LINGUIST List 8.1409

Thu Oct 2 1997

Disc: Author's reply to Review of Yngve 1996

Editor for this issue: Andrew Carnie <carnielinguistlist.org>

Directory

1. Victor H. Yngve, Re: Reply to review

[Editor's note: The following is the Author's Reply to the
 Review by Pius ten Hacken of the Book:

Yngve, Victor H. (1996) From Grammar to Science: New Foundations for
General Linguistics. Amsterdam/Philadelphia: John Benjamins

Which appeared in issue 8.1277. This review can be found at:
http://linguistlist.org/issues/8/8-1277.html ]


Reply writeen by Victor Yngve <v-yngveuchicago.edu>


0. Introduction

I would like to thank Pius ten Hacken for separating his description of the
book from his own comments. The description is reasonably accurate, but it
does require some correction. His criticisms, however, appear to stem
mainly not from the book itself but from false stereotypes of standard
science found in some writings in the philosophy of science.

1. The question being asked

The preface of the book begins in this way: "How is it that when I open my
mouth and some sound comes out that you can, most of the time, understand
what I am saying? How does it work?" This initial question is elaborated
throughout the book. Yet the reviewer complains that the book never
addresses the question of what the theory is to explain.

Traditional and current linguistic theories, if they try to approach this
question at all, try to approach it not directly as a legitimate question
in a natural science focused on people and sound waves etc. but indirectly
through some reworking of ancient semiotic-grammatical theory. But this
ancient theory was originally designed as a philosophical theory of
knowledge and elaborated for the purposes of normative grammar. It is
inappropriate for a scientific task and does not serve the need. As science
it is severely ontologically crippled or, if you prefer, ontologically
challenged. We have been reduced to this indirect approach because there
has been no alternative. There has been no foundational body of theory for
linguistics available other than in the semiotic-grammatical tradition--and
theories in that tradition cannot be scientifically justified at the most
basic level in spite of any claims to the contrary.

The book, culminating several decades of research, tries to remedy the
deficiency by laying new foundations on which linguistics can build a true
natural science focused on real people embedded in the real physical world.
Further, it provides a scientifically justified notation and a survey of
how the notation can be applied to the study of a variety of linguistic
phenomena. This is the only scientifically justified notation available in
linguistics. The possibility of having a scientifically justified notation
is one of the rewards of moving to a study of the real world.


2. Philosophy of science

Another reward for moving to a study of the real world is that we do not
need to innovate. We do not need to rework the foundations of science laid
by Galileo and the other greats who showed the way and I do not intend to
try. With three degrees in physics and several publications in that field,
I think I know what science in the real world is. The review admits that
"many practicing scientists may agree with Yngve's concept of science..." I
rest my case. Virtually all agree with it: It's quite standard in physics,
chemistry, and most of biology. The many advances in our understanding of
the natural world that these sciences have given us rest on it. That's good
enough for me. We can accept these foundations as they are if we move to a
study of the real world and make linguistics a natural science rather than
a discipline focusing on invented objects and abstract models that have no
contact with reality at the most basic level.

Yes, chapter 8 is devoted entirely to standard science, a chapter that I
would not have had to include for an audience of natural scientists. It's a
shame that many otherwise well-trained linguists do not have any serious
background in the natural sciences. Thoughtful or concerned linguists have
had a difficult time trying to make a science where no standard science is
possible. Some philosophers and philosophical linguists, trying to be
helpful I presume, have obliged by proposing to change the rules of the
game and redefine science so that it does apply to the study of invented
objects and would no longer require contact with reality! If science were
to tolerate this sort of thing it would be set back centuries. It would
lose its hard-won ability to turn back superstition, myth, untested folk
theory, arbitrary assumptions, sheer nonsense, charlatanism, and false or
fraudulent claims. Medical research, for example, would be disastrously
shackled in its life-and-death struggle against disease.

The reviewer also concurs "that a lot of disagreement exists among
philosophers of science." If linguistics moves to study the real world,
philosophical acrobatics will be unnecessary and we will not have to
"concentrate on general epistemological questions." May I reiterate that
one would not go to a music critic to learn to play the violin, no matter
how insightful his critical writings may have been.

The review misrepresents my view and the standard view of science in
important respects. The reviewer's background in philosophy has apparently
led him to conflate and confuse the standard view of science with certain
philosophical straw men in spite of my best efforts in chapter 8. A symptom
of this is the non sequitur of claiming I would have to commit myself to
the view that musical taste is permanent and universal. This reflects a
complete misunderstanding.

The review states, "Thus in Yngve's view scientific knowledge is permanent
and represents truth. ... Even logical positivists such as Ayer
acknowledged that strong verifiability, leading to permanent certainty of a
statement, is only possible in tautologies."

Certainly I never championed strong verifiability nor is it the position of
standard science. Where did this criticism come from? It sounds like one
philosopher disagreeing with what he thinks another philosopher's position
is. What I did say in chapter 8 (8.2) is:

"Our theories are always held to be tentative, although many of them are
surely correct and will almost certainly never have to be abandoned. The
fact that scientific "truth" is always tentative does not mean that we can
never get correct answers, or that we are condemned to continual revisions
in science, as is the case in the grammatical tradition. Science is surely
correct in stating that the earth is not flat, as was once thought, but
shaped like a ball: there is plenty of evidence for that and even for the
details of its shape. And it is surely true that malaria is not caused by
bad air as was earlier thought (hence its name), but by a plasmodium
organism with a life cycle partly in a mosquito. The fact that these pieces
of validated scientific truth are tentative does not mean that we can't
have finally achieved a correct understanding of the matter."

What could be clearer? And who would object to it? It is simply not correct
to say that "Thus in Yngve's view scientific knowledge is permanent and
represents truth." or to imply that I hold to "strong verifiability,
leading to permanent certainty of a statement."

The review also seems to confuse logical truth, which is supposed to be
absolute, and truth about the real world, which is always tentative.

Also, nowhere do I claim or imply "that Einstein's theory of relativity is
a straight continuation of the permanent foundations laid by Newton's
science." In fact, nowhere in the book do I even mention Einstein or
Newton. This seems to me to be the result of conclusion jumping coming
perhaps from some philosophical dispute the reviewer has in mind.

Perhaps it's a small matter but I think it may be symptomatic. The review
states (1.1): "In stoic philosophy, reality was divided into physical and
logical domains."

What the book says is (2.1): "The Stoics saw fit to divide philosophy into
three parts, the physical, the logical, and the ethical. It is the
distinction between the physical and the logical that will engage our
attention."

It was not reality but philosophy that the Stoics divided. This may reflect
a confusion about "reality," possibly stemming from confusions in the
philosophical literature where "reality" is sometimes unreal or in the
social-science literature where "reality" is sometimes "constructed."

The reviewer's negative judgment of my views and of standard science is
simply misinformed. When he comes to understand them, he may change his
mind and turn his analytic abilities, training, and insight to help with
one or more of the exercises suggested below, which now take on
considerable urgency as research questions.


3. Criteria of science

The book introduces the criteria of acceptance of science in the following
words (8.3):

"The standard criterion of acceptance of hypotheses or theories in science
when doubts arise is the ability of their predictions to pass tests against
the real world by means of careful observations and experiments."

"The standard criterion of acceptance of observational and experimental
results in science is their reproducibility when questioned."

The Nobel Laureate in physics, the late Richard P. Feynman put it in the
following words, which I selected as the epigraph to the book:

"The principle of science, the definition, almost, is the following: The
test of all knowledge is experiment. Experiment is the sole judge of
scientific "truth." But what is the source of this knowledge? Where do the
laws that are to be tested come from? Experiment, itself, helps to produce
these laws, in the sense that it gives us hints. But also needed is
imagination to create from these hints the great generalizations--to guess
at the wonderful, simple, but very strange patterns beneath them all, and
then to experiment to check again whether we have made the right guess"
(Feynman, Leighton, and Sands 1963:1-1).

(By "experiment," Feynman includes observation.)

Perhaps the reviewer does not agree with this standard view of science,
seeing it as "naive." I think it is a simple and straightforward position
developed through over 400 years of experience in trying to find out about
nature. It is a mistake to think that we can ever have a proper science of
invented or assumed immaterial objects where these criteria cannot be
applied. It is misguided to propose other ad hoc criteria to apply only in
respect to language and grammar, such as "descriptive adequacy" and
"explanatory adequacy," and then represent the result as scientific.

As an aid to further study and to move us forward, may I suggest the
following exercise:

Exercise (type 1). For your favorite linguistic approach (or your favorite
disfavorite) identify the criteria of acceptance that are being proposed or
that are tacitly invoked.


4. Assumptions

It is well known that no result can be more secure than the assumptions on
which it is based. If the assumptions are false or nonsense, any results
based on them fall under suspicion of being false or nonsense. Science has
a long history of doubting its assumptions and discarding all but the bare
essentials. We are left today with the following four standard assumptions
of science, introduced in chapter 8 in the following way and further
explained there (8.4):

"The first assumption of science is an ontological assumption, that there
actually is a real world out there to be studied."

"The second assumptions of science is a regularity assumption, that the
real world is coherent so we have a chance of finding out something about
it."

"The third assumption of science is a rationality assumption, that we can
reach valid conclusions by reasoning from valid premises, that we can trust
our ability to calculate predictions from our theories for comparison with
the real world."

"The fourth assumption of science is a causality assumption that observed
effects flow from immediate real-world causes."

Thus astronomers assume (1) that the objects they observe are real, (2)
that the astronomical world is coherent so there is a possibility of
finding out something about it, (3) that they can trust their ability to
calculate in making orbital predictions on the basis of their theories and
thus predict observable phenomena, and (4) that from the phenomena they
observe they can infer causes in the properties and motions of the real
astronomical objects and thus test their theories against reality.

Science routinely casts doubt on any proposed additional assumptions.
Efforts would be made to convert them into hypotheses and test them. If
they did not survive the tests they would be given up. If they could not
even be tested, they would not be accepted into science but, at best,
placed in the realm of interesting speculation. If they did survive the
tests, they would be removed from the realm of assumptions and moved into
the realm of empirical results.

The grammatical tradition has always rested on a number of assumptions
differing from the above four. As pointed out in the book (3.5) Saussure
raised questions about the ontological status of linguistic objects, that
they were not given in advance in nature like the objects of study in the
other sciences but that they were created by a point of view. It was
Bloomfield (3.6) who pointed out that one could not have a science of
language above the level of phonetics, so he explicitly introduced the
traditional objects of study into the discipline by assumption, his
"fundamental assumption of linguistics." If we were to convert this
assumption into an hypothesis and try to test it, it would fail, as
Bloomfield knew. So we have suspected since Saussure and known since
Bloomfield's 1926 paper and his 1933 book Language that linguistics in its
tradition of studying language and the objects of language cannot be a
proper science. Bloomfield explicitly took his assumption to be tentative
until that time when a proper alternative would become available. He
mistakenly thought it would require the prior perfection of a number of
other sciences.

That it has taken so long to see through the situation is a testament to
the strength of the semiotic-grammatical tradition not only in linguistics
but also in philosophy and in our everyday concepts invoked in talking
about talk. It is also a testament to the strength of the illusion of
language. Saussure exclaimed in his unpublished notes that the illusion of
things naturally given in language is profound. It certainly is. The source
of this profound illusion can be explained in a linguistics built on the
new foundations (22.9).

Exercise (type 2). For a selected approach to linguistics, identify the
explicit assumptions put forth on which it is based.

Exercise (type 3). (Requires more analytic skill) For a selected approach
identify the implicit, tacit, or hidden assumptions relied on.

The new foundations for general linguistics rest only on the standard
criteria of science and accept only the four standard assumptions of
science. Thus linguistics built on the new foundations is aligned with the
physical and biological sciences in this regard and becomes a genuine
natural science in its own right.


5. Why did I not discuss current attempts?

The book does not try, however, to build a new body of linguistic theory on
the new foundations as the reviewer seems to have hoped. I appreciate the
implied complement, but that is way beyond my capabilities. I have refused
in the book to guess at linguistic structures without carefully obtained
evidence. I try to follow Galileo's admonition that in the natural sciences
"one must take care not to place oneself in the defense of error" (Drake
1953:53-4).

To build such theory on the new foundations is a job for a number of clever
linguists who are similarly committed to scientific truth. In this task, we
certainly will want to make use of the many insights already won by
linguistics. However, such insights will all have to be reexamined, recast,
and carefully tested against the evidence so as to exclude those
traditional elements they all contain that make them at present unfit for
inclusion in a proper science. This is not unlike the situation linguistics
has already faced in studying indigenous languages. Any existing earlier
descriptions would be relevant to the task even though cast in Latinate
grammatical categories and overlooking important phenomena. Although
useful, they would have to be completely reexamined and reworked. The point
is, we don't have to start from scratch but no existing result can be
included automatically: we must be extremely careful. This is not exactly
"sweeping away ... the entire linguistic research of the last 2000 years."
It's rebuilding linguistics on new and more secure foundations.

This rebuilding will have to be carried out by a number of linguists drawn
from all the subdisciplines of general linguistics and some neighboring
disciplines as well. In this the new scientifically justified notation,
which is applicable across this complete area, will be invaluable. Since it
refers to a real physical reality that can be studied from various points
of view, it will allow scientifically verified findings from all these
subdisciplines to be combined into a coherent whole. It will allow gains
made in one subdiscipline to contribute to other subdisciplines. This
should immeasurably improve the efficiency with which linguistics can
conquer its territory with scientifically established results. Thus the new
notation will not become like just another trademark of yet another
approach to language; it will belong to everybody.

We can learn from the chemists who have long had a standard notation also
referring to physical reality. It has allowed discoveries of new chemical
compounds and new chemical reactions to be communicated to other chemists
in other areas of chemistry and other parts of the world and applied
immediately in their own research. It has allowed chemists everywhere to
cooperate in the common task of building a coherent unified and
scientifically testable view of nature from the chemical point of view.

Exercise (type 4). For a selected approach, what are its major findings or
insights we should consider as worthy candidates for preservation?

Exercise (type 5). To what extent do these insights, or the way in which
they have been stated, simply reflect the inappropriate criteria and
assumptions relied on, and to what extent might they reflect actual truths
about nature?

Exercise (type 6). Can any of these truths be simply moved over onto the
new foundations?

Exercise (type 7). For those that cannot simply be moved over, what
additional information about plex structure would be needed to accommodate
them?

Exercise (type 8). What observational or experimental evidence would be
needed to be able to develop plex structures preserving these truths?


6. Points not well covered in the review.

In a short review it is impossible to do justice to all the material in the
book. Readers of the review may thus not be in a position to understand or
appreciate some of the most important features and advantages of the new
foundations that accrue from moving to a study of the real world.

The new foundations offer two separate domains of theory, one focused on
individuals, the other on groups, and a theory of their interrelation. The
tradition, however, has provided us with only one, grammar, and it has
never been clear how the same theory can simultaneously cover individual
phenomena and group phenomena, which are clearly different.

With these two domains of theory (1) variation in the individual and
variation in the community can be separately accommodated; (2) theory
becomes available for treating the linguistic development of the child and
the adult in the context of the family and the community in which they find
themselves; (3) the way is open to explaining phenomena at the group level
as flowing in part from phenomena at the individual level and phenomena at
the individual level as flowing in part from phenomena at the group level.
The way is open to develop improved theory (4) for treating communicative
interactions; (5) for treating the embedding of communicative tasks in the
noncommunicative tasks they serve to coordinate; (6) for treating the
individual and social dimensions of ceremonies and rituals; and (7) for
developing improved theory for historical linguistics.

The natural sciences are not stand-alone and autonomous as linguistics has
been. They enjoy a unity in their interconnections, an important one of
which is called scientific reduction. (See 19.6.) The combining properties
and chemical reactions of various substances were first studied and
explained in chemistry on the basis of postulated atoms, molecules, and
chemical bonds having different properties and strengths. Later, atomic and
molecular physics explained the nature of the chemical bond through the
quantum theory and it has become possible to calculate the configurations
and chemical properties of molecules in terms of elementary particles
obeying the laws of physics. Similarly, the biological nature and operation
of organisms and their parts, including the workings of biological
inheritance, are now being explained in terms of lower-level theories in
chemistry and physics. The relationship between psychology and biology is
now under intensive investigation--it has been thought for a long time that
explanations of psychological phenomena are to be found in the
physiological workings of the brain.

It has also been thought for a long time that linguistics stands between
the individual psychological sciences and the social sciences, but a
linguistics rendered autonomous and separate from the rest of science
through its focus on language and grammar has been unable to fill this
important role. A linguistics build on the new foundations can. Its theory
focused on the individual person as an object of nature is solidly among
the individual psychological sciences. Its theory of groups is firmly among
the social sciences. Its detailed treatment of the relation between
individual theory and group theory can be seen to provide a reductive link
between the social sciences and the individual psychological sciences. This
allows linguistics to finally fulfill its natural function and connect the
social sciences through an unbroken reductive hierarchy all the way down to
the physical sciences. I think this may be one of the most important
contribution that linguistics can make to science in general, and to the
social sciences in particular by helping them to improve their scientific
integrity. It will allow us to erase the troublesome distinction between
soft and hard sciences and it will complete the long-sought unity of
science. Still another reward for moving to a study of the real world.

It has long been understood by linguists that people are all different, a
fact that could not be captured adequately in grammatical theories, which
have focused on ideals of uniformity or ideals of perfection. It is these
very observed differences in people and their similarities that the new
theory is built on. In the new framework we can capture the unity in human
diversity. We can exhibit the insight that we can have a science while yet
recognizing the uniqueness of the individual and the particularity of
situations honored in the humanities. Thus a linguistics on the new
foundations may not only better serve the humanities but may be able to
bridge what has been perceived as a wide intellectual gap, another perhaps
surprising reward for moving into the real world.

Exercise (type 9). What additional possible investigations of plex
structure or relations to other disciplines come to mind, not necessarily
related to the findings of one particular existing approach?


7. Other questions raised

Let me correct a few points in the review. In the second paragraph of 1.1
the review says that I take human communicative behavior to be "a given
entity in the real world, as required by science." This is a bit confused.
The new framework is not a version of behaviorism. What is assumed to exist
in the real world under the first standard assumption of science is the
people who communicate and their communicatively relevant surroundings
including the physical energy of sound and other energy flow associated
with their communicative behavior.

In the same paragraph, it is important to understand that it is not only
the sound waves that we can observe but also the other means of energy flow
and the surroundings and importantly the people who communicate, including
their noncommunicative behavior. The new foundations can in fact make use
of such "nonlinguistic" evidence. For example, in a cooperative task of
hanging a picture, if one person says "a little higher" it counts as
evidence that the other person moves the picture a bit higher on the wall,
and this can in fact be reflected in the notation if desired.

In the middle paragraph of 1.2, the review is not correct in stating that
it is an assumption of the theory that properties do not change without
reason. We must investigate the causes of stability as well as the causes
of change. See section 12.5 in the book. Setting procedures are set up on
the basis of three empirically supported general laws of communicative
behavior. They do not rest on any additional assumptions.

The reviewer's likening of Chomsky's theory of linguistics to Newton's
theory of astronomy is specious. Chomsky does not adhere to the standard
criteria and assumptions of science, Newton did. I see no reason to grant
the status of a science to any of Chomsky's approaches to linguistics.
Models in science are models of something in the real world, the solar
system in the case of Newton's theory of astronomy, and they can be tested
against observations of the real world. Chomsky's models are not and
cannot. They are abstract models of assumed objects, also sometimes
conceived as being abstractions. To see them as models of people is a
fallacy and a blatant category mistake. See Yngve (1986) for further
discussion of these issues that I did not feel obliged to repeat in the
book under review. If the reviewer had read the 1986 book he might not have
been so ready to fault me for not including more discussion of these issues
that seem to be of particular professional interest to him.

The reviewer asks in his conclusion: "Assuming, however, that several
different frameworks, approaching linguistics from different perspectives,
may each be legitimate, the question remains as to how valuable Yngve's
theory and framework would be in competition with existing ones."

In the first place, one can't just blithely assume this. If one wishes to
claim that several different approaches to linguistics may each be
legitimate, each must adhere only to the standard criteria of science and
accept only the standard assumptions of science. None of the existing
frameworks that make assumptions about language or the objects of language
can qualify. From this point of view none is legitimate. The new
foundations provide the only framework that does meet these requirements
for legitimacy as a science. None of the claims of existing theories of
language or grammar can be accepted into science as they stand. This is not
an a priori rejection of these claims, as the reviewer appears to think,
but a rejection on principle. No matter how insightful they may be, these
claims have not been shown to have any scientific validity. That's why work
on the suggested exercises is so important to the advancement of
linguistics.

Nevertheless, some linguists may wish to compare particular approaches to
linguistics with the new foundations in greater detail. To assist in such
projects I have prepared a short article identifying seventeen points of
comparison giving chapter and section citations to the book for each. This
article, presented at the 1997 LACUS conference, has been preprinted and
placed on the web for your convenience:

http://humanities.uchicago.edu/humanities/linguistics/faculty/yngve.html

Exercise (type 10). (Requires considerable knowledge of a selected
approach) How does the selected approach stack up against the new
foundations in light of the seventeen points?


References

Drake, Stillman. 1953. tr. Galileo Galilei. Dialogue Concerning the Two
Chief World Systems--Ptolemaic & Copernican. Berkeley and Los Angeles:
University of California Press.

Feynman, Richard P., Robert B. Leighton, and Matthew Sands. 1963. The
Feynman Lectures on Physics. Reading, MA: Addison-Wesley.

Yngve, Victor H. 1986. Linguistics as a Science. Bloomington &
Indianapolis: Indiana University Press.

- -------------------------------------------------------
Victor H. Yngve Professor Emeritus,
 mailing address: Linguistics & Psychology
28 Crest Drive Dune Acres Department of Linguistics
Chesterton IN 46304 University of Chicago
phone: (219) 787-8340 1010 East 59th Street
e-mail: v-yngveuchicago.edu Chicago, IL 60637