Re: Obsessed with Mouth Noises

David's message (which makes a number of good points) came through
without line breaks, making it a little hard to read in my client;
I've reformatted it and reproduce it below for others who might have
the same problem.  (I've also added right-chevron-indentation to
Gary's passages, in addition to David's pseudo-guillemet-quotes - I
hope he doesn't mind.)

David Peterson wrote at 2004-04-10 03:53:47 (-0400)
 > There are a lot of assumptions you've made that you might not
 > realize you've made here.  So let's go through all this:
 >
 > Gary wrote:
 >
 >> <<Granted, pronunciation can be interesting, but I'llnever
 >> understand the seeming obsession with trivialnuances that seems to
 >> grip some linguists.Pronunciation is nearly irrelevant when it
 >> comes tofulfilling the primary purpose of
 >> language;communication.>>
 >
 > Assumption Number 1: If language is only a tool for communication,
 > then all linguists should be concerned with, with respect to
 > phonetics, is whether or not person A can communicate with person
 > B.  Or possibly: All phoneticians are interested with are dialectal
 > differences which have no bearing on communication.  Both of these
 > are, of course, completely untrue.  Leaving production problems
 > aside (i.e., people that have to work with a speech pathologist in
 > order to be able to produce recognizable sounds), one of the main
 > things phoneticians do with their studies is make
 > predictions.  These predictions can be predictions about why a
 > certain sound change occurred, which ones are more likely to occur,
 > what distinctions are unstable, etc.  Why do all this?  Well, for
 > example, at one point in time, German speakers and English speakers
 > spoke the same language.  Now they don't.  Part of this has to do
 > with syntax, morphology, etc., but a lot also has to do with sound
 > changes.  The way we learn about sound changes is studying dialect
 > variation that's occurring right now.  After all, different
 > dialects can eventually grow into different languages.  Now,
 > whether or not this whole realm of study is frivolous is not
 > something that can really be argued, because that could lead to a
 > whole discussion of, "Well, if you think *that's* frivolous, then
 > what about this?"
 >
 >> <<Two people walk into a restaurant.  One orders "frahdchikin
 >> pleez" and the other orders "vroit jigun pliss"and the both get
 >> the same meal delivered to theirtable.>>
 >
 > Let's assume your right about this.  We can extend this to syntax
 > and general grammar.  A propose that a conlang should have no fixed
 > grammar, but merely a wide range of acceptable variants.  After
 > all, you can say "He went to the store", or "He go store", and
 > they're both perfectly understandable to a native English
 > speaker.  So why impose arbitrary rules such as tense,
 > prepositions, articles, etc.?  If you do that, it becomes too hard
 > to speak "correctly".
 >
 > One thing that the example I gave above and your example have in
 > common is that they're hard to screw up.  If you have an idea of
 > how the language is supposed to sound, you'll get it somewhat right
 > (i.e., you won't say "The to went store he", or [vli:t s@idZo~
 > BjejS] ="fried chicken please").  What about little differences,
 > though?  So, the stereotypical Hindi speaker who speaks English
 > overgeneralizes the progressive, so they say things like, "I am
 > thinking that's not such a good idea."  So, one thing such a
 > speaker might say is, "I am seeing him."  Does that mean they're
 > dating him, or that they can see him?  Presumably they also know
 > the expression "to be seeing someone" (after all, we're dealing
 > with fluent speakers who just have a different way of speaking, not
 > language-learners).  So which is it?  Same thing with
 > pronunciation.  I can't think of any examples off hand, but one
 > problem that my grandmother has is she can't distinguish [I] and
 > [i].  So both "seat" and "sit!" are pronounced [sit].  I can tell
 > you from experience that there are times that I've genuinely been
 > confused about what she was talking about, and the only reason was
 > pronunciation.  Now, the difference between [i] and [I] is, quite
 > frankly, really just a nuance of the mouth.  The sounds are *so*
 > close that they're indistinguishable in most languages.  They are
 > distinguished in English, though, so this little nuance of the
 > mouth isn't simply a nuance.
 >
 > Also, why is it that you've only focused on seemingly non-native
 > speakers?  Native speakers have varying pronunciations, and
 > studying the variance can tell you a lot about the history of the
 > language, and where the language is going.  I attended an
 > interesting talk last year, in fact, about how there's a kind of
 > vowel harmony developing in Scots English.  Did I have trouble
 > understanding them?  No.  Is it important to communication?
 >  No.  Is it, therefore, uninteresting, and a waste of time to
 > study?  If you say "yes", then I'd really like to hear why.  It's a
 > linguistic phenomenon, and linguists study linguistic
 > phenomenon.  From a conlanging point of view, it's something that
 > happens in real language, so if you want to make a realistic
 > conlang, you should keep it in mind.
 >
 > But I think your main point is that there's no point in specifying,
 > in a conlang, that X sound needs to pronounce exactly thus, and Y
 > vowel needs to be pronounced in just this way.  Fair enough.  But
 > why do they do it real grammars of real languages?  Should it be
 > considered unimportant?  After all, one of the things that every
 > good grammar will say right off the bat is what dialect they
 > describe, and all the particulars in the grammar will be understood
 > to be specific only to that dialect.  Would you say that they
 > should do something else?  Maybe say, "In X language, you can kind
 > of pronounce this vowel like this, this, this, this, this, this,
 > this, this, this, and this, depending on where you are."  Maybe if
 > the point was to help you be understood in that language, that
 > would be fine, but if the point is to describe the language, then
 > it would be totally irresponsible.
 >
 >> <<It seems to me that it is sufficient for any given language to
 >> define the range of acceptable phonetic values for a given meaning
 >> and leave it at that.>>
 >
 > Back to conlanging, let's consider this point.  If you intend your
 > conlang to be spoken by actual people, then, yes, maybe you might
 > do this.  In fact, this is exactly the kind of thing they talk
 > about on Auxlang.  When I learned Esperanto, for example, we were
 > told how all the sounds should be pronounced, and encouraged to not
 > pronounce them in an English-like manner.  In fact, for every
 > language I've ever learned this has been the case.  The reason is
 > that the not-100%-Esperanto-way of pronouncing, for example, /t/ in
 > the U.S. (alveolar, aspirated), will be different from the
 > not-100%-Esperanto-way of pronouncing /t/ in India (I'd guess that
 > the latter'd be dental, unaspirated?).  Then when you consider
 > complicated letters like /l/ and /r/, pretty soon you could imagine
 > people not being able to understand one another.  Unless you're
 > creating an IAL, though, you're probably imagining a conlanging
 > audience, and so you can get away with giving a range of
 > pronunciations for a given sound.
 >
 > So that's for a language one intends to be spoken.  None of my
 > conlangs (with the exception of maybe Kele), however, are supposed
 > to be spoken languages.  I mean, they could be, but I'm not going
 > out trying to recruit speakers.  So I really don't care how people
 > pronounce, or what pronunciations would be acceptable and what
 > wouldn't.  What I'm doing is describing a dialect.  (In my cases, I
 > don't have concultures, so they're not dialects from particular
 > groups, but any description will be of a dialect.)  The intent is
 > that this is the way that some people would or could pronounce that
 > language.  If I were to make a TY book, then I would include
 > information about how a given sound could be pronounced (isn't that
 > what they do, anyway?  "The dotless i in Turkish sounds a little
 > bit like the 'u' in 'could', but less round"), and the goal would
 > be for a native speaker to simply understand.  But that's not what
 > my conlang descriptions are.  As for how a native speaker of
 > another language would pronounce these sounds, while kind of
 > interesting, why devote any time to it?  After all, it's pretty
 > formulaic.  Like you said here:
 >
 >> <<Their English was excellent but their pronunciation and theyway
 >> that placed accents on different sylables took alittle getting
 >> used to.  But once I go used to it they were perfectly
 >> understandable.>>
 >
 > The reason you could get used to their pronunciation was because it
 > was systematic.  To figure out how a native speaker of another
 > language would pronounce a conlang would be a simple matter of
 > creating a formula: This sound will be pronounced like this, this
 > one like that, this one like that...  Further, if you consider all
 > the languages in the world, you could probably get every sound
 > having twenty different variants--maybe more.  Would you want to
 > include all these in a description?  Perhaps a separate page
 > devoted to each natural and created language?  How a native
 > Esperanto/American English/British
 > English/Mandarin/Spanish/Italian/Japanese/Hawaiian/Russian/Tagalog,
 > etc. speaker would pronounce X language.  I don't see why this
 > would be even desirable--by anybody.
 >
 > Also, consider you're notion of a nuance of the mouth.  You're
 > claim is that a native English speaker can understand anyone who
 > speaks with an accent.  This isn't controversial: We all can do it
 > (after all, we *all* speak with an accent).  You say, then, that
 > since this is the case, such phonetics shouldn't be studied.  But
 > notice: Don't you find it interesting that we *can* do this?  When
 > listening to a person speaking with an accent, it can sound either
 > odd or unintelligible at first, but after awhile, we "get used to
 > it".  Something allows us to adjust to understand them, and to do
 > so *very* quickly.  How the heck do we do that?  What would be too
 > jumbled to adjust to?  And further, if you got somebody with a
 > really thick accent (doesn't matter what), and told a native
 > English speaker that this person spoke no English at all, would the
 > native English speaker still be able to make the adjustment, since
 > they expect this person to be speaking another language?  These, to
 > me,are all fascinating questions, and they're dealt with by
 > phoneticians, and all of them have to do with what you've referred
 > to as nuances.
 >
 > So that's partly why I think nuances are interesting and
 > relevant.  Does this address your concern, or did you mean
 > something slightly different?  (I know it's easy to misunderstand
 > intention over the internet: I've done it tons of times, and I know
 > I'll do it again.)
 >
 > -David*******************************************************************
 > "sunly eleSkarez ygralleryf ydZZixelje je ox2mejze.""No eternal
 > reward will forgive us now for wasting the dawn."
 > -Jim Morrison
 > http://dedalvs.free.fr/

Reply