[ The Sokal Affair | Searching | Background Material | Guestbook | Recent Additions ]

[ Top : Articles : "THE SOKAL-WEINBERG THEORY OF READING" : Other Articles ]


Gabriel Stolzenberg



   1.  DOUBLE TALK.  An unpublished letter to the New York
       Review of Books about Steven Weinberg's "Sokal's hoax."

       STS posting about creative readings of a thirty year
       old innocuous and deeply insignificant, off the cuff
       remark by Jacques Derrida.

   3.  GIVE US THE MAN & OTHER COMMENTS.  Comments to Steven
       Weinberg about his replies to criticism of "Sokal's
       hoax" by George Levine and M. Norton Wise.  This is
       primarily about Weinberg's role in the invention of
       Derrida the physics faker.

                         0. Introduction.

                        A Cautionary Tale

   In the early 50's at Yale, Arthur Herschman, a physics graduate
student rooming with Harold Bloom--the same Harold Bloom who later
argued brilliantly for the importance of creative misreading, took
his German exam.  Relying more on courage than knowledge, Herschman
soon satisfied himself that the text to be translated into English
was about how a submarine works.  After this, everything fell into
place--no matter that it happened to be about a Wilson cloud chamber.

   Caution.  This happens within a language too, and when it does,
it often is less amusing.  People have strong feelings about being
understood correctly.

              The "Only One Reading" Theory of Reading

   The pieces I present here are about bad reading--by Alan Sokal,
Steven Weinberg and others--of social constructivists and Jacques
Derrida, one of the world's great readers.  (Yes, there are great
readers, just as there are great physicists.)

   Unlike Arthur Herschman's delightful transformation of a cloud
chamber into a submarine, these bad readings make the authors look
ridiculous and are accompanied by expressions of scorn.  In short,
these are bad readings with an attitude--one that says,

   "If it strikes me as nonsense, then it is and that's the end
    of it.  If it strikes me as being confused, contradictory or
    illogical, then it is.  Furthermore, this has nothing--much
    less everything--to do with how I am reading it."

   Notice the denial of an appearance/reality distinction.  Here,
being obvious--that most subjective of grounds for accepting that
something is the case--trumps all other considerations.  In other
words, everything that seems obvious is true.  Can anyone imagine
science, or mathematics, being conducted this way?

   A way to make sense of this practice is to see it as one among
many that follow from what I shall call here the "Only One Reading"
theory of reading.  According to this theory, yes, the meaning and
intentions of the author are paramount.  But if we find the meaning
plain (it's about how a submarine works), there is neither need nor
reason to consult the author--not even if we find the text confused,
illogical or nonsensical.  Why?  Because there is only one reading,
only one meaning.  And we are in possession of one.  So it must be

   Does the author disagree with the reading?  Of course, she does.
She's so confused that she thinks a submarine is a cloud chamber.  Or
if she does understand, then she is unwilling to admit that she has
been caught in a contradiction.  But couldn't she possibly be right
and we wrong?  Are you kidding?  Look, she says right here in black
and white, blah blah blah.

   Some readers agree with the author.  Them?  They're just as bad as
she is!  Those people have no respect for the truth.  Anyway, they're
so blinded by their ideology that they can't see what is staring them
in the face.  Etc., etc., etc..

   Now, any of these things could be true.  However, they are not true
in any interesting sense merely by virtue of being asserted.

   (For an example of another bad-reading practice that can be made
sense of in this way, see Paul Boghossian's mistreatment of a
remark that the archeologist, Roger Anyon, made to a reporter for the
New York Times.  This is in Boghossian's, "What the Sokal hoax ought
to teach us," TLS, December 13, 1996.)

                          Program Notes

   "Double Talk" is self-contained.  Its modest aim is to explain,
in 800 words or less, why every assertion that Sokal and Weinberg
take to be a refutation of social constructivism--even those they
haven't thought of yet--instead reveals, always in the same boring
way, the confusion that infects their thought.

   Consider, for example, Sokal's context-sensitive assertion,

        "Deny that non-context-dependent assertions can
         be true and you throw out the Nazi gas chambers,"

in the afterword to the hoax.  It is supposed to show the absurdity
of Andrew Ross's culturalist claim that there is

        "[a] stand-off between the empiricist's claim that
         non-context dependent beliefs exist and that they
         are true, and the culturalist's claim that beliefs
         are only socially accepted as true."

   However, for Ross's culturalist, the standoff is exemplified by
the fact that Sokal does not find his own assertion absurd.  There
is a standoff because each position is irrefutable from within.  (I
am over-simplifying here.)  This is but one of the indefinitely many
examples that can be analyzed by the simple method I describe in my
letter.  Try it yourself on this one.  A social constructivist parody
of Sokal would be hard-pressed to improve on Sokal himself trying to
be clever about the nature of reality.

   The story of "Inventing Jacques Derrida" and "Give Us the Man"
begins in 1966 at a question and answer session following a lecture
by Derrida that had nothing to do with science but everything to do
with his notion of a center for a structure.  Derrida was asked if
the speed of light is a center for special relativity (qua structure).
He replied that it is not, and could not be, a center for special
relativity.  He then elaborated a little.  The reply required almost
no physics.

   Matters rested there for a quarter of a century, until Ernest Gallo
creatively read Derrida's answer to be that the world of physics is not
governed by rules!  His article, "Nature faking in the humanities," was
published in the summmer 1991 issue of the not very Skeptical Inquirer.

   Then Gallo begat Gross and Levitt--whose Derrida says that c is
not constant, who begat Sokal, who begat Weinberg, who, when taken
to task by George Levine for forgetting that he needed an argument,
made the terrible mistake of trying to prove that Derrida was faking,
by an examination of the text from which Derrida's remark was taken.
Cloud chambers.  Submarines.  Why not leave the physics to Weinberg
and the interpretation of texts to Derrida?

                         1.  DOUBLE TALK

       [This was written as a letter to the New York Review of
        Books in response to Steven Weinberg's "Sokal's hoax."]

   Steven Weinberg [NYR, August 8] and Alan Sokal let their two-fold
use of expressions like "discover" and "law of physics" fool them into
finding obvious something for which their evidence is bogus.  Social
constructivists say that a law of physics is a social construction.
Some of them also say that a law of physics is an impersonal truth
about objective reality.  To Weinberg and Sokal, these two assertions
are obviously inconsistent.  This would indeed be so if the two uses
of "law of physics" were the same.  But they are not.  Weinberg and
Sokal, like almost everyone else, use "law of physics" to refer both
to an objective truth and to a statement that we accept as such because
we have compelling scientific reasons to do so.

   This two-fold use is normally harmless, even though, on Sokal and
Weinberg's conception of the world, only the first is correct.  The
second one is more common because, although we can never know that a
statement is a true law of physics, we often have compelling scientific
reasons to accept one as such.  When we do, physicists talk about "the
discovery of a physical law" and social constructivists talk about the
"the social construction of a physical law."  "Discovery" and "social
construction" are metaphors for a complex process that I prefer to call
"acceptance."  The upshot is that it is not inconsistent to speak of a
physical law as a social construction (provided it is the kind of thing
that physicists talk about discovering) and to affirm with Sokal and
Weinberg that a genuine law of physics (the kind we can never know we
are looking at) is an impersonal truth about objective reality.

   I will illustrate the point with two examples.  The first is from
"Sokal's Hoax."

      "The choice of scientific question and the method of
       approach may depend on all sorts of extrascientific
       influences, but the correct answer when we find it
       is what it is because that is the way the world is."

   Here Weinberg is confusing himself by his own way of talking.
What we accept as a "correct" answer to a scientific question may
well be mistaken.  So we cannot claim that it truly captures "the
way the world is."  "Finding the correct answer" is a metaphor for
-accepting- some statement as the answer--for what we believe are
compelling scientific reasons.  Can the acceptance sometimes depend
on extrascientific influences?  I hope not, but who knows?  On the
face of it, this is an empirical question, not a logical one.

   The second example is from Sokal's "A Physicist Experiments With
Cultural Studies."

     "Fair enough: anyone who believes that the laws of physics are
      mere social conventions is invited to try transgressing those
      conventions from the windows of my apartment.  (I live on the
      twenty-first floor.)"

   "Mere social conventions" is Sokal taking liberties with "social
constructions."  ("Transgressing those constructions" isn't funny.)
His "laws of physics" are plainly statements that physicists accept
as such for compelling scientific reasons--which is also what the
social constructivists are talking about.  Hence, Sokal is inviting
them to reject something that they have compelling reasons to accept.
Why would they do that?  There also are excellent reasons not to jump
that have nothing to do with knowledge of laws of physics: observations
and reports of observations.  Furthermore, it is not implausible that
a fear of jumping from heights was selected for in evolution, perhaps
before the existence of homo sapiens.

   Readers who need more convincing are invited to apply the same
analysis to the many other examples in Weinberg's "Sokal's Hoax" and
Sokal's "A Physicist Experiments With Cultural Studies."  (According
to Sokal, an analysis of this kind downplays the practical relevance
of objective realities.  Readers are invited to apply the analysis to
downplay this assertion too.)

   It is understandable that Weinberg sees social constructivists
as relativists.  They often study science as belief.  To some, this
falsifies science by ignoring its most important feature--that it is
true.  Unsurprisingly, some religious communities have made the same
objection when objectively minded scholars (some of them believers!)
have studied their beliefs as beliefs.  Distortion is one concern, but
there may also be a fear that if we think too much about our beliefs
-as- beliefs, it will weaken our confidence in their truth.

   Finally, I am not the only one to have remarked that the naive
attempts of Weinberg and Sokal to refute the metaphysical position
they attribute to the social constructivists have gotten them into an
embarrassing situation.  These questions have been discussed since
antiquity, with great philosophers weighing in on different sides.
(There are more than two sides.)  Now is an especially active time.
Subtle issues have been brought to light.  (Weinberg almost stumbles
on one when he says that he has to assume that there is a one-to-one
correspondence between statements and aspects of reality.)  But there
is no recognition of any of this in Sokal's rhetoric or Weinberg's
arguments, which work equally well against Sandra Harding and Immanuel
Kant.  Weinberg and Sokal write as if our everyday, unreflective conception
of external reality suffices to settle these issues.  If they really
believe this, perhaps they should go back to school.  If they do not,
they should give us a sign.

                                           Gabriel Stolzenberg
                                           Cambridge, Massachusetts


       [This is a revised version of a posting to the STS
        discussion group in December 1996.  References to
        to other STS postings are from the same period.]

      I.    The Silly Quote.
      II.   Val and Roger Set Good Examples, Alan Does Not.
      III.  Clearly Norman Levitt.

                   I.  The Silly Quote

   In his posting about Roger Hart's, The Flight From Reason,
Val Dusek refers to "the famous silly quote from Derrida."  And
in his reply, Alan Sokal remarks that "in truth, it's nonsense."

   A colleague said the same thing to me, nearly two years ago.
He said, "Levitt is right."  I said, "What do you mean, Levitt
is right?"  He said, "It's nonsense."  I said, "Yes, to you it
is nonsense, because you don't understand it."  He said, "How
long would it take me to understand it?"  I thought about it a
moment and replied, "It's hard to say.  Derrida is a difficult
read.  So if you want to really understand it, it could take a
long time."  He said, "That can't be."

   A little later, with the text open in front of us, I asked my
colleague if he was saying that, in 1970, Jacques Derrida did not
believe that the speed of light is constant.  I put it this way
because it's what Gross and Levitt seemed to be saying.  He said,
"Yes."  I said to myself, "I can't believe this is happening."

   The next day, to force the matter, I proposed a $2000 bet.  The
bet was not accepted but I was invited to explain what the passage
is about.  Which I did, relative to the concept of a center for a
structure.  As I read it, Derrida was asked if the speed of light
is a center for special relativity.  He replied that it is not, that
it and a center are very different -kinds- of things, that the sense
in which a center is invariant is different from the sense in which
a physical constant is invariant.  And he elaborated on this a bit.

   How much physics is needed to know that c is not a center?  None.
As I read the text, it is enough to remark that a center is outside
the structure for which it is a center.

        II.  Val and Roger Set Good Examples, Alan Does Not

   In a recent posting about related matters (Gross, Levitt, Derrida
et al, Thu 19 Dec), Val Dusek showed that recanting can be done with
class.  I was going to invite him to go a little further, and recant
"the famous silly quote," but he beat me to it in a message I received
this afternoon.

   For another classy performance, here is Roger Hart, in a recent
message to me, reconsidering his statement in The Flight From Reason
that Gross & Levitt identified an "error of scientific fact" in the
Derrida passage.

You are, of course, absolutely correct that assertions that Derrida's
statement is "an error of scientific fact" or "nonsense" require an
argument: the burden of proof must be on the person who makes any such
assertion; it is inadequate simply to claim that one does not understand
Derrida.  You have convinced me that these assertions have been made by
dismissing Derrida without offering any argument whatsoever, and shown that
I too was guilty of dismissing Derrida's statement without evaluating
whether or not it might be correct under a charitable interpretation.

   In the same message, Roger also makes the following observation about
the passage in question.

On the other hand (but perhaps reasonably enough since he was responding to
a question in a discussion) Derrida did not offer much explanation of his
statement or argument for it (at least as I remember the passage).  I
continue to think that Derrida does not understand enough relativity theory
to make the kind of deconstructive criticisms of it as a system that he
offers for Levi-Strauss's strutural anthropology.  So in this sense I remain

   I quote this here to point out the radical difference between this
critical remark and Alan Sokal's "in truth, it's nonsense"--or Sokal's
"Who knows -what- it means?" on National Public Radio.  Hart engages
the text, possibly to refute it.  By contrast, Sokal believes that we
can tell that the passage is nonsense without any reflection, without
inquiring into the meaning of presumptively technical terms that appear
in it.  (Terms like "center" and the second appearance of "constant.")

   My evidence for this is again Sokal's performance on NPR.  As I
heard it, Sokal introduced the passage as a choice piece of nonsense,
and then presented a proof of this.  The proof consisted in reading
the passage aloud.  If I am mistaken about this--I think Sokal is in
the best position to know--I'm happy to stand corrected.  But I think
it will take a heck of an explanation.

   As for the content of Hart's statement, I would expect Derrida to
agree with it.  However, I do not see any deconstructionist criticism
in the passage.  So, in -this- sense, I too remain agnostic.

                 III.  Clearly Norman Levitt

   When Higher Superstition was pressed upon me--"You should read
this, it's important"--it was the Derrida passage that was offered
as a choice example of the good things in it.  (Exactly as in the
NPR interview with Alan Sokal.)  My response, and that of someone
else who read the passage, was, "How am I supposed to know what this
means?  I need to see the full text."

   Nine months later, I inquired about this and was told, "Oh yes,
I spoke to Norm about it.  He says he has 50 examples that are as
bad or worse."  (This was in early 1995.  Has anyone seen them?)
I explained that I didn't want 50 more examples, I wanted the text
for this one.  So I wrote to Levitt asking for the reference.  I
explained that I was not persuaded by his reading of the passage
and wanted to form my own opinion about it.  In his reply, Levitt
asserted that Derrida -clearly- was faking and should have passed
on the question to which the passage is part of the answer.

   One may have to be a mathematician to fully appreciate the irony
in Levitt's use of "clearly."  In math talk, it too often means "I
hope nobody asks me about this because I have no argument, just a
feeling."  In other words, it too often means "I'm faking it here,
so go easy on me."

                3.  GIVE US THE MAN & OTHER COMMENTS

   [These comments have more bite if they are read together with the
relevant parts of Weinberg's replies to George Levine and M. Norton
Wise in the October 3, 1996 New York Review of Books.  A few changes
have been made, for the sake of clarity and to respect privacy.]

Dear Professor Weinberg,

   After reading your reply to replies, I have three more comments.
The first is long, but the other two are not.

                       10.  Give Us the Man.

   I confess that I found your "explication de texte" appalling, but
it was also revealing.

   a)  There is a wonderful naivete in thinking that a layman talking
to a layman about physics should use only expressions that a physicist
meets in his work.

   b)   You find the notion of a center obscure.  Yet you believe you
understand it well enough to say, "I did not see how [a number] could
be the `center' of anything."  So, apparently you do not find it -too-
obscure to have opinions about it.  It would have been fascinating to
hear your account of where in the text you learned enough about what a
center is to be able to have the thought, "I did not see how a number
could be one."  But I suppose some things will always remain mysteries.

   c)  When you make your "I did not see how" statement above, are
you counting on your readers not to notice that you are agreeing with
Derrida?  (I'm not the only one who did notice it.  It's pretty bald.)
Derrida tells Hyppolite that the constant that he's talking about is
-not- a center.  Your position on this should be, "Right, and I'll bet
it's simply because it's a number."

   d)  You make the wild claim that the 1st sentence of a 29 sentence
discussion about the meaning of "center" is Derrida's explanation of
it.  Even though, immediately after presenting this 1st sentence, you
comment, "This was not much help."   Why do you do this?  Does Derrida
say, "The 1st sentence is my explanation"?  Of course not.  Moreover,
it doesn't -look- like an explanation.  So again, why do you do this?
It's outrageous.  Would you dare to do it if you knew the readers had
the text in front of them?  I doubt it.

   e)  And then, astonishingly, you say, "Lest the reader think that I
am quoting out of context," when -of course- you are doing just that!
(But maybe this is a case of bad writing, so I won't make any more of
it here.  You probably meant to say something like, "Lest the reader
think the problem is that I am quoting out of context....")

   f)  Yes, Hyppolite asked Derrida to clarify the notion of a center.
But if you read carefully the full text of his question, you will see
that Hyppolite does have some understanding of the notion of a center.
He uses it confidently in several sentences, and at the end of one of
them, he asks, "Is this what you wanted to say, or were you getting at
something else?"

   g)  Here you give away the store.  "It was Hyppolite who introduced
`the Einsteinian constant' into the discussion."  This is false, as you
can see for yourself.  However, what matters is not that you misremembered,
but why.  (I also misremembered, until I checked the text again, just to
make sure.)  Why did you think it was Hyppolite?  Here is my hypothesis,
which I find compelling.

   What -does- Hyppolite say?  He says, "With Einstein,..., we see a
constant appear, a constant which is a combination of space-time."  In
this context, it is utterly natural for Derrida, in his answer, to refer
to the constant Hyppolite is talking about as "the Einsteinian constant."
And it is likewise utterly natural for someone like you or me to forget
that Derrida's expression is only a paraphrase of what Hyppolite said--
because it is such a good paraphrase.

   The point is that if you had traced how "the Einsteinian constant"
entered the discussion, you would have seen what I just pointed out,
and would not have been bothered by laymen using an expression "which
I had never met in my work as a physicist."  (You shouldn't have been
bothered anyway, but this would have clinched it.)  Instead, you would
have grasped that Derrida used "the Einsteinian constant" simply as a
convenient expression for referring to the constant that Hyppolite was
talking about when he said, "With Einstein...we see a constant appear."

   h)  Jean Hyppolite says that he admired Derrida's presentation and
discussion.  Why then is it "poor" Hyppolite?  You don't know what he

   i)  What then is this Einsteinian constant?  Well, here is a report
on 12 answers to this question.  I did not solicit any of them.

   Two physicists, you and Sidney Coleman, and also my brother-in-law,
a molecular biologist, said that they don't know.  Like you, Sidney
noted that it is not an expression used by physicists.

   The other 9 people all said that it's the speed of light.  Who are
they?  There's Ernest Gallo, in "Nature Faking in the Humanities," the
first one to lift the passage out of context and ridicule it.  (Except
that Gallo's has 3 extra sentences.)  Then there are Gross and Levitt,
in "Higher Superstition."  There is also a colleague, who I offered to
bet ,000 that in 1970 Jacques Derrida--not his cousin, Bernard--knew
that the speed of light is constant.  He said that Derrida did not know
this, taking the passage to be evidence of it.   Then there is Susan
Carey, a cognitive psychologist, who described it this way.  "First I
thought, `the Einsteinian constant, I don't know what that is.'  And
then, `The speed of light!'"  Barry Mazur's account was similar.  To
my wife, Nancy Nancy Kopell, an applied (and also pure) mathematician,
it just was the natural answer.  Which is how I feel too.  As does Mark
Bridger, a mathematician who asked to be added to the list when I told
him about it yesterday.

   The point is that it seems to be natural for educated laymen to
take "the Einsteinian constant" to mean the speed of light.  And if so,
it's reasonable to suppose that this too is what Hyppolite and Derrida

   (By the way, I discovered yesterday that the talk was held not in
1970 but in 1966.  Thirty years ago!)

   j)  "Derrida just started talking about the Einsteinian constant,
without letting on that (as seems evident) he had no idea of what
Hyppolite was talking about."  This is outrageous.  As seems evident!
It doesn't get any more subjective than this.  You make this claim
without offering an iota of evidence.  No doubt this is because you
have none.

   k)  Finally, if you don't find my readings plausible, do remember
that it is you who are in the soup, not me.  If you don't yet see how
ridiculous it looks for the bunch of you (Gallo, Gross, Levitt, Sokal
and now you) to try to use these few sentences (from an answer to a
question following a talk given thirty years ago) to expose Derrida as
a physics babbler, maybe these remarks will serve as a wake up call.

   A friend of mine was disposed to believe the charge about Derrida.
But when she looked at the text, and found that you are making this
Big Deal about an unclear answer to an unclear question at the end of
a talk that had -nothing- to do with physics, her reaction was (this
is not verbatim but it's faithful to the spirit of it), "That's it??
This is all they have on him??"

   Finally, this lack of balance between your charge and your evidence
reminds me of a conversation between Alik Volpin and Vladimir Bukovsky,
two great Soviet dissidents, back in the USSR.  Alik was talking about
the the importance of obeying the laws, to which Bukovsky replied, "But
doesn't the KGB always say, give us the man and we will find the charge?"

          11.  I knew Tom Kuhn.  Tom Kuhn was a friend of mine.

   Invoking your friendship with Tom (who had almost as many friends as
Imre Lakatos) works against you, because if you really were a friend of
Tom, as I'm sure you were, how did you fail to take in the full picture?

   Tom was conflicted.  He was passionately committed to objectivity, but
never found an adequate language for talking and thinking about it.  As I
see it, the main difference between you and Tom is that this bothered Tom.

                       12.  Avoidance 101.

   Finally, I was impressed by the way you avoided replying to a remark
by M. Norton Wise that is closely related to the main point I made in my
letter to the NYR.

   Wise noted that if your position is "If it works, it's real," then
every anti-realist is a realist.  What Wise said next (when he turned
to a different subject) inspired you to require also that there be lack
of multiplicity.  But this changes nothing.  Wise's point holds so long
as the criteria for "It's real" are observable--and we can be as generous
as we like about what counts as an observation of something.  We observe
lack of multiplicity by reading a proof of it.  Right?

   So, as Wise observed, on your account of things, every anti-realist
is a realist.  However, I don't think this will bother you because you
will take the attitude that, because you don't mean it to be so, it is
not.  Nevertheless, whatever rationalization you choose, in fact, you
are in exactly the same kind of pickle that you say Sandra Harding is in.

   Sincerely yours,

   Gabriel Stolzenberg

Gabriel Stolzenberg is a Professor of Mathematics at Northeastern University. He has an A.B. from Columbia, a doctorate from MIT, and an honorary master's degree from Brown. He has been a fellow of the Alfred P. Sloan Foundation and the John Simon Guggenheim Foundation. He spent one year at the Institut des Hautes Etudes Scientifiques in France. He has been a Visiting Professor at the University of Paris, Orsay, the University of Caifornia at San Diego and MIT.

His best known mathematical work is "Volumes, Limits and Extensions of Analytic Varieties" (Lecture Notes in Mathematics, No. 19, Springer 1966). His best known works about mathematics are his critical review of Errett Bishop's, "Foundations of Constructive Analysis" (Bulletin of the American Mathematical Society, March 1970) and "Can an Inquiry into the Foundations of Mathematics Tell Us Anything Interesting About Mind?" (in Psychology and Biology of Language and Thought: Essays in Honor of Eric Lenneberg, edited by Eliza Lenneberg and George A. Miller, Academic Press, 1978).

His non-mathematical interests include metaphysics, belief formation, interpretation of texts and legal theory. His main research has been on constructivist mathematics and the classical/constructivist gestalt shift.

[ The Sokal Affair | Searching | Background Material | Guestbook | The Top of this Article ]
Last Modified: 24 November, 1997