.
If you were wearing very slightly red tinged glasses looking at a white wall, it would look red. You are concluding that this means the wall is red. This is obviously nonsense. If I stood in front of the same white wall wearing very slightly green tinged glasses, the white wall would look green.
You have given several cites that say singing strips out regionalisms. You say singing sounds like your own accent. You conclude that this means your accent has no regionalisms. What you are overlooking is the more likely conclusion namely that when we hear something stripped of regionalisms, we perceive it to sound like ourselves.
None of us hear our own accents. We sound neutral to ourselves. So when we hear something that sounds neutral to us (no regionalisms) we perceive it as sounding like us.
Leaving aside singers who manage to inject their regional vowels into their singing, singers to me sound like me ie neutral. So clearly an Australian accent is “no accent”.
Your entire OP would work equally well if it were written by an English or Australian or South African person from their perspective, but to an entirely contradictory conclusion than that which you reach.
For one thing, it seems to me like an unaccented form of a language developing several centuries and an ocean away from its origin goes against common sense. I may be wrong about this, as I am no linguist.
But I’ve read through all the posts and I’m a little unclear as to what “Standard American English” sounds like. Can someone give me an example of someone who speaks in Standard American English that I may be able to use for comparison? I’m not trying to deny its existence, but every speaker and every singer I’ve ever heard sounds to me like there is some kind of idiosyncratic pronunciation present in some of the things they say, whether it’s identifiably regional or not.
Minimizing the differences between vowel sounds is not the same thing as removing an accent. Vowels that all sound the same are as much of an accent as vowels that sound wildly different.
When white guys sing the blues, though … they still sounds pretty damn white, or rather, SAE. Strange how white rappers will affect AAVE, but white blues singers sound about as different from old black men as you can get.
On a related note, it can be interesting to hear the accents of educated people who are very fluent in English, but as their second language. When I hear Dutch people speak English, it sounds a LOT more American/Canadian than British. However, if you can cut through the sing-song of an “Indian newscaster” accent, there’s often some Brit behind it.
What seems to be happening here is conflation of the concepts of ‘neutral’ and ‘normal’ (or possibly ‘average’).
There is no particular reason, logical or otherwise, why the normal case of anything need be a neutral one. For example, if 99% of people adore chocolate and 1% dislike it, the normal position is not neutral indifference, neither is the average.
So yes - a standard American English accent is normal for popular singing, but that doesn’t mean anything other than what it says. This is like insisting that Mean, Median and Mode all mean the same thing.
No wonder. To minimize your second language accent you have to listen a lot to that language. People in Dutch- and Scandinavian-speaking regions of Europe generally have less of an accent than other non-native-English-speaking Europeans. These are also the only language regions (that I know of) in non-English-speaking Europe where foreign language movies and TV shows are subtitled, not dubbed, so we get to hear a lot of English just by watching imported TV shows and movies.
Then consider that most English-language TV shows and movies shown in Europe - perhaps excluding the UK - are made in the US. Which English dialect will have the strongest impact?
There are also fairly obvious historical reasons why the British variety of English has had a rather more significant influence on Indian English than the American variety…