Language Universals

skilaatara · Post by **skilaatara** » Sun Nov 16, 2003 12:19 pm

Would it be unreasonable to break language universals?

gach · Post by **gach** » Sun Nov 16, 2003 12:32 pm

I don't see any bad breaking "universals". It's of cource a good indicator that a structure works if a natlang uses it. But it doesn't say anything if there just aren't any langs to use it. It's good enoung when you can justify your language reasonably.

Aurora Rossa · Post by **Aurora Rossa** » Sun Nov 16, 2003 12:56 pm

If it's a human language, then yes. If 6,000 languages follow a rule, it's probably not just coincidence. If it's not a human language, then violate as many universals as you want. A truely alien language will have little in common with our languages.

Whimemsz · Post by **Whimemsz** » Sun Nov 16, 2003 1:43 pm

Eddy the Great wrote:If it's a human language, then yes. If 6,000 languages follow a rule, it's probably not just coincidence. If it's not a human language, then violate as many universals as you want. A truely alien language will have little in common with our languages.

I see no problem violating a universal, even if it IS a human language. For one thing, as Glenn points out, we are creating these people, and we should be able to do anything with their language we want (in theory). Secondly, many of our conworlds seem to be in alternate universes/realities. Hence, I'd say we can break the rules there.

And finally, these "language universals" are not concrete rules set in stone. They are simply characteristics that all of the languages on Earth studied so far share. There is absolutely no reason why a certain language wouldn't have, say, just the stop /t/. It's just never happened before.

Aurora Rossa · Post by **Aurora Rossa** » Sun Nov 16, 2003 1:51 pm

I find it hard to believe that these universals are simply chance. There has to be a reason why 6,000 languages follow these rules.

Whimemsz · Post by **Whimemsz** » Sun Nov 16, 2003 1:52 pm

Eddy the Great wrote:I find it hard to believe that these universals are simply chance. There has to be a reason why 6,000 languages follow these rules.

These are patterns, not rules. There is absolutely no reason why a language couldn't break them.

Aurora Rossa · Post by **Aurora Rossa** » Sun Nov 16, 2003 1:59 pm

Then why hasn't any language violated any of them?

Mecislau · Post by **Mecislau** » Sun Nov 16, 2003 2:02 pm

Eddy the Great wrote:I find it hard to believe that these universals are simply chance. There has to be a reason why 6,000 languages follow these rules.

It's probably better to compare families. Languages within one family will have shared characteristics, including a full stop system, if there is one. Also, as languages interact, the differences between them can shrink as well. The only way to truly find out is to restart human history, and that's exactly what we're doing.

Whimemsz · Post by **Whimemsz** » Sun Nov 16, 2003 2:03 pm

Eddy the Great wrote:Then why hasn't any language violated any of them?

Why? Because no languages exist right now that don't follow the patterns. We have no way of knowing whether they have been violated in the past. And I recall Jeff saying that one of the Indian languages is very close to breaking that universal (the 2 of /p/, /t/, /k/, /?/ universal).

6,000 languages sharing a certain characteristic does not mean that a language COULDN'T lack that characteristic.

jburke · Post by **jburke** » Sun Nov 16, 2003 2:52 pm

Whimemsz wrote:And I recall Jeff saying that one of the Indian languages is very close to breaking that universal (the 2 of /p/, /t/, /k/, /?/ universal).

The distribution of /p/ and /k/ in Cheyenne is very sparse; most Proto Algonquian free-standing *p and *k were lost (resulting in the very distinctive geminate vowels of the language). Almost every /p/ and /k/ in Cheyenne is a reflex of a PA consonant cluster (and many clusters involving *k simply became glottal stops, decreasing the distribution of /k/ even more). It wouldn't be unreasonable to say that, in a few hundred years, Cheyenne might not have either a /p/, /k/ or both.

Glenn · Post by **Glenn** » Sun Nov 16, 2003 3:53 pm

I must say, I knew that my previous post would probably spark a discussion (and thus I added it with trepidation

), but I'm very glad that said discussion received a separate thread of its own. Thanks, skilaatara!

Everyone seems to have done a good job of stating their positions so far; carry on!

p@,
Glenn

butsuri · Post by **butsuri** » Sun Nov 16, 2003 4:00 pm

First of all, most "universals" aren't universal. They're tendencies, statistical correlations. See e.g. the Universals Archive at the University of Konstanz (the archive itself seems to be down again, but bookmark that page and check back later) or a draft of an article on word-order universals by Matthew Dryer.

As for those universals to which no exception is known (such as, presumably, this one about stops), well, languages violating them will probably always be rare, but there is only a finite number of languages currently spoken on Earth and not all of them have been well described by linguists. Until the late 1970s there were no languages known in which object preceded supject and verb in basic word order. Then Hixkaryana (OVS) was discovered. There are only a very few OVS and OSV languages reported (and in several cases the basic order is disputed), and none of them have very large populations. It's easy to imagine a world in which these languages died out without being studied, and then we would not know that they were possible. But thanks to Hixkaryana we do.

It's probably a good idea to know which features of your conlang are typologically unusual, and to remark on it in your grammar. But you shouldn't necessarily think of them as unbreakable laws. Rather, they're probabilistic, and a feature can have a finite probability which is sufficiently low that it isn't guaranteed to show up in a sample of 6000. (Of course, there may be some truly absolute universals too. But you can't identify them simply by observing that a feature doesn't exist in any documented language.)

So Haleza Grise · Post by **So Haleza Grise** » Sun Nov 16, 2003 4:54 pm

I don't mean to sound like a hopeless Chomskyite (which I'm not), but I believe there are universals that all languages follow (and a great many other ones that are not true universals).

I don't think, for example that things like say basic sentence structure should be tampered with - unless the conlanger has a great deal of experience and knowledge. My view is, stick to the basics to start with, and the more practice you get, the more you will be able to make your languages unearthly. There's nothing worse, in my view, than a conlang that attempts to be inhuman but just ends up being an unreal mess.

skilaatara · Post by **skilaatara** » Sun Nov 16, 2003 5:35 pm

All my conlangs are for humans; unlike Zomp or Tolkien I have little use for nonhumans. However, I believe this is largely irrelevant (except when it comes to sounds).

I don't think there is any genetically inherent grammar in our bloods that causes us to use those language universals, anymore than there is genetic determination of which language we can speak. The language universals that I've run across have all seemed basically about convenience and clarity: /p,t,k/ for example because of the distinctness of the sounds and the high contrastiveness of stops with vowels. If this "convenience" can be discarded in various languages, than so can "conviences" of the sort that determine "markedness" of this and that grammatical feature.

I know it was odd of me to start this thread with a question on a topic where I already have a strong opinion, but I know a lot of you have more of a linguistic background than I do, and I'm curious as to everyone's opinions.

Soap · Post by **Soap** » Sun Nov 16, 2003 6:32 pm

Actually voiceless stops are pretty hard to distinguish, and even voiced stops are pretty hard to distinguish, as the person who showed us the McGurk effect helped me realize. Again, I'm going to use my baby theory, that is, ... humans will distinguish whatever babies distinguish, plus some more innovations. That's just a hunch though, it could be totally wrong.

jburke · Post by **jburke** » Sun Nov 16, 2003 6:52 pm

skilaatara wrote:All my conlangs are for humans; unlike Zomp or Tolkien I have little use for nonhumans.
Most of Tolkien's nonhumans are, well, human-like. The elves, the dwarves, and esp. the hobbits.--all essentially human in character. The elves and dwarves are rather more like exotic cultures than radically different races.

I don't think there is any genetically inherent grammar in our bloods that causes us to use those language universals, anymore than there is genetic determination of which language we can speak.

No, the language intsinct is a bunk hypothesis. (Chomsky despises--or did at one time--Whorf's theories, which Chomsky analyzed as hopelessly deterministic. The irony of it all is that Chomkey's language instinct nonsense is far more deterministic than anything Whorf ever said.)

jburke · Post by **jburke** » Sun Nov 16, 2003 6:57 pm

Mercator wrote:Actually voiceless stops are pretty hard to distinguish, and even voiced stops are pretty hard to distinguish, as the person who showed us the McGurk effect helped me realize. Again, I'm going to use my baby theory, that is, ... humans will distinguish whatever babies distinguish, plus some more innovations. That's just a hunch though, it could be totally wrong.

p,t,k all sound pretty distinct to me, esp. when aspirated. I have to wonder, though, just how much the quest for distinctness lies behind phonological universals. After all, languages are full of grammatical indistinctness--AKA ambiguity--but context and other cues smooth all that over. Why should phonology be any different?

jburke · Post by **jburke** » Sun Nov 16, 2003 7:03 pm

Addendum:

E.g., consider the most common tri-vowel set: i,a,u. There are other sets of three vowels that are just as distinct (e,a,u, with the /a/ as a low back vowel, e.g.), but they're nowhere near as common.

Aurora Rossa · Post by **Aurora Rossa** » Sun Nov 16, 2003 7:10 pm

And why is that? Coincidence?

jburke · Post by **jburke** » Sun Nov 16, 2003 7:12 pm

Eddy the Great wrote:And why is that? Coincidence?

And why is what?

Nuntar · Post by **Nuntar** » Sun Nov 16, 2003 7:14 pm

Why is it that other vowel sets aren't near as common?

We had a discussion on vowels a while ago, so I can answer this one. Nobody knows.

Soap · Post by **Soap** » Sun Nov 16, 2003 7:34 pm

My theory is that it is because we unconsciously gravitate towards vowel patterns that babies prefer. Why babies prefer those vowels is another unknown.

jburke · Post by **jburke** » Sun Nov 16, 2003 7:37 pm

Mercator wrote:My theory is that it is because we unconsciously gravitate towards vowel patterns that babies prefer. Why babies prefer those vowels is another unknown.

That's quite a bizarre theory, and I don't see any particular plausibility in it.

Neek · Post by **Neek** » Sun Nov 16, 2003 8:20 pm

But don't babies emulate speech habits of their parents? So wouldn't that mean that vowel systems don't change, but they change more often (and sporadically) than consonants?

Dudicon · Post by **Dudicon** » Sun Nov 16, 2003 8:27 pm

The most plausible explanation for the ubiquity of the a i u system is simply that it makes the most efficient use of the vowel space of any three vowels. Essentially, those three vowels have the most distance between themselves of any set of three vowels.