Sociolinguistics wtf?

[Torco]

Elsewhere it has been asked

So what made them suddenly change between two quite different sounds one day?

This raises an interesting question

do we know anything about *why* sound change happens?
I mean we know some stuff about how it happens. more or less regularly, gradually, lenition's more common than fortition, some SCs are more likely than others, like 4>m>T_l isn't *impossible*, its just very unlikely.

But why do people suddenly change their speech and, most importantly, why does it stick? why do some or other trait of a guy's ideolect spread to other individuals? what's up with that?

[faiuwle]

This is probably a huge oversimplification, but AIUI, the two major motivations are 1) shit's hard to pronounce, so people start doing it a shorter/simpler/easier way, and 2) words start to sound the same, so people make special efforts to distinguish them, which usually has the opposite effect.

[Yng]

Yes. All sorts of different things. Poor reproduction of sounds (thus [θ] > [f] etc) that enough people do at a given time for it to become an accepted pronunciation, ease of pronunciation (thus lenition, anticipatory changes, assimilation and dissimilation). People change their speech to mimic a prestige dialect, too, obviously - there is a tale of some village up north where the local notables suffered from a speech impediment, leading to an extremely strange rhotic being used.

[linguoboy]

Torco, let me make an evolutionary analogy to make sure I understand your question correctly: You're not looking for an explanation of how mutations (i.e. actual sound changes) arise, but for what drives the equivalent of natural selection, right? Since there's no obvious evolutionary advantage of saying [R] rather than [r] like there is for having brown eyes rather than blue eyes.

Isn't there an entire branch of sociology devoted to explaining this mechanism? I remember reading an article years ago[*] that talked about a sociological study which focussed on the spread of improved seed varieties among American farmers in a rural community. The researches ended up segmenting the population according to whether they were innovators, early adopters, late holdouts, and so forth. This became the basis for a thousand marketing studies, mostly aimed at identifying trendsetters and figuring out how to your product into their hands so that it would quickly spread to the bulk of the target population.

When I saw this model laid out, my first thought was how well it would explain linguistic change as well. I just know what research has been done in order to confirm that it does.

Edit: Here's a recent paper on just that subject from researchers at the University of Murcia: http://www.um.es/dp-filologia-inglesa/i ... -Conde.pdf.

[*] Looks like what I came across was a summary of Everett Roger's 1962 work Diffusion of innovations.

[zompist]

I vaguely recall that Labov has investigated sound changes to the level of finding individual culprits, and his results do seem quite parallel to the kind of person-by-person adoption linguoboy is talking about.

My notion is that people adopt sound changes for the same reason they adopt idioms or lexical changes: they sound cool. How can a sound change be cool? By being adopted by a cool person.

(One difference with sound changes though: there seems to be much less interest in adopting sound changes heard from the media. The cool person has to be a local.)

[Salmoneus]

I don't think that's true at all. Cf spread of cockney/estuary innovations in unrelated and not always adjacent english dialects - h-dropping, th-fronting, l-vocalisation etc. I think it's not about geography at all - it's about adopting the speech of the leaders of the in-group. Often groups have a geographical element, but they also have a class-based element. And in some places probably religious or ethnic elements.

[linguoboy]

I don't think that's true at all. Cf spread of cockney/estuary innovations in unrelated and not always adjacent english dialects - h-dropping, th-fronting, l-vocalisation etc. I think it's not about geography at all - it's about adopting the speech of the leaders of the in-group. Often groups have a geographical element, but they also have a class-based element. And in some places probably religious or ethnic elements.

But people are highly mobile--particularly those who are likely to be perceived as "cool". It would be very revealing to investigate these discontinuous areas and pinpoint who were the first individuals in each to exhibit Estuary features. I'd be very surprised to learn that they were locals imitating what they'd heard on the telly as opposed to transplants.

[finlay]

I don't think that's true at all. Cf spread of cockney/estuary innovations in unrelated and not always adjacent english dialects - h-dropping, th-fronting, l-vocalisation etc. I think it's not about geography at all - it's about adopting the speech of the leaders of the in-group. Often groups have a geographical element, but they also have a class-based element. And in some places probably religious or ethnic elements.

I don't think h-dropping is a london innovation. It's occurred throughout almost all of England for quite a long time. It's certainly tied to class though.

[Cathbad]

I don't think that's true at all. Cf spread of cockney/estuary innovations in unrelated and not always adjacent english dialects - h-dropping, th-fronting, l-vocalisation etc. I think it's not about geography at all - it's about adopting the speech of the leaders of the in-group. Often groups have a geographical element, but they also have a class-based element. And in some places probably religious or ethnic elements.

Leaders for prestige, but also attempts at differentiation (status vs. solidarity, etc.). I guess the reasons would basically be similar to those for any other type of linguistic change (at least sociolinguistically speaking).

Along with innovators, holdouts etc., maybe one should also differentiate sound changes 'adopted' in childhood (as the 'mother dialect'), adopted later in life as part of one's dialect, and sound changes one 'acquires' through 'realizing' what the 'correct' pronunciation is for a given 'standard' (which may or may not be congruent with, say, what the standard was 25 years ago, and can be diffused through various means). There may be different reasons for these types of changes - acquisition, peer pressure, desires for social advancement, or just plain elitism, as in my own attempts to "reverse" the Ljubljana dialect sound change of accented /@/ into /E/ (so even colloquially, I'm now always conscious to say /"m@gla/ instead of /"mEgla/). (This may not diffuse much further than me, unless I become some sort of Paragon for speaking Proper Ljubljana Dialect, but still.)

[CaesarVincens]

People modulate their speech based on many factors (which is why it's hard to say what's going on), but just who you are talking to and where and why can make a major difference.

In my Socio-linguistics class last semester, we discussed many examples from several languages, a few studies by Labov featured prominently.

One major factor for middle to upper middle class is trying to imitate the perception of the upper class speech. That is, social climbers want to sound like they belong with those they want to belong to.

Ultimately, the variations that are always present just become codified as the way one talks; preferred variants are generally those used more often. A banker sounds like a banker because in part he uses the banking register/dialect/whatever so much, even if he happened to have been born in a lower-class.

[Torco]

Well yes, its intuitive that people trying to sound like a banker talk like bankers, and those trying to sound like sportsmen sound like sportsmen, and girls trying to sound sexy try and speak the way they guess they'll sound more sexy. In that regard there's probably something that could be called auditory semantics, like how do certain sounds and sound substitutions are encoded by listeners into paralinguistic information. I mean why does Sean Connery's speech sound so awesome in english, like serious and major-chord-y so to speak? its a similar deviation, from my perspective, to peninsular spanish, also having S for s and a much more open centerpoint for vowels [as in all vowels are more open], and it connotes similar things, larger-than-lifeness, extrovertion, loudness.

But his might just be my own noisy biases... might there actually be sound universals? kinda like [e] sounds smaller than [A] and stuff? it looks like it, buba-kiki effect and all. likely the outcome is complex, and its unfortunate the experiment is buba-kiki and not buba-bibi or kuka-kiki, which would help in determining which element means what.

[Torco]

Okay, so I made the survey

Answer it. NAO!

[LinguistCat]

Okay, so I made the survey

Not related to the linguistics aspect, but I'm a little cheered up after taking this.

[Ossicone]

I wish I could have written in the names. I didn't feel particularly inclined to any of those given, but I made up some pretty cool ones.

[Torco]

Not related to the linguistics aspect, but I'm a little cheered up after taking this.

So the claim that naming little abstract figures promotes good health is true?

most excellent

[zompist]

I don't think that's true at all. Cf spread of cockney/estuary innovations in unrelated and not always adjacent english dialects - h-dropping, th-fronting, l-vocalisation etc. I think it's not about geography at all - it's about adopting the speech of the leaders of the in-group. Often groups have a geographical element, but they also have a class-based element. And in some places probably religious or ethnic elements.

The reason I think sound changes are picked up in person, besides half-remembered Labov, is that it seems clear that people don't pick up accents from the media. You can hear upper-class accents all day long— or any other type of cool accent— from the radio or TV or internet without adopting them.

Some of the changes you mention are so old that they're way past the initial acquisition stage I thought Torco was asking about; today there are millions of in-person exemplars people can pick them up from.

[Torco]

I think the low impact of media in speech has to do with the fact that SCs are registry-dependent. I observe as much in my various registries: I have t4 as an affricate in colloquial but not in formal speech, I aspirate plosives when angry but not when not, and even have different realizations of /s/ depending on my mood and the topic of conversation, so... yeah, its probably because TV people speak TV language, and one doesn't use the TV registry very often.

I've seen accent adquisition vary widely between people. One immigrant I knew indicated that accent adquisition is voluntary: people who want to assimilate take up the accent in its entirety, people who want to reafirm their own identity as foreigners keep their speech patterns those of a foreigner. That's probably true, and is coherent with what has been suggested, but it still doesn't explain the direction of SC.

again, it could be basically random.

[LinguistCat]

So the claim that naming little abstract figures promotes good health is true?

most excellent

Eh, something like that.

[Salmoneus]

I don't think that's true at all. Cf spread of cockney/estuary innovations in unrelated and not always adjacent english dialects - h-dropping, th-fronting, l-vocalisation etc. I think it's not about geography at all - it's about adopting the speech of the leaders of the in-group. Often groups have a geographical element, but they also have a class-based element. And in some places probably religious or ethnic elements.

The reason I think sound changes are picked up in person, besides half-remembered Labov, is that it seems clear that people don't pick up accents from the media. You can hear upper-class accents all day long— or any other type of cool accent— from the radio or TV or internet without adopting them.

But upper-class accents aren't cool. Quite the contrary. People aspire to the symbols of success within their group. Upper class accents aren't something for working-class people to aspire to.

Some of the changes you mention are so old that they're way past the initial acquisition stage I thought Torco was asking about; today there are millions of in-person exemplars people can pick them up from.

But this isn't really born out in reality. These changes spread fastest to those with least geographical mobility (because they're working-class, and the working classes are less geographically mobile). Taking, for instance, th-fronting - the gradual movement into estuary english, and up into the middle classes in the south-east, can certainly be explained through in-person contact, but its imitation in more and more working class areas since the 1980s surely cannot, given how little physical contact there is between the working classes in these areas (and given that it has not spread contiguously or via the most mobile people). In Glasgow, th-fronting, which has been occuring at a rapid and accelerating pace since around 1990, occurs only among working-class adolescents with low social and geographical mobility. These are precisely the people with the least face-to-face contact with th-fronters. There is, geographically, very little contact between the working-class populations of glasgow and london - they are, as it were, at opposite ends of the language. Yet th-fronting is rapidly 'tunelling' through this barrier, without affecting intervening dialects, and reversing the local trends toward th-debuccalisation. While it's impossible to say that this is happening ENTIRELY through TV (we can never rule out some, sporadic face-to-face contacts, glasgow isn't an island), it does make it hard to argue that people only adopt the speech of locals. [And glasgow is only the remotest example - it's happening in islands in most working-class urban groups]

[CaesarVincens]

But upper-class accents aren't cool. Quite the contrary. People aspire to the symbols of success within their group. Upper class accents aren't something for working-class people to aspire to.

But the more mobile middle class often do try to imitate perceived upper class accent features, to the point of hypercorrection in some cases. ('whom' for 'who', 'between you and I', etc.)

But this isn't really born out in reality. These changes spread fastest to those with least geographical mobility (because they're working-class, and the working classes are less geographically mobile). Taking, for instance, th-fronting - the gradual movement into estuary english, and up into the middle classes in the south-east, can certainly be explained through in-person contact, but its imitation in more and more working class areas since the 1980s surely cannot, given how little physical contact there is between the working classes in these areas (and given that it has not spread contiguously or via the most mobile people). In Glasgow, th-fronting, which has been occuring at a rapid and accelerating pace since around 1990, occurs only among working-class adolescents with low social and geographical mobility. These are precisely the people with the least face-to-face contact with th-fronters. There is, geographically, very little contact between the working-class populations of glasgow and london - they are, as it were, at opposite ends of the language. Yet th-fronting is rapidly 'tunelling' through this barrier, without affecting intervening dialects, and reversing the local trends toward th-debuccalisation. While it's impossible to say that this is happening ENTIRELY through TV (we can never rule out some, sporadic face-to-face contacts, glasgow isn't an island), it does make it hard to argue that people only adopt the speech of locals. [And glasgow is only the remotest example - it's happening in islands in most working-class urban groups]

This tunneling is because Glasgow and London are major cities. The same thing happens with say Los Angeles and New York City. NYC youth have certain features (quotatives mostly, I think) in common with LA youth, but the mid-westerners in between have yet to adopt. This is less noticeable over time as more youth adopt these features, but I remember that my class looked at studies showing this pretty strongly. (Canadian youth also picked up the quotatives.)

[Soap]

Yes. All sorts of different things. Poor reproduction of sounds (thus [θ] > [f] etc) that enough people do at a given time for it to become an accepted pronunciation, ease of pronunciation (thus lenition, anticipatory changes, assimilation and dissimilation). People change their speech to mimic a prestige dialect, too, obviously - there is a tale of some village up north where the local notables suffered from a speech impediment, leading to an extremely strange rhotic being used.

Wasnt that a myth? Not that we could ever conclusively know, but I thought it was just folklore. THere's a similar story regarding the origin of the Spanish /T/ for 'c' being due to the lisping habit of one of the kings ... but it doesnt make sense because they still contrasted it with /s/, it was just the "apical" sibilant that changed to /T/.

[Yng]

Yes. All sorts of different things. Poor reproduction of sounds (thus [θ] > [f] etc) that enough people do at a given time for it to become an accepted pronunciation, ease of pronunciation (thus lenition, anticipatory changes, assimilation and dissimilation). People change their speech to mimic a prestige dialect, too, obviously - there is a tale of some village up north where the local notables suffered from a speech impediment, leading to an extremely strange rhotic being used.

Wasnt that a myth? Not that we could ever conclusively know, but I thought it was just folklore. THere's a similar story regarding the origin of the Spanish /T/ for 'c' being due to the lisping habit of one of the kings ... but it doesnt make sense because they still contrasted it with /s/, it was just the "apical" sibilant that changed to /T/.

Well, I did say 'tale' - I don't know whether it's true or not. It's not completely out of the bounds of possibility, though, although I agree that the Spanish story is.

[chris_notts]

I read a book about this recently. It's called "Linguistic Diversity", by Daniel Nettle. See:

http://www.amazon.com/Linguistic-Divers ... 0198238576

It's quite cool that the author developed a simple computer simulation to demonstrate the point. He found that random mutation at the individual level alone isn't enough to create any kind of significant divergence between groups, if new members of the group are assumed to have "averaging" behaviour when learning and there is any significant level of migration. In order to get divergence when there is migration between groups in his simulation, you need to add high status individuals whose errors are more likely to be copied than the average member of the population.

Of course, his computer simulation is very simplistic, so it is more illustrative than a proof of everything. But his description of it is quite interesting, and I'm tempted to make my own simulation program just for fun.

He also makes an interesting proposal that typologically odd or unlikely features should be more likely to arise and be maintained in small groups, because the influence of any one high status individual is much greater in such societies. However, I'm not sure that there's enough evidence to test this, and the author himself only suggests this, presumably because he lacks enough evidence to support it.

[linguoboy]

He also makes an interesting proposal that typologically odd or unlikely features should be more likely to arise and be maintained in small groups, because the influence of any one high status individual is much greater in such societies. However, I'm not sure that there's enough evidence to test this, and the author himself only suggests this, presumably because he lacks enough evidence to support it.

I remember coming across that hypothesis before. One of the bits of evidence was the emergence of homoorganic stops before nasals in both Cornish and Surmeiran--an unusual change arising independently in two very small speech communities.

[dunomapuka]

This tunneling is because Glasgow and London are major cities. The same thing happens with say Los Angeles and New York City. NYC youth have certain features (quotatives mostly, I think) in common with LA youth, but the mid-westerners in between have yet to adopt. This is less noticeable over time as more youth adopt these features, but I remember that my class looked at studies showing this pretty strongly. (Canadian youth also picked up the quotatives.)

Interesting... are you referring to "like" as a quotative? Or "go?" (I grew up using both of these and later found it odd that this was considered a California feature.)