10.1283, Disc: Universal Word Order

[LINGUIST Network]

LINGUIST List:  Vol-10-1283. Thu Sep 2 1999. ISSN: 1068-4875.

Subject: 10.1283, Disc: Universal Word Order

Moderators: Anthony Rodrigues Aristar: Wayne State U.<aristar at linguistlist.org>
            Helen Dry: Eastern Michigan U. <hdry at linguistlist.org>
            Andrew Carnie: U. of Arizona <carnie at linguistlist.org>

Reviews: Andrew Carnie: U. of Arizona <carnie at linguistlist.org>

Associate Editors:  Martin Jacobsen <marty at linguistlist.org>
                    Ljuba Veselinova <ljuba at linguistlist.org>
		    Scott Fults <scott at linguistlist.org>
		    Jody Huellmantel <jody at linguistlist.org>
		    Karen Milligan <karen at linguistlist.org>

Assistant Editors:  Lydia Grebenyova <lydia at linguistlist.org>
		    Naomi Ogasawara <naomi at linguistlist.org>
		    James Yuells <james at linguistlist.org>

Software development: John H. Remmers <remmers at emunix.emich.edu>
                      Chris Brown <chris at linguistlist.org>
                      Qian Liao <qian at linguistlist.org>

Home Page:  http://linguistlist.org/


Editor for this issue: Anthony Rodrigues Aristar <aristar at linguistlist.org>

=================================Directory=================================

1)
Date:  Thu, 02 Sep 1999 11:51:56 -0700
From:  Roger Billerey <billerey at ucla.edu>
Subject:  Universal word order?

2)
Date:  Thu, 2 Sep 1999 14:06:52 -0700 (PDT)
From:  Dan Moonhawk Alford <dalford at haywire.csuhayward.edu>
Subject:  Re: 10.1263, Disc: Universal Word Order[, or The Great Whorf Hypothesis  Hoax Revisited]

-------------------------------- Message 1 -------------------------------

Date:  Thu, 02 Sep 1999 11:51:56 -0700
From:  Roger Billerey <billerey at ucla.edu>
Subject:  Universal word order?

Dear all,

In his The Antisymmetry of Syntax monograph, Richie Kayne proposes "a
very specific theory of word order" whereby "asymmetric c-command
invariably maps into linear precedence ... [Underlyingly,] complements
must always follow their associated head and ... specifiers and adjoined
elements must always precede the phrase that they are sister to." The
various surface word orders found in Malagasy, English, Korean, Welsh,
etc. would be determined by movement. A good read.

-
Roger Billerey
Graduate student
Department of Linguistics
UCLA
http://www.humnet.ucla.edu/linguistics/people/grads/billerey/billerey.htm

Discipline is never an end in itself--
Only a means to an end.


-------------------------------- Message 2 -------------------------------

Date:  Thu, 2 Sep 1999 14:06:52 -0700 (PDT)
From:  Dan Moonhawk Alford <dalford at haywire.csuhayward.edu>
Subject:  Re: 10.1263, Disc: Universal Word Order[, or The Great Whorf Hypothesis  Hoax Revisited]


[moonhawk]

Pardon me for butting in, but butt in I do when I see serious
misrepresentations or misunderstandings of Whorf fly by on this list, or
when someone's peddling the old, tired Hoax again.

Aren't people tired of kicking Whorf's corpse yet? You can't imagine, and
I mean that literally for some, what's actually in his essays when you
read them! You can even take off the Hypothesis Hoax filters and read them
for yourself -- essays that many Native Americans and quantum physicists
these days take much more seriously than most linguists do!

[John Thiels to Sean Witty]

In a message dated 99-08-30 19:42:26 EDT, you write:

<< Now, suppose the woman eats the apple and visits her boyfriend, who
offers to cook dinner for her. According to the SVO rationale and
Sapir-Whorf, the woman realizes that she has already eaten before she
realizes that the apple is what she ate. Thus, cognitive perception of the
preterit, 'to eat', precedes perception of the direct object, 'the apple'.
SOV speakers, therefore, must modify the order of cognitive perception to
fit the word order demands of their languages. >>

I think it is important to clarify here what one can actually find out
from the Sapir-Whorf hypothesis and what it refers to;

[moonhawk]

And if it even exists! See my "Demise of the Whorf Hypothesis" at my
webpage (at the end) for why there is no "the" (Sapir-) Whorf Hypothesis
since Whorf never wrote a hypothesis anywhere I can find -- only a
"principle of linguistic relativity," a different scientific concept than
a hypothesis.

"The hypothesis" came from researchers who developed their own hypotheses
(which they *generously* named after Whorf, being either too shy or afraid
to label them with their own names) from the principle (which they seldom
if ever cite), tested them, found them wanting, then blamed Whorf for
their own failures to construct testable hypotheses from his principle,
which they mostly misunderstood anyway, as shown next.

[John Thiels]

there are two versions, the strong version (which a careful reading of
Sapir will dispel) and a weaker version which is not usually contested as
such.

[moonhawk]

See same webpage for why this two-version theory, neither one of which
fits Whorf, makes these researchers even more deterministic than their
whipping boy Whorf, who was into complementarity (a la modern physics)
rather than an outmoded 19th-century version of science using Newtonian
'physical' monocausal determinism (as with billiard balls) in a totally
inappropriate way on mostly 'non-physical' language, cognition,
perception, worldview, etc. The fundamental error of social sciences has
been modeling themselves after 19th- instead of 20th-century science.

See the same website soon for a transcription of the Quantum Linguistics
Roundtable Discussion at the Flagstaff "Quantum Approaches to
Consciousness" conference on 8/1/99. Quantum Linguistics, launched at a
quantum physics conference to some acclaim, shares with quantum physics
and many Native worldviews the foundations of: consciousness (different
languages for different states of consciousness); non-locality (Human
language is a subset, a special case, of what we sometimes call telepathy
and rapport); and relativity (the language you use has a lot to do with
the way you think, and vice versa). Notice while there Nobel Laureate
Brian Josephson's comment that the "paranormal" realm [which would include
telepathy] is the quantum realm with details filled in that are averaged
out by science.

I am continuing Whorf's quest to get linguists and physicists dialoguing
together, especially given our common 20th-century base of structuralism,
which comes in handy when we are trying to understand each other.

[John Thiels]

The following are from my class notes from linguistics with Dr. Judith
Irvine, who has compiled many of Sapir's collected works and reconstructed
his lectures on the Psychology of Culture.

The strong version holds that language determines perception and therefore
expression...this is not what Sapir ever wrote.  Whorf said some things
which could be read that way, but even then, considering that his articles
were written for engineers not acquainted with linguistics, it is
important to read very carefully and consider what his inclusion of the
diagrams meant.

[moonhawk]

Agreed completely, except for the 'considering' clause. This reflects the
phase where Whorf had given up on linguists understanding him, and he was
casting his readership net more widely.

[John Thiels]

The hypothesis makes three propositions about language and thought:

First, there is the issue of linguistic relativism:

[moonhawk]

"Relativity" is the word Whorf consistently used, along with "principle",
both echoing Einstein, from whom he was trying to reclaim the Humboldtian
concept for linguistics. (See my "A Hidden Cycle in the History of
Linguistics," "Is Whorf's Relativity Einstein's Relativity," and "Stealing
the Fire" for fuller details -- at website). "

Relativism" comes from anthroplogical and other critics, and it obscures
Whorf's important link with physics for no good reason.

[John Thiels]

1) Ways of thought are intimately related to the structure of language
(Questions:  what is meant by thinking?  What are the relevant linguistic
structures?)

2) Relative linguistic structures significantly differ between language

[moonhawk]

we agree to here, but next you lapse into monocausal determinism, which
Whorf was not a believer in.

[John Thiels]

3) There is a causative link between some structures of language and the
thinking of its speakers.  Does that link influence only habitual,
inattentive thinking (weak version) or does it absolutely determine or
control or prevent alternatives (stronger version)

[moonhawk]

Challenge: cite page number of Whorf which advocates either strong or weak
determinism. Did your professor provide you with those? ;-) These two
forms are merely by-products of the analytical tool used, having nothing
whatever to do with what Whorf atually wrote or their antecedents in the
Humboldtian tradition per se -- only the linear-dominant minds trying
vainly to comprehend this "hidden cycle in the history of linguistics".

The only way I could repair your first sentence would be "There is a
causative link of mutual interdependence between some structures of
language and the thinking of its speakers." But even that isn't good
enough, because the KIND of thinking this most aligns with is the thinking
for speaking. Neither language nor thinking are monolithic.

[John Thiels]

The geneology of this idea is through Boas, Sapir and Whorf.

[moonhawk]

of WHICH idea? Maybe 1 and 2, but for 3 you must add the names of the
specific critics who interjected this 19th-century notion of science some
time after the mid-point of the 20th-century and its still ill-understood
scientific insights concerning reality.

[John Thiels]

Boas, who was one of the first to argue against the randomness of sounds
made by speakers of indigenous American languages (oh, how far we've come)

[moonhawk]

oh, PLEASE! Have you worked on a Native language? There are two levels of
arbitrariness, one absolute and the other variable. Yes, in principle,
there is no inherent connection between sounds in an arbitrary sound
system and the world, but on the second level, Native languages, just like
American Sign Language, are profoundly and intentionally iconic while we
cling to our arbitrariness. Boas was correct in this, to a large degree,
and the baby was thrown out with the bathwater when arbitrariness was
generally adopted.

[John Thiels]

... [Boas] spoke a lot about 1 and 2, and very little about three

[moonhawk]

exactly the same with Whorf and Sapir.

[John Thiels]

...where he came close was in the obligatoriness of certain categories for
speakers of particular languages. Jakobson took this up in his paper on
Boas, pointing out for example, that languages that force the speaker to
reveal the gender of a friend (amigo/amiga) do so...English, for example
does not and if someone asks, "Is it a male or female friend", the English
speaker can reply "It's none of your business."  English does not,
however, prevent one from noticing the gender of one's friend.  (Neither
does Portuguese, either)

[moonhawk]

oh, PLEASE again! Just because our system is covert (see Whorf)  instead
of overt doesn't prevent us from having to notice enough to be able to
assign gender correctly when using a 3rd person singular pronoun! Is this
still from your class notes? Yikes!

[John Thiels]

The strongest point Sapir himself made in this direction was in stating
that language is an essential part of the determination of SOCIAL life,
but he did not say the material world or the perception of it.  Whorf, who
came closest to stating #3 in strong terms,

[moonhawk]

Ahem -- where??! This is crucial: where??! Even most of his critics agree
he didn't hold the strong version, so where did you get this understanding
-  from your teacher?

See, this is what happens when we for decades virtually prohibit our
graduate students from reading Whorf in the original, or ever discussing
his ideas out loud in class without all thought being siderailed by the
academic hoax, with the result that Whorf's actual words and ideas seldom
get a fair academic hearing in most classrooms and gatherings.

[John Thiels]

... {Whorf] was not writing for linguists, and was talking about what
language forced you to notice about the material world and the strength of
convention in habitual thinking and expression.  These are hardly
earth-shaking propositions if, admittedly, difficult to test by today's
standards.

[moonhawk]

It was "earth-shaking" enough when Einstein -- who got the Humboldtian
idea of linguistic relativity from the Humboldtian-trained linguist Jost
Winteler, who owned the rooming house Einstein lived in as a grad student
in Geneva -- saw and proved that you can't describe an integrated
4D-spacetime with a 3D-space language, that you need a 4D-spacetime
language for doing it, which is a mathematical version of Humboldtian
linguistic relativity which led to new life-threatening physical realities
for us all.

If this knowledge isn't powerful (and dangerous), then why have all the
'social sciences' for decades ganged up on Whorf, who conveniently can't
defend himself, in their quest to be scientistic? Why have generations of
linguists been disuaded from reading him, and why has his reputation as a
careful scientific mind been so viciously trashed as our 'scientists'
tried to make a name for themselves?  Whorf was prophetic about where
linguistics needs to go, and it's definitely not in the direction of the
latest version of the Chomskyan/Pinker linguistic order. See the webpage
for my rebuttal to Chomsky in his only known (to me) frontal attack on
Whorf.

[John Thiels]

Much of the testing that has been done has been with categories directly
linked to lexical items and color terms that are as close as possible to
hard-wired in.  These are probably the least interesting aspects of
language to investigate and also very easily manipulated by consciousness.

[moonhawk]

As if we know what consciousness is? And they also had little to do with
anything Whorf ever wrote, but a lot to do with how the researchers
understood Whorf.

[John Thiels]

There is so much ink spilled in condemning or misunderstanding Sapir for
his supposed hypothesis,

[moonhawk]

let's say it straight -- somebody accused our honorable ancestor Sapir of
having something to do with the Unnamed Critics' Hypotheses spawned after
his death by other people trained theoretically rather than
anthropologically, in the tradition of Boas and Sapir; that's the only
supposing going on.

[John Thiels]

 ... but actually testing it is difficult.  If people are actually
interested, John Lucy has published two books on the hypothesis and his
experiments with speakers of a Mayan language in classifying certain kinds
of objects.  I have not read his books carefully but I have heard they are
wth a look. There is also a collection of papers looking at S-W called
"Rethinking Linguistic Relativity" (I have neither "bold" nor "underline"
right now) that you can look at with a multiplicity of views...

[moonhawk]

I agree. Don't forget Penny Lee's magnificent The Whorf Theory Complex.

[John Thiels]

Michael Silverstein of the University of Chicago has published a couple of
papers which might be interesting on this point.  One from 1984 is called
"The Limits of Awareness" and another is "Shifters, Linguistic Categories
and Cultural Description," republished in a book called Language, Culture
and Society (Waveland Press).

[moonhawk]

He's done fine work. See also Dan Slobin's excellent work on the Thinking
for Speaking notion (which obviates much Whorf criticism) -- that of the
many ways of thinking, there is one we use when rehearsing something to
say or in thinking out loud which is highly determined by the language we
use. This is the most fruitful context I know for fully exploring what
Whorf really said about the relationship of language to thinking (other
than my own, of course). ;-)

[John Thiels]

By the way, what about the VSO languages (such as Irish Gaelic) and OSV
(Malagasy)... German also makes you wait for the main verb quite often,
although its basic formula is SVO (Like this contribution, you have to
wait until the end to get to another point...)

[moonhawk]

And what about languages with topic/comment and other syntactic schemas --
just conviently ignore them in a quest for universals?

****

As to the original to which you were responding (LINGUIST 10.1263),

Date:  Sun, 29 Aug 1999 13:20:34 KST From:  "Sean Witty"
<wittysan at hotmail.com> Subject:  Intuitions on universal word order

Recently I read an article concerning the existence of a universal word
order. The author analyzes one argument for a universal SVO word order,
discusses what is wrong with that argument, and then proceeds to argue in
favor of a universal SOV word order. The most notable features of both
arguments presented are

1. The SVO argument belongs to a native speaker of an SVO language and is
based solely on other SVO languages. The author of the article, and a
proponent of the SOV argument, is a native speaker of an SOV language, the
argument is based on other SOV languages and flaws in the SVO argument.

2. Despite validity questions and obvious flaws present in both arguments
(i.e., selection of languages used in argumentation and obvious exceptions
that were overlooked), they are equally compelling.

I spent some time afterwards pondering over the issue of word order. Is
there a universal word order? If so, what is it? If not, how can
non-native speakers of a language casually acquire languages of the other
word order type? Why do human beings use two separate word orders to
achieve the same effect?

[moonhawk]

Wow -- only two human choices, eh? No topic/comment? That's the whole of
syntactic choices?

[witty]

According to the Sapir-Whorf hypothesis,

[moonhawk]

again, whose version of this academic hoax?

[witty]

a person's perception of the world, culture, is influenced by that
person's native language.

[moonhawk]

Except for "and vice versa," it's true as stated, IMO, but show me the
page number where either Whorf or Sapir wrote that.

[witty]

One hundred years earlier, Humboldt suggests that culture, as its
benefactor, influences language.

[moonhawk]

That's the "and vice versa" Sapir and Whorf were aware of, and also
probably true -- but where is it from?

[witty]

While there is plenty of evidence to support both sides of this argument,

[moonhawk]

as in, they are complementary aspects of the same, larger, complex
phenomenon which can only be analyically separated at great risk?! It is
in this higher, holistic level of thinking that Whorf functioned.

[witty]

neither is actually antithetical to the other and it is sufficient to say
that language and culture have an independent, yet mutually symbiotic,
relationship.

[moonhawk]

yes -- in a complementary, mutually determining, interdependent way. Take
a look at the yin and yang of the Tao to see graphically what you said
verbally.

[witty]

Regardless of cultural or linguistic affiliation, all "normal" humans
possess the same five senses and a brain that works, more or less, the
same way.

[moonhawk]

yes, guess, after we average out significant differences. See
Merleau-Ponty, The Phenomenology of Perception, for a wider notion of
perception, unshackled from the five-senses-only concept which we
otherwise receive from our culture.  "Normal" means ... what -- with the
extraordinary differences averaged out? When you DON'T average them out,
you get things like synaesthesia, as in The Man Who Tasted Shapes.

I find the differences in perception, cognition, and language at least as
interesting as I find the commonalities, as I believe did Humboldt, Boas,
Sapir, and Whorf. Before the hegemony of universals began demonizing
relativity (and the horse it rode into town on), universals and
differences were both "cool", in complementary opposition as principles of
linguistics. Some linguists like researching universals, some like
differences, and some like both. Whorf, for instance, had 3-4 different
explanations and statings of his single relativity principle, but probably
has 6-7 statements of different universals (by my vague recollection),
which universalists have never explored after labeling him a relativist.

[witty]

Generally, cultural affiliations determine one's perception of the world
and one's linguistic affiliation; linguistic affiliations determine how
one communicates with others.

[moonhawk]

 ... if one wants to wallow in 19th-century determinism more suited to
billiard balls than to invisible mental/linguistic forms, that is.

[witty]

Languages, therefore, represent cognitive perception after cultural
influence. This would suggest, therefore, that it is not an individual's
perception of the world that is influenced by culture, but the way in
which that world is expressed.

[moonhawk]

This is too simplistic -- and outmoded 19th-century science. You're
forgetting about the net of meaning we cast over our reality, probably at
the speed of light. But that's another story about Einstein's relativity.

Language and culture are also complementary aspects of each other -- a
fact for anthropological linguists which got lost in autonomous
Chomskymania.

[witty]

Many people have had thoughts that were difficult to put into words, and
many languages do not have forms that other languages do. The important
thing is that people still have these thoughts and ideas, even if the
language does not.  Thus, whatever influence culture and language have on
each other, it starts at the deep structure, where cognition meets
language.

[moonhawk]

see Dan Slobin on Thinking for Speaking. This has been a non-issue for a
long time. It's probably not Chomsky's deep structure, that intersection.

[witty]

If the universal word order is SVO/SOV (choose one), then all speakers of
SOV/SVO (choose the other) languages would necessarily be speaking a
language in opposition to their cognitive processes. This frame of logic
begs three questions:

1. What would prompt a culture group to adopt a linguistic affiliation
that is antithetical to cognitive reality?

2. What advantage is gained by doing so?

3. How does one explain the linguistic processes involved?

[moonhawk]

and what if there is no universal word order? Do you mean to imply that
cognitive reality is fixed and has nothing to do with language?

[witty]

In the original article, questions 1 and 2 are ignored.  Perhaps this is
appropriate, since linguists hardly ever concern themselves with
non-linguistic questions of "why". When faced with two equally plausible
arguments, however, it is always preferable to select the simpler of the
two. In this case, not only are the arguments equally plausible, they are
equally complcated.  Thus, questions 1 and 2 become appropriate because
both sides have presented "equal" arguments as answers to question 3.

[moonhawk]

both equally fatally flawed.

[witty]

The answers to the first and second questions are quite easy: they would
not. People are generally lazy and, as such, they prefer language systems
that are simple and easy to master. If the opposite were true, then the
Roman alphabet would never have developed and the Northern Semitic
alphabet [sic!] might still be used.

[moonhawk]

I must have missed the article or book demonstrating that the Roman
alphabet is simpler and easier to master than a *syllabary* for speakers
of simple-syllabic languages. Can you provide a cite? Otherwise, this just
seems to prefer what you are used to.

[witty]

Many people attest to changes in language, but our languages really do not
change -- simply our usage of them. In almost all cases of linguistic
change, the new form represents a simplification of the older form (for
example, enclitic mutation, pictographic vs. phonetic writing systems, and
metathesis). Without any advantage to be gained, a clumsy linguistic
system that is antithetical to the cognitive process would quickly become
extinct in favor of a less complicated, more convenient language.

[moonhawk]

huh? I can't buy the argument, which simply ignores other changes which do
not simplify.

[witty]

Since both sides provide convincing arguments, yet attribute behavior that
is inconsistent with human nature, perhaps they are both right AND they
are both wrong?

Suppose a woman sees an apple on a table. According to the SOV rationale
and Sapir-Whorf,

[moonhawk]

or somebody's version of the hoax, at least, unless you can provide a
Sapir or Whorf citation.

[witty]

she realizes the apple before realizing that she sees it, and few would
argue differently. Thus, cognitive perception of the direct object, 'the
apple', precedes perception of the preterit, 'to see'. SVO speakers,
therefore, must modify the order of perception to fit the word order
demands of their languages.

Now, suppose the woman eats the apple and visits her boyfriend, who offers
to cook dinner for her. According to the SVO rationale and Sapir-Whorf
[sic!], the woman realizes that she has already eaten before she realizes
that the apple is what she ate.

[moonhawk]

HUH?! See how crazy this gets without supporting quotes?!

[witty]

Thus, cognitive perception of the preterit, 'to eat', precedes perception
of the direct object, 'the apple'. SOV speakers, therefore, must modify
the order of cognitive perception to fit the word order demands of their
languages.

[moonhawk]

I hope this makes more sense to others than it does to me! And what does
this mean for speakers of languages where you "can talk all day long and
never utter a single noun"? They never know what they eat?

[witty]

In truth, it is impossible to say that, 100% of the time, perception is in
accordance with the word order of one's native language. As such, it makes
sense that every language, as a universal rule, would have a primary word
order (SOV/SVO) and linguistic processes for dealing with perception tha
does not conform to this order.

[moonhawk]

nonsense! still only two choices?

[witty]

In the case of the universal SVO word order argument, the formulae to
explain the derivation of the SOV surface form do not support universal
SVO, but do explain how SOV languages deal with SVO perceptions. Similar
formulae in the SOV argument neither explain the derivation of the SVO
surface form nor support universal SOV, but do explain how SVO languages
deal with SOV perceptions.

Since there exist occasions when cognitive perception can be either SVO or
SOV, it seemikely that bot forms are uiversal at the cognitive level. When
these perceptions are converted into linguistic forms, the deep structure
of the language forces conversion of perceptions that do not conform to
the word order of the language. The only universal truth about word order,
then, is that the subject must precede the verb -- but this is another
issue.

[moonhawk]

nonsense again, leaving out topic/comment, among others.  Subject/object
discriminations are projected onto reality by users of the linguistic
structure and then SEEN there as real! Sometimes even by linguists!

What if you spoke a language with 80 permissible syllables, each of which
iconically embodied root-level kinesthetic feelings of biological
movement, process, and relationship?  These roots combine and recombine
endlessly to create what we woulld variously label a single word OR a
sentence, with a morphosyntax that has so far defied sensible analysis
because of universalist blinders. A language with no separate subjects and
objects, no tense system (evidentials instead), a kind of "vector" system
instead of our pronouns -- where you can talk all day long and not utter a
single separate noun, and you don't ever use metaphors because the
transparent-root system nearly forces you to make up completely novel
WORDS (and be understood) with the same ease with which we utter novel
sentences. At least, that's how the doctorate-educated Natives explain it
to me, from the insides of such languages.

Can either of you fit that into a universalist pipe and smoke it? ;-)

warm regards, moonhawk

Visit Moonhawk's new webpage at
<http://sunflower.com/~dewatson/alford.htm>
for recent presentations and hard-to-find classic articles.
(111,111,111 x 111,111,111 =  12,345,678,987,654,321)




---------------------------------------------------------------------------
LINGUIST List: Vol-10-1283