Re: Human Language. (long post)

Michael McBroom (
Mon, 06 Jan 1997 14:32:14 -0500

John A. Halloran wrote:

> >But if
> >we're really going to compare early hominids to carnivores that live and
> >hunt in packs, why not hyenas or lions, since humans originated in
> >Africa?
> You are right that it would help to find a study on the extent to which wolves
> range widely in comparison to the animals that you mention. Gorillas and
> chimpanzees range over no more than 15 square miles, while even the most
> primitive humans range over hundreds of miles.

Geez, here we go with the wolves again.

> The principal question is for what reason have other species developed a vocal
> tract that permits or enhances long-distance vocalizations. This can throw
> light on the evolution in the human vocal tract.

Oh really. Your principal question is germane to the issue only if we
assume that the human vocal tract evolved due to selectional pressures
that favored long-distance communication. There is clear evidence that
it did not. When's the last time you've tried to carry on a
conversation by shouting to somebody down the street? Pretty hard to
do, huh? The hard consonants don't travel, and even the ones that do,
don't travel that far. What you're left with are open vowels and vowels
that are "colored" by the nasals and the retroflex [r]. One of the
principle advantage we possess as users of a full-blown system of
language is the ability to communicate rapidly with each other. You
just can't do this sort of thing at a distance.

> >> It is a fact that wolves have developed a vocal tract that permits
> >> long-distance vocalizations. I believe this is related to their hunting
> >> lifestyle, in which they are able to range widely apart but come together when
> >> prey is being attacked.
> >Do wolves "announce" the attack to the rest of their far-ranging pack?
> >Or do they attack in relative silence? I always thought they kept
> >rather silent when they got down to serious business, and that they were
> >too busy snarling and snapping at each other when it was time to eat.
> >Hardly the kind of "come and get it!" behavior that you seem to be
> >suggesting.
> It takes more than one wolf and more than one hominid to bring down a large
> mammal.

So, therefore, humans would have been required to develop the same type
of communication systems that wolves possess. Of course. I see now.

> >> Do you mean anything by this other than descent of the larynx?
> >Thank you. Back on topic. The descent of the larynx is that which
> >leads to the choking hazard I had mentioned before. They descent of the
> >larynx also serves to open up the pharyngeal area, and allows the tongue
> >to move downward. But the arching of the basicranial area is perhaps
> >even more crucial, since it is what gives the tongue the freedom to
> >produce the three vowels with the most acoustic separation ([i], [a],
> >[u]) and the back consonants.
> >
> >> You know what throws cold water on Liberman's elaborate theory about the
> Homo>> sapiens sapiens vocal tract developing to be able to produce more
> vowels? The>> assertion by historical linguists that Proto-Indo-European only
> had one vowel!>> In "The Primitive Features of a Protolanguage", Bernard H.
> Bichakjian asserts>> that it had "a very limited number of vowels, presumably
> only one in its first>> reconstructed phase.", namely the vowel e.
> >Bichakjian's conclusions are highly controversial and have been almost
> >universally rejected.
> References please?

Don't have 'em handy, I'm afraid. I'll have to go dig 'em out. It'll
be in the historical linguistics literature. Look for peer review
commentaries to these assertions. BTW, Bichakjian has written
extensively on the subject of language evolution, which, you must
understand, is a different research field from historical linguistics.
I haven't as yet had the opportunity to read any of his writings on the
matter. But I will.

> But even if we grant for the sake of argument
> >that he's correct (which he isn't), we cen stell meke erselves endersted
> >quete well even ef we heve enle ene vewel en er enventere ef sends. So
> >what? Arabic only has three phonemic vowels. But in Arabic script,
> >vowels are largely ignored altogether. Same with Hebrew. So, if one
> >can ignore vowels entirely in one's writing system, which do you think
> >are more important as carriers of information, vowels or consonants?
> >Unfortunately, consonants don't carry very well at all over distances,
> >so there goes your wolf theory.
> In Comparative Indo-European Linguistics, by Robert S.P. Beekes, 1995, there
> is the statement, "PIE had only two vowels: e and o." p. 137. Bichakjian
> claims that the o developed as an ablauted e.

Ablaut is a common enough ocurrance, especially in Indo-European
languages. But since you seem to be fixated on Indo-European, grab a
copy of the American Heritage College Dictionary, and turn to the very
back. You will find a dictionary of Indo-European roots there, compiled
by Calvert Watkins, an internationally renowned scholar of IE studies,
along with an introduction written by Watkins. (BTW, you'll need to
locate a copy of either the 1st or 3rd edition, the roots dictionary was
left out of the 2nd edition) Watkin states that PIE contained the vowel
sounds [i], [e], [a], [o], and [u], as well as the semivowels [w] and
[y], and the diphthongs [ei], [oi], [ai], [eu], [ou], and [au]. So, I
guess it's all a matter of who you believe. Those environments which
Beekes and Bichakjian suggest that existed are highly unlikely, although
not impossible. Parsimony would side with Watkins, since this situation
is much more common. Thus one should exercise caution in adopting the
views of Bickakjian and Beekes without careful consideration of the
arguments on both sides.

> Consider the logic of Lieberman and Laitman's claims. They are saying that
> the ability to produce distinctive vowels drove the descent of the larynx.
> But if this was true, it can have no relation to abstract language because
> when you look at an early language like PIE, the vowels are unimportant.

They were not, but even if they were, it still doesn't matter. Look,
you need to understand that PIE, a language that existed some 5,000
years ago, would have been indistinguishable from so-called "modern
languages" in the sense that it would be just as good of an information
carrier for its native speakers as any other natural language. PIE
doesn't hold any greater evolutionary significance than Proto-Semitic or
Classical Chinese. It simply isn't old enough. In fact, most
historical linguists will tell you in no uncertain terms that it is
impossible to determine much of anything about languages that existed
prior to about 5,000 ya or so. I dunno if I believe this entirely, but
that is the received wisdom at the moment.

The fact remains that we *have* the capability to produce a wide variety
of vowel sounds, and that many examples of words exist which are
different in meaning, but identical in sound except for their vowels
(e.g., beet, bit, bet bait, bat, bot, but, boot), which strongly
indicates that a supralaryngeal airway that was able to produce these
distinctions became a selectional advantage. You cannot offer up a
single language in the hopes that it will be old enough to refute this
characteristic of human anatomy.

> This does not mean that improved vowel production was not
> important to nonlinguistic vocalizations, however. But you have not dealt
> with the issue that descent of the larynx also aids in producing deeper,
> lower-pitched vocalizations.

I have stated repeatedly that non-linguistic vocalizations were not what
our vocal tract has been selected for. I have also stated that the
descent of the larynx was for different reasons than the specific
production of lower-pitched vocalizations. In other words, I reject
your claims on the basis that there is no evidence to suggest that the
human vocal tract evolved for the uses you indicate.


Michael McBroom
CSUF Linguistics