Why do different languages sound different?
November 8, 2010 9:50 AM Subscribe
Do we know anything about why different languages sound fundamentally different? What started the people who would eventually become French going "zhoh" in the first place?
My partner is studying Swedish at the moment and it got me wondering - we know that different languages use different types of sounds, and that each language has a sort of characteristic set of sounds (which can be known even by non-speakers of that language, as discussed on AskMe before). What I want to know is why?
Are there physical reasons that different groups of people started speaking in different ways, perhaps having to do with geographical features? Did speakers in cold climates prioritize sounds that could be made without opening the mouth very much? Or are there perhaps physical features of some groups of people that contributed to features of language (ie, is everyone physically capable of rolling their Rs or can some people genetically just not do it)?
Obviously there are a ton of socio-political reasons for the way languages sound today, having to do with who had the power and who conquered who. But I'm interested in how the distinct groups formed in the first place. Does linguistic science/history know anything about that?
My partner is studying Swedish at the moment and it got me wondering - we know that different languages use different types of sounds, and that each language has a sort of characteristic set of sounds (which can be known even by non-speakers of that language, as discussed on AskMe before). What I want to know is why?
Are there physical reasons that different groups of people started speaking in different ways, perhaps having to do with geographical features? Did speakers in cold climates prioritize sounds that could be made without opening the mouth very much? Or are there perhaps physical features of some groups of people that contributed to features of language (ie, is everyone physically capable of rolling their Rs or can some people genetically just not do it)?
Obviously there are a ton of socio-political reasons for the way languages sound today, having to do with who had the power and who conquered who. But I'm interested in how the distinct groups formed in the first place. Does linguistic science/history know anything about that?
Best answer: Or are there perhaps physical features of some groups of people that contributed to features of language (ie, is everyone physically capable of rolling their Rs or can some people genetically just not do it)?
IANALinguist, but from what I've read there's no genetic correlation with the ability for any specific language or pronounciation, e.g. a child of 100% Korean parents raised in Spain by Spaniards will speak perfect Spanish, rolled Rs and all.
posted by signal at 10:10 AM on November 8, 2010
IANALinguist, but from what I've read there's no genetic correlation with the ability for any specific language or pronounciation, e.g. a child of 100% Korean parents raised in Spain by Spaniards will speak perfect Spanish, rolled Rs and all.
posted by signal at 10:10 AM on November 8, 2010
Best answer: As far as I'm aware there aren't any physiological differences that contribute to language change (except for birth defects, everyone is physically capable of making the sounds in every language) and geography is not a factor.
One driver of language change seems to be a push and pull between easily being able to say what you want to say, and being easily able to make yourself understood. The natural tendency is to slur sounds together or make other sound changes that are easier to say. But as that happens, it gets harder for other people to understand you, and you end up compensating in other ways -- as a simplified example, in some southern US dialects "pen" and "pin" are homonyms, and people will disambiguate by saying "inkpen" for "pen."
There's also simply random change. If you take two populations and isolate them from each other, then they will gradually evolve to be different from each other even if the change isn't in any way adaptive or beneficial. This happens with genes, and I'm willing to bet is happens with languages as well.
posted by Jeanne at 10:10 AM on November 8, 2010 [1 favorite]
One driver of language change seems to be a push and pull between easily being able to say what you want to say, and being easily able to make yourself understood. The natural tendency is to slur sounds together or make other sound changes that are easier to say. But as that happens, it gets harder for other people to understand you, and you end up compensating in other ways -- as a simplified example, in some southern US dialects "pen" and "pin" are homonyms, and people will disambiguate by saying "inkpen" for "pen."
There's also simply random change. If you take two populations and isolate them from each other, then they will gradually evolve to be different from each other even if the change isn't in any way adaptive or beneficial. This happens with genes, and I'm willing to bet is happens with languages as well.
posted by Jeanne at 10:10 AM on November 8, 2010 [1 favorite]
Best answer: I recall a study (Google fu failing me) that hypothesized that some English accents simply require less muscle action and less caloric energy to speak with. The classic Brooklyn accent, for example, includes an "r" sound at the end of some words that end with vowels ("pizza" becomes "peetzer"). This can make it easier to transition to the beginning sounds of the next word in the sentence.
posted by Cool Papa Bell at 10:31 AM on November 8, 2010
posted by Cool Papa Bell at 10:31 AM on November 8, 2010
marginaliana - Are you asking about different regional pronunciations or entirely different sounds?
Any Romance or Germanic language will sound like language to me. On the other hand Vietnamese or other Austro-Asiatic languages are completely different sound structures.
Are you asking more about regional variations in a given language family or why are the different language families sound so friggin' different.
posted by 26.2 at 10:42 AM on November 8, 2010
Any Romance or Germanic language will sound like language to me. On the other hand Vietnamese or other Austro-Asiatic languages are completely different sound structures.
Are you asking more about regional variations in a given language family or why are the different language families sound so friggin' different.
posted by 26.2 at 10:42 AM on November 8, 2010
You could see this happening a bit in Maine where I grew up. It was the Down East accent and certain letters get dropped from words because they are utterly useless, in addition to "pick ups" from French. They tend to focus on 'r' up there but let's take 'c':
In my first two sentences it appears 5 times: could, accent, certain, and because. In each case it doesn't make a single unique sound. In 'could', 'accent', and 'because' it's a 'k' noise. In 'certain' it's a 's' noise. Not until you combine it with other letters does it become useful, such as with 'change'. S or K wouldn't cut it there, it's a unique noise.
In other cases words are repurposed as needed, such as with 'wicked'. 'Uniquely good' doesn't really reflect the sense of awe and wonder something is, so 'wicked' replaced 'uniquely' originally because it was an attention grabber: Hey, they used a word that means evil but applied it to something good! It must really have been special!
We started with something more like the San clicking language. It's amazing we have the variety we have, unless if you consider something like Spencer Wells' Journey Of Man research and it's byproduct, the National Genographic Project's Atlas of the Human Journey. Navigate via the yellow boxes under the times in the flash app.
Eventually someone is going to roll mythology into that atlas or something like it because there's only one way for half the societies on Earth to have flood mythology happen in such a fundamental part of their self views: it happened early in our history, before we split into separate groups. So you can compare myths between groups and look for where they *start* and thus actually establish a date and rough estimate of *where* they happened (between this spot and the previous one whose modern descendants don't have myth X).
We've been out there, doing the same crap Down Easters are doing now, for two hundred thousand years (5 zeros). That kind of force to a language's change rate applied constantly over that long of a period is going to yield massive changes. Massive.
posted by jwells at 10:53 AM on November 8, 2010
In my first two sentences it appears 5 times: could, accent, certain, and because. In each case it doesn't make a single unique sound. In 'could', 'accent', and 'because' it's a 'k' noise. In 'certain' it's a 's' noise. Not until you combine it with other letters does it become useful, such as with 'change'. S or K wouldn't cut it there, it's a unique noise.
In other cases words are repurposed as needed, such as with 'wicked'. 'Uniquely good' doesn't really reflect the sense of awe and wonder something is, so 'wicked' replaced 'uniquely' originally because it was an attention grabber: Hey, they used a word that means evil but applied it to something good! It must really have been special!
We started with something more like the San clicking language. It's amazing we have the variety we have, unless if you consider something like Spencer Wells' Journey Of Man research and it's byproduct, the National Genographic Project's Atlas of the Human Journey. Navigate via the yellow boxes under the times in the flash app.
Eventually someone is going to roll mythology into that atlas or something like it because there's only one way for half the societies on Earth to have flood mythology happen in such a fundamental part of their self views: it happened early in our history, before we split into separate groups. So you can compare myths between groups and look for where they *start* and thus actually establish a date and rough estimate of *where* they happened (between this spot and the previous one whose modern descendants don't have myth X).
We've been out there, doing the same crap Down Easters are doing now, for two hundred thousand years (5 zeros). That kind of force to a language's change rate applied constantly over that long of a period is going to yield massive changes. Massive.
posted by jwells at 10:53 AM on November 8, 2010
Response by poster: 26.2, I was thinking of differences between languages like French and Chinese. Why does Chinese sound the way it does and French sound the way it does? Maybe this is betraying a woeful misunderstanding of human history (and if so, tell me!) but the very first human beings who could speak "language" lived in the same area and sort of spoke the same "language," right? And then they split off into groups which lived in different places and the languages diverged, and many many many years later became French and Chinese. Right? So my question is what happened at the point where they stopped living in the same place that caused them to prioritize certain sounds which eventually led to there being French and Chinese? Does that make sense?
My thought is that surely we can use some information about the ways in which dialects are changing now to extrapolate as to why language changed way back then. But maybe that's not the case.
posted by marginaliana at 11:13 AM on November 8, 2010
My thought is that surely we can use some information about the ways in which dialects are changing now to extrapolate as to why language changed way back then. But maybe that's not the case.
posted by marginaliana at 11:13 AM on November 8, 2010
Best answer: It's random.
All languages are subject to sound change. Now there are general tendencies of sound change: we can see the same kinds of changes occuring all over the world. Palatalization of velar stops (k and g), for instance, is seen both in Mandarin (Peking > Beijing*) and Latin /kwinkwe/ > French cinq. There's something called a lenition hierarchy which includes quite a few common sound changes (e.g. unvoiced stops tend to become voiced, voiced ones fricativize; any consonant can become h; any consonant can disappear).
Many sound changes can be described as reducing effort. E.g. voicing an unvoiced stop (like k > g, e.g. Latin focus > Spanish fuego) reduces effort as you keep the vocal cords running instead of having to stop the voicing during the consonant. (There are some compensatory processes thad increase the effort though... e.g. vowel reduction can create new consonant clusters.)
But any particular sound change is unpredictable. The French go in one direction, the Italians in another. Accumulate enough changes over two thousand years and you've got separate languages.
posted by zompist at 11:48 AM on November 8, 2010
All languages are subject to sound change. Now there are general tendencies of sound change: we can see the same kinds of changes occuring all over the world. Palatalization of velar stops (k and g), for instance, is seen both in Mandarin (Peking > Beijing*) and Latin /kwinkwe/ > French cinq. There's something called a lenition hierarchy which includes quite a few common sound changes (e.g. unvoiced stops tend to become voiced, voiced ones fricativize; any consonant can become h; any consonant can disappear).
Many sound changes can be described as reducing effort. E.g. voicing an unvoiced stop (like k > g, e.g. Latin focus > Spanish fuego) reduces effort as you keep the vocal cords running instead of having to stop the voicing during the consonant. (There are some compensatory processes thad increase the effort though... e.g. vowel reduction can create new consonant clusters.)
But any particular sound change is unpredictable. The French go in one direction, the Italians in another. Accumulate enough changes over two thousand years and you've got separate languages.
posted by zompist at 11:48 AM on November 8, 2010
* The asterisk meant a footnote, which I forgot to supply. "Peking" wasn't meant as a direct representation of the Middle Chinese. But the name of the city used to have a k in Mandarin, and still does in Cantonese.
posted by zompist at 11:51 AM on November 8, 2010 [1 favorite]
posted by zompist at 11:51 AM on November 8, 2010 [1 favorite]
Best answer: Zompist is right (he usually is about this stuff), with one caveat. It's very common for languages to borrow phonological features from their neighbors. It's that sort of process that leads to geographic clusters of similar-sounding languages. For instance, tone appears to have "spread" from one language to another in East Asia — in fact, Chinese itself originally wasn't a tonal language — so that now it's an overwhelmingly common feature of languages in that part of the world.
If a bunch of French speakers settled in rural China and fell out of touch with mainland France, it would be a pretty decent bet that in a few hundred years their language would become tonal. Not a sure bet by any stretch of the imagination, but the odds would be in your favor.
If you set aside neighborhood effects like that, though, then yeah — this stuff is basically not predictable, and asking "What caused this sound change?" is very often missing the point. Entropy caused it, that's what.
posted by nebulawindphone at 11:56 AM on November 8, 2010 [1 favorite]
If a bunch of French speakers settled in rural China and fell out of touch with mainland France, it would be a pretty decent bet that in a few hundred years their language would become tonal. Not a sure bet by any stretch of the imagination, but the odds would be in your favor.
If you set aside neighborhood effects like that, though, then yeah — this stuff is basically not predictable, and asking "What caused this sound change?" is very often missing the point. Entropy caused it, that's what.
posted by nebulawindphone at 11:56 AM on November 8, 2010 [1 favorite]
And the randomness of language change also has a political dimension. For instance, there's the story explaining differences between regional varieties of Spanish that claims back in the day King Ferdinand (famous as the patron of Christopher Columbus) happened to speak with a lisp, but because he was the King and ruled Spain for over 35 years, the elite classes began to imitate his style of speech, eventually giving it the status of a prestige dialect among Old-World varieties of Spanish, even though other dialects emerged with the colonization of the New World and other locales.
posted by 5Q7 at 12:08 PM on November 8, 2010
posted by 5Q7 at 12:08 PM on November 8, 2010
Best answer: If there was something in our environment that strongly favoured a particular set of phonemes, then you might expect everyone in a particular environment to have similar-sounding languages. Morover, people in similar environments but in geographically very distant regions would've come up with very similar languages. This doesn't seem to be true.
However, if there isn't a strong environmental pressure for a language to use a particular set of phonemes, you'd expect each isolated group (whether isolated geographically, or by the cultures not particularly liking each other) to drift in a random direction, picking up a suite of phonemes that work well together (don't sound too alike, don't require wild tongue contortions to string together, etc) and slowly changing by a sort of "random walk" into something that sounds nothing like its neighbours. Which sounds a lot like the situation we have now. So while I'll obviously defer to an expert who contradicts me, I'd suggest that there doesn't need to be a reason why a language sounds a particular way. All languages drift and, when groups are separated, there's no reason why they would drift in the same direction.
I couldn't really guess at the rules for which sets of phonemes work well together. Avoiding tricky tongue gymnastics might be a factor (for example, part of the distinctive sound of spoken Welsh is the formally-defined mutations to certain word combinations that are basically there to make them flow together for smoother pronounciation), as might being easily distinguishable from each other by people already familiar with the rest of the language. I'd imagine that another big factor might be the borrowing of words or pronounciation styles from other languages as a consequence of mixing cultures (e.g. the incorporation of Latin words into early UK English thanks to the Romans, the borrowing of huge amounts of French vocabulary when our royal court was very closely tied to the French, and now a drift toward US English speech patterns and slang, presumably as a result of all the films, TV, music, etc. we get).
Finally -- and perhaps most interestingly -- this summer I spoke to a grad student working on the evolution of language. Part of the project she was presenting is explained online, with a fun little flash game.
As I understood it, they were interested in seeing what types of tone sequences (analogous to new words) people tended to be good at remembering, and building up a "volcabulary" of tone sequences that could be (a) easily remembered and (b)fairly easily distinguished from each other. The idea was that this could give an insight into the pressures acting on language formation. I forget the details of the rules that they came up with (other than that shorter sequences tend to be easier...), perhaps they're on the website. But most interesting to me was that these rules seemed to be the same for everyone they tested: people raised speaking wildly different languages tended to settle on vocabularies containing the same kinds of patterns, perhaps suggesting that there isn't some innate difference in the brain of a Chinese-speaker vs. an English-speaker that has caused the evolution of two very differently structured languages (or conversely, suggesting that growing up speaking differently structured languages has not conditioned them to be good at handling different patterns).
posted by metaBugs at 12:18 PM on November 8, 2010
However, if there isn't a strong environmental pressure for a language to use a particular set of phonemes, you'd expect each isolated group (whether isolated geographically, or by the cultures not particularly liking each other) to drift in a random direction, picking up a suite of phonemes that work well together (don't sound too alike, don't require wild tongue contortions to string together, etc) and slowly changing by a sort of "random walk" into something that sounds nothing like its neighbours. Which sounds a lot like the situation we have now. So while I'll obviously defer to an expert who contradicts me, I'd suggest that there doesn't need to be a reason why a language sounds a particular way. All languages drift and, when groups are separated, there's no reason why they would drift in the same direction.
I couldn't really guess at the rules for which sets of phonemes work well together. Avoiding tricky tongue gymnastics might be a factor (for example, part of the distinctive sound of spoken Welsh is the formally-defined mutations to certain word combinations that are basically there to make them flow together for smoother pronounciation), as might being easily distinguishable from each other by people already familiar with the rest of the language. I'd imagine that another big factor might be the borrowing of words or pronounciation styles from other languages as a consequence of mixing cultures (e.g. the incorporation of Latin words into early UK English thanks to the Romans, the borrowing of huge amounts of French vocabulary when our royal court was very closely tied to the French, and now a drift toward US English speech patterns and slang, presumably as a result of all the films, TV, music, etc. we get).
Finally -- and perhaps most interestingly -- this summer I spoke to a grad student working on the evolution of language. Part of the project she was presenting is explained online, with a fun little flash game.
As I understood it, they were interested in seeing what types of tone sequences (analogous to new words) people tended to be good at remembering, and building up a "volcabulary" of tone sequences that could be (a) easily remembered and (b)fairly easily distinguished from each other. The idea was that this could give an insight into the pressures acting on language formation. I forget the details of the rules that they came up with (other than that shorter sequences tend to be easier...), perhaps they're on the website. But most interesting to me was that these rules seemed to be the same for everyone they tested: people raised speaking wildly different languages tended to settle on vocabularies containing the same kinds of patterns, perhaps suggesting that there isn't some innate difference in the brain of a Chinese-speaker vs. an English-speaker that has caused the evolution of two very differently structured languages (or conversely, suggesting that growing up speaking differently structured languages has not conditioned them to be good at handling different patterns).
posted by metaBugs at 12:18 PM on November 8, 2010
Best answer: If you're familiar with complexity (aka chaos) theory, you must have heard the theory that complex systems like weather can give you radically different outputs from even tiny differences in the input or prevailing conditions.
Language works basically the same way. It's not completely random, obviously, or everyone would speak a different language and no-one would be able to talk to each other. But it is ever-changing, and slightly different conditions in valley A vs valley B can produce slightly different results over one generation that add up to hugely different results over ten or a hundred generations.
On the other hand, there are factors that can make languages more similar to each other, too, even if they aren't necessarily genetically related. Examples would include the phenomenon of the Sprachbund and, more generally, common sound changes that tend to occur very frequently in human language (these tend to be changes that simplify certain sound sequences or otherwise make the language easier/quicker to say... although in most cases the results are eventually combined into new complex/long sequences, which then get simplified, and so on until the language goes extinct).
Searches for external, non-social factors driving language change, e.g. this country has a lot of sweeping plains so you have to be able to shout a long way and so all the vowels are like this, have not produced much in the way of provable results IIRC.
posted by No-sword at 1:04 PM on November 8, 2010
Language works basically the same way. It's not completely random, obviously, or everyone would speak a different language and no-one would be able to talk to each other. But it is ever-changing, and slightly different conditions in valley A vs valley B can produce slightly different results over one generation that add up to hugely different results over ten or a hundred generations.
On the other hand, there are factors that can make languages more similar to each other, too, even if they aren't necessarily genetically related. Examples would include the phenomenon of the Sprachbund and, more generally, common sound changes that tend to occur very frequently in human language (these tend to be changes that simplify certain sound sequences or otherwise make the language easier/quicker to say... although in most cases the results are eventually combined into new complex/long sequences, which then get simplified, and so on until the language goes extinct).
Searches for external, non-social factors driving language change, e.g. this country has a lot of sweeping plains so you have to be able to shout a long way and so all the vowels are like this, have not produced much in the way of provable results IIRC.
posted by No-sword at 1:04 PM on November 8, 2010
We started with something more like the San clicking language
This is semi-OT, but is there research that shows this is true? Do we know something about the San language that indicates that it's older than other languages? I know that anthropologists are fairly sure that the San ethnic phenotypes are "older" than the more predominant Bantu ethnic phenotypes of the surrounding areas (and have other ways of knowing that Bantu groups spread out relatively recently to dominate sub-saharan Africa). But I'm not sure that translates to language at all, or if modern San people even speak the same language that their prehistoric counterparts spoke. Or how we would even know that. But I'm suddenly really, really interested in the answer...
posted by Sara C. at 1:06 PM on November 8, 2010 [2 favorites]
This is semi-OT, but is there research that shows this is true? Do we know something about the San language that indicates that it's older than other languages? I know that anthropologists are fairly sure that the San ethnic phenotypes are "older" than the more predominant Bantu ethnic phenotypes of the surrounding areas (and have other ways of knowing that Bantu groups spread out relatively recently to dominate sub-saharan Africa). But I'm not sure that translates to language at all, or if modern San people even speak the same language that their prehistoric counterparts spoke. Or how we would even know that. But I'm suddenly really, really interested in the answer...
posted by Sara C. at 1:06 PM on November 8, 2010 [2 favorites]
Best answer: If you accept the premise that inbreeding is generally selected against-- and the plethora of mechanisms which seem to encourage outbreeding among animals, plants and single celled organisms appear to argue for that-- then you could guess heritable traits which cause languages to diverge in geographically separated populations would be selected for simply because such divergence would tend to be an indicator by proxy of an underlying genetic divergence (though the operation of that selection would seem to have the effect of making it a poorer proxy, interestingly).
If we have in the main been patrilocal (females migrate away from home to reproduce), and humans do indeed conform to the general mammalian rule of displaying a strong component of female mate choice, perhaps that could explain my experience that women tend to find a foreign or regional accent very attractive-- much more than men do.
posted by jamjam at 2:08 PM on November 8, 2010
If we have in the main been patrilocal (females migrate away from home to reproduce), and humans do indeed conform to the general mammalian rule of displaying a strong component of female mate choice, perhaps that could explain my experience that women tend to find a foreign or regional accent very attractive-- much more than men do.
posted by jamjam at 2:08 PM on November 8, 2010
One thing I've noticed is that there's an apparent desire to sound like popular people. Kids trying to be macho sound like Schwarzenegger (they'll ham it up for the catch phrases but what I'm more talking is a residual pattern where Ts and Ds sound the same), and I've heard Americans who who are fans of Richard Dawkins suddenly acquire a slightly posh accent.
So speech patterns might be partly random, but popular figures seem to also have an effect and 5Q7's story rings true.
posted by holloway at 2:24 PM on November 8, 2010
So speech patterns might be partly random, but popular figures seem to also have an effect and 5Q7's story rings true.
posted by holloway at 2:24 PM on November 8, 2010
Best answer: So, to answer your above-the-fold example question... French people say /j/ and Italian people say /gi/ because the "original" sound (in the most recent common ancestor to both languages) evolved that way in those two languages. How that evolution went down -- whether it was a straight "easier-to-say" change from an original ancestor like /g/ or whether contact with other language groups influenced the change in one (or both) cases, I don't know, but you can bet there was some precursor (even if it can't be traced any more), and interactions within the languages may also have helped (maybe changes in French vowels made the /j/ easier to combine with them. Some people believe that environmental factors can also be involved (e.g. sweeping plains call for lots of shouting, so your vowels change like so) but IIRC there is no good evidence for this.
It's harder to compare French vs Chinese because they aren't related in any way yet detected, but if hypothetically they both sprung from the same ancient ancestor language, the same is true: gradual change adding up to huge differences.
So the question is, if "everything at time N" comes from "everything at time N-1" + "catalyst," where did everything at time N-1 come from? N-2 + catalysts... and so on for as far back as scholarship can go. As far as we can tell, we aren't yet anywhere near the "dawn of language," and barring amazing new techniques or ideas may never get there.
posted by No-sword at 2:33 PM on November 8, 2010
It's harder to compare French vs Chinese because they aren't related in any way yet detected, but if hypothetically they both sprung from the same ancient ancestor language, the same is true: gradual change adding up to huge differences.
So the question is, if "everything at time N" comes from "everything at time N-1" + "catalyst," where did everything at time N-1 come from? N-2 + catalysts... and so on for as far back as scholarship can go. As far as we can tell, we aren't yet anywhere near the "dawn of language," and barring amazing new techniques or ideas may never get there.
posted by No-sword at 2:33 PM on November 8, 2010
Best answer: So.... let's see if I have this straight... If we could do an experiment and start all over again, then the evolution of languages would likely be quite different? The occupants of what we call France would speak a language that sounds nothing like what we know as French? It's all just a big random evolution of sounds traded among neighbors?
posted by exphysicist345 at 7:06 PM on November 8, 2010
posted by exphysicist345 at 7:06 PM on November 8, 2010
Best answer: Assuming that events we'd agree are "random accidents of history" really are, and there's not some larger force driving world events (e.g. it's not some sort of law of physics that the Celts inhabited Gaul, followed by Roman and subsequent Frankish invasions), then yes, probably.
posted by Sara C. at 7:15 PM on November 8, 2010
posted by Sara C. at 7:15 PM on November 8, 2010
And the randomness of language change also has a political dimension. For instance, there's the story explaining differences between regional varieties of Spanish that claims back in the day King Ferdinand (famous as the patron of Christopher Columbus) happened to speak with a lisp, but because he was the King and ruled Spain for over 35 years, the elite classes began to imitate his style of speech, eventually giving it the status of a prestige dialect among Old-World varieties of Spanish, even though other dialects emerged with the colonization of the New World and other locales.
And yet Castilian Spanish has both an "s" sound and a "th" (if looking at the language from English), so this doesn't make sense. Actually it's an urban myth. And more here. The money quote that explains all of this:
"Firstly, the ceceo is not a lisp. A lisp is the mispronunciation of the sibilant s sound. In Castilian Spanish, the sibilant s sound exists and is represented by the letter s. The ceceo comes in to represent the sounds made by the letters z and cfollowed by i or e.
"In medieval Castilian there were two sounds that eventually evolved into the ceceo, the ç (the cedilla) as in plaça and the z as in dezir. The cedilla made a /ts/ sound and the z a /dz/ sound. This gives more insight into why those similar sounds may have evolved into the ceceo."
posted by ob at 8:10 PM on November 8, 2010 [1 favorite]
And yet Castilian Spanish has both an "s" sound and a "th" (if looking at the language from English), so this doesn't make sense. Actually it's an urban myth. And more here. The money quote that explains all of this:
"Firstly, the ceceo is not a lisp. A lisp is the mispronunciation of the sibilant s sound. In Castilian Spanish, the sibilant s sound exists and is represented by the letter s. The ceceo comes in to represent the sounds made by the letters z and cfollowed by i or e.
"In medieval Castilian there were two sounds that eventually evolved into the ceceo, the ç (the cedilla) as in plaça and the z as in dezir. The cedilla made a /ts/ sound and the z a /dz/ sound. This gives more insight into why those similar sounds may have evolved into the ceceo."
posted by ob at 8:10 PM on November 8, 2010 [1 favorite]
Best answer: Great question! Short answer: No, there are no identifiable physical reasons why certain languages sound the way they do. Detailed answer hits on a bunch of different subfields of linguistics.
First of all: At a first approximation, when a sound changes in a language, it changes systematically across that language (subfield: historical phonology). We tolerate small numbers of exceptions—either the few words in which the sound change starts out, or a few words that resist it—but we get uncomfortable if the language is arbitrarily divided down the middle. We'd rather handle comparable words in analogous ways (subfields: psycholinguistics, analogy). This is why a language retains a particular recognizable sound: It uses the same phonemes over and over, instead of letting every word mutate independently into whatever sound it happens to acquire.
(At a second approximation, sound change isn't strictly uniform because geography gets involved. One sound change may spread outward from one place, and another from another place, so that they reach different territories at different times (or maybe don't reach certain territories at all). If one of those changes happens to either create or destroy the sounds that the other change acts upon, then the order in which they arrive is important, and those two sound changes will accordingly create a patchwork of territories that speak a little differently from one another. Subfield: dialectology.)
One of the first regular sound changes discovered was Grimm's Law (as in the Brothers Grimm, yes), which describes how voiced stop consonants (such as b) turned into voiceless fricative consonants (like f) in the Germanic languages. No sooner had the law been discovered than someone suggested that the sounds had changed because the early Germanic settlers were routinely out of breath from running up and down the Alps! In comparison to other mountainous areas, the theory has not stood up, nor have subsequent efforts succeeded in predicting language features from either geographic or genetic factors. Closest you can get is to predict language features from features of languages nearby, where there is a lot of interaction—turns out even unrelated languages sometimes acquire similar features, if their speaking populations interact or overlap enough. (Subfield: areal linguistics. Note that the terrain can have an effect on the size of a linguistic area, or even on whether one gets established at all; navigable coasts are conducive to interaction, impassable mountains to isolation. But this still doesn't predict what the areal features will be.)
Jeanne and zompist pretty much already covered the tension between ease of pronunciation and differentiation of different words, so I'll just add that the subfield is called Optimality Theory. In Optimality Theory, rules that tend toward easy pronunciation are called markedness constraints, rules that tend to differentiate one pronunciation from another are called faithfulness constraints. The remainder of the theory proposes how the two kinds of constraints combine to shape a 'space of possible grammars' large enough to cover the variety in human language, but small enough that a child can search it and learn which of those grammars to use.
But, all that being said, exactly which sound changes will be picked up, in the course of this struggle to speak easily and still be intelligible, is not strictly a linguistic question. It's an anthropological and sociological one: Speech communities tend to follow the lead of whoever has prestige. That doesn't necessarily mean money. In a community where social ties are important, it may be the best-connected person. In a community that defines itself in opposition to a moneyed class, it may be the person who speaks least posh. You influence the speech standard insofar as you embody the values. (Subfield: sociolinguistics. There are other factors besides prestige, such as the shopkeeper effect: You tend to speak like your prospective customers. I recall reading that for a while, Protestant shopkeepers with Catholic customers were pretty much the only route for Catholic-to-Protestant linguistic diffusion in Belfast. But prestige is still the big one.)
posted by eritain at 11:47 PM on November 8, 2010
First of all: At a first approximation, when a sound changes in a language, it changes systematically across that language (subfield: historical phonology). We tolerate small numbers of exceptions—either the few words in which the sound change starts out, or a few words that resist it—but we get uncomfortable if the language is arbitrarily divided down the middle. We'd rather handle comparable words in analogous ways (subfields: psycholinguistics, analogy). This is why a language retains a particular recognizable sound: It uses the same phonemes over and over, instead of letting every word mutate independently into whatever sound it happens to acquire.
(At a second approximation, sound change isn't strictly uniform because geography gets involved. One sound change may spread outward from one place, and another from another place, so that they reach different territories at different times (or maybe don't reach certain territories at all). If one of those changes happens to either create or destroy the sounds that the other change acts upon, then the order in which they arrive is important, and those two sound changes will accordingly create a patchwork of territories that speak a little differently from one another. Subfield: dialectology.)
One of the first regular sound changes discovered was Grimm's Law (as in the Brothers Grimm, yes), which describes how voiced stop consonants (such as b) turned into voiceless fricative consonants (like f) in the Germanic languages. No sooner had the law been discovered than someone suggested that the sounds had changed because the early Germanic settlers were routinely out of breath from running up and down the Alps! In comparison to other mountainous areas, the theory has not stood up, nor have subsequent efforts succeeded in predicting language features from either geographic or genetic factors. Closest you can get is to predict language features from features of languages nearby, where there is a lot of interaction—turns out even unrelated languages sometimes acquire similar features, if their speaking populations interact or overlap enough. (Subfield: areal linguistics. Note that the terrain can have an effect on the size of a linguistic area, or even on whether one gets established at all; navigable coasts are conducive to interaction, impassable mountains to isolation. But this still doesn't predict what the areal features will be.)
Jeanne and zompist pretty much already covered the tension between ease of pronunciation and differentiation of different words, so I'll just add that the subfield is called Optimality Theory. In Optimality Theory, rules that tend toward easy pronunciation are called markedness constraints, rules that tend to differentiate one pronunciation from another are called faithfulness constraints. The remainder of the theory proposes how the two kinds of constraints combine to shape a 'space of possible grammars' large enough to cover the variety in human language, but small enough that a child can search it and learn which of those grammars to use.
But, all that being said, exactly which sound changes will be picked up, in the course of this struggle to speak easily and still be intelligible, is not strictly a linguistic question. It's an anthropological and sociological one: Speech communities tend to follow the lead of whoever has prestige. That doesn't necessarily mean money. In a community where social ties are important, it may be the best-connected person. In a community that defines itself in opposition to a moneyed class, it may be the person who speaks least posh. You influence the speech standard insofar as you embody the values. (Subfield: sociolinguistics. There are other factors besides prestige, such as the shopkeeper effect: You tend to speak like your prospective customers. I recall reading that for a while, Protestant shopkeepers with Catholic customers were pretty much the only route for Catholic-to-Protestant linguistic diffusion in Belfast. But prestige is still the big one.)
posted by eritain at 11:47 PM on November 8, 2010
Best answer: Maybe this is betraying a woeful misunderstanding of human history (and if so, tell me!) but the very first human beings who could speak "language" lived in the same area and sort of spoke the same "language," right? And then they split off into groups which lived in different places and the languages diverged, and many many many years later became French and Chinese. Right?
You (and most of the people who replied with explanations of language change) are presupposing here that language began in a certain location at a certain time amongst all humans and that this is the principle reason for the difference between languages. This is not necessarily the case! Most linguists reject the idea of an "Adamic" language (i.e. a single language which is the progenitor of all languages). Rather it seems that language arose (near-) spontaneously amongst several groups of people (and of course immediately began changing). This may sound strange, but evolutionarily speaking, this sort of thing is common.
Next, languages are really not so different as you may think. There are a closed group of sounds that make up all of human language, so although some have tone and other do not and some have clicks and others do not, there are really only so many sounds in the first place.
Moreover, there are correlations between sounds and groups of sounds, so languages with sound X usually don't have quality/sound/process Y (I guess languages with prenasalized stops [mp, nd, Ng]do not usually have contrastive voiced stops [b, d, g]).
I guess my main point is that they are not as different as they seem.
posted by mateuslee at 5:08 AM on November 9, 2010
You (and most of the people who replied with explanations of language change) are presupposing here that language began in a certain location at a certain time amongst all humans and that this is the principle reason for the difference between languages. This is not necessarily the case! Most linguists reject the idea of an "Adamic" language (i.e. a single language which is the progenitor of all languages). Rather it seems that language arose (near-) spontaneously amongst several groups of people (and of course immediately began changing). This may sound strange, but evolutionarily speaking, this sort of thing is common.
Next, languages are really not so different as you may think. There are a closed group of sounds that make up all of human language, so although some have tone and other do not and some have clicks and others do not, there are really only so many sounds in the first place.
Moreover, there are correlations between sounds and groups of sounds, so languages with sound X usually don't have quality/sound/process Y (I guess languages with prenasalized stops [mp, nd, Ng]do not usually have contrastive voiced stops [b, d, g]).
I guess my main point is that they are not as different as they seem.
posted by mateuslee at 5:08 AM on November 9, 2010
One thing that could possibly be a culture-based factor in making languages sound a particular way might be 'conservatism' in a linguistic sense, i.e., the value the culture places on preserving a fixed pronunciation, vocabulary, canon of classic poems, etc. I'm thinking of the way the Arab world has perpetuated knowledge of classical Arabic and preserved it's letter-sounds, even as pronunciation shifted around in the colloquial dialects. So young people in every generation still absorb the ability to pronounce far-out sounds, e.g., ق, that, on the basis of the natural evolution of language, would have disappeared long ago. Maybe, that results in Arabic in general maintaining a certain sound? I know that still doesn't explain what the profile of that sound might be and why, so maybe this doesn't have a point...
posted by Paquda at 9:18 AM on November 9, 2010
posted by Paquda at 9:18 AM on November 9, 2010
I think that's more like the difference between how we have Latin, a classical "dead" language nobody really speaks anymore (and which may or may not have been spoken as written even during classical times), and then we also have Italian, which is what happened to Latin over a few millenia of language change. An Italian student in a seminary being trained for the priesthood in 1950 would have had to learn the classical pronunciations in order to say mass in Latin, in the same way a Spanish or French student would. None of which has any bearing on how language changed in those three regions between Roman times and 1950.
The difference in this situation between Latin and its offshoot languages, and Arabic and its offshoot languages, as far as I can tell, is that we call Italian, French, Spanish, Provencal, Portuguese, etc. "languages", while we call all the offshoots of Arabic "dialects". Which is most likely a total accident - there is no hard and fast dividing line between what should be considered a separate language and what should be considered a dialect.
There may be some other accidents of culture involved, for instance maybe there's been a stronger influence on lay people learning classical Arabic where that is no longer true for Latin. Or maybe it has something to do certain aspects of Islam which have not been true of the Catholic church over the centuries (they have very different historical relationships to literacy, for example). But it's almost certainly got nothing to do with innate qualities of European vs. Middle Eastern people.
posted by Sara C. at 9:39 AM on November 9, 2010
The difference in this situation between Latin and its offshoot languages, and Arabic and its offshoot languages, as far as I can tell, is that we call Italian, French, Spanish, Provencal, Portuguese, etc. "languages", while we call all the offshoots of Arabic "dialects". Which is most likely a total accident - there is no hard and fast dividing line between what should be considered a separate language and what should be considered a dialect.
There may be some other accidents of culture involved, for instance maybe there's been a stronger influence on lay people learning classical Arabic where that is no longer true for Latin. Or maybe it has something to do certain aspects of Islam which have not been true of the Catholic church over the centuries (they have very different historical relationships to literacy, for example). But it's almost certainly got nothing to do with innate qualities of European vs. Middle Eastern people.
posted by Sara C. at 9:39 AM on November 9, 2010
Hi, Sara C. What I meant was, the preservation of letter sounds that would otherwise be lost...in Latin, whatever the original sounds were, people in each country just ended up pronouncing the letters as they would in their native language: Italians pronouncing 'c' sometimes as 'ch', Englishmen pronouncing it sometimes as 's'--but there's not 'stretching' of one's repertoire of sounds involved. (I know that over the last century, there's been a movement to reform the pronunciation of latin taught in schools, on the basis of what's been pieced-together about how ancient romans pronounced it, but I don't think that's a phenomenon that would have any wider effect.) So, whereas almost all languages are subject to evolutionary pressures that morph high-effort sounds into easier ones, in Arabic there might be this counter-force that keeps them around.
posted by Paquda at 6:47 AM on November 10, 2010
posted by Paquda at 6:47 AM on November 10, 2010
The reason Arabic keeps them around is probably that there is an unbroken line from classical times right up to the present day - scholars can know how the words used to be pronounced.
With Latin, we don't really know how it was pronounced, because there were centuries of cultural collapse between classical times and the time when anybody would have even cared. (probably the Renaissance?) Now we only have the words as written.
And, again, it also probably has a lot to do with differences between Islamic and Christian religious practice. For centuries the Catholic Church emphasized the idea that the average lay person should have little to no actual contact with scripture as written, and everything should go through a few priestly elites (even in terms of actually understanding what was being said in the Mass). I don't think Islam has that tradition.
It probably doesn't have much if anything to do with innate qualities Arabic as a language. ALL languages are subject to the same "evolutionary forces". The differences are external (culture, religion, politics, etc).
posted by Sara C. at 10:41 AM on November 10, 2010
With Latin, we don't really know how it was pronounced, because there were centuries of cultural collapse between classical times and the time when anybody would have even cared. (probably the Renaissance?) Now we only have the words as written.
And, again, it also probably has a lot to do with differences between Islamic and Christian religious practice. For centuries the Catholic Church emphasized the idea that the average lay person should have little to no actual contact with scripture as written, and everything should go through a few priestly elites (even in terms of actually understanding what was being said in the Mass). I don't think Islam has that tradition.
It probably doesn't have much if anything to do with innate qualities Arabic as a language. ALL languages are subject to the same "evolutionary forces". The differences are external (culture, religion, politics, etc).
posted by Sara C. at 10:41 AM on November 10, 2010
"The difference in this situation between Latin and its offshoot languages, and Arabic and its offshoot languages, as far as I can tell, is that we call Italian, French, Spanish, Provencal, Portuguese, etc. "languages", while we call all the offshoots of Arabic "dialects". Which is most likely a total accident - there is no hard and fast dividing line between what should be considered a separate language and what should be considered a dialect. "
While it's a fair point to note that Arabic dialects might better be termed languages, if only because not all "dialects" of Arabic are mutually intelligible, there are a couple of things that should be noted when you want to compare Arabic with Latin. The first is that Latin's expansion was much wider, and much longer ago, than Arabic's. Spanish is just as influenced by Basque, Catalan, Arabic, and any number of other local languages as French is influenced by the Gauls and Celts, etc. Arabic spread much more recently, in a much more interconnected world, and spread a lot more totally (in that local languages tended to be more dominated by Arabic). So the differences in Arabic are more due to divergent evolution, whereas the differences between Romance languages are in large part because they had much more varied inputs in creating them.
Or at least, that's what I remember from my linguistics class years ago.
posted by klangklangston at 10:59 AM on November 22, 2010
While it's a fair point to note that Arabic dialects might better be termed languages, if only because not all "dialects" of Arabic are mutually intelligible, there are a couple of things that should be noted when you want to compare Arabic with Latin. The first is that Latin's expansion was much wider, and much longer ago, than Arabic's. Spanish is just as influenced by Basque, Catalan, Arabic, and any number of other local languages as French is influenced by the Gauls and Celts, etc. Arabic spread much more recently, in a much more interconnected world, and spread a lot more totally (in that local languages tended to be more dominated by Arabic). So the differences in Arabic are more due to divergent evolution, whereas the differences between Romance languages are in large part because they had much more varied inputs in creating them.
Or at least, that's what I remember from my linguistics class years ago.
posted by klangklangston at 10:59 AM on November 22, 2010
Oh, and a neat current version is Austrian's divergence from German. Austrians speak German as a sort of official second language, but if you don't have any use for it, you tend to slip back into the vocabulary of Austrian, which is much more influenced by Slavic languages and the Turks. Likewise, because so many of the villages in Austria are isolated, a lot have developed local dialects that really only apply to that valley. I was at a party once where a bunch of Austrians that didn't know each other showed up, and one went through the party asking the others to say a couple phrases, after which he would guess what village they grew up in (he had some sort of student job that took him all over rural Austria, so he got good at picking out where people were from).
posted by klangklangston at 11:02 AM on November 22, 2010
posted by klangklangston at 11:02 AM on November 22, 2010
There are parts of the Arab world where Arabic is just as strongly influenced by substrate languages: Berber in Morocco, for instance. But yeah, I gather it's not so much like that places like Egypt or most of the Arabian peninsula.
posted by nebulawindphone at 11:03 AM on November 22, 2010
posted by nebulawindphone at 11:03 AM on November 22, 2010
This thread is closed to new comments.
posted by Ys at 10:01 AM on November 8, 2010