A Data Point: Sociologists Smarter than Physicist...

Cosma Shalizi, sociologist, has a truly remarkable intellectual encounter with a... with an I'm-not-sure:

2006 08: On the Superiority of Sociology to String Theory: One of my friends in graduate school had an adviser with a gift for memorable expressions, at least for a theoretical physicist, especially when dismayed by some stupidity. Two of his put-downs which stuck with me were "I could go crazy tomorrow and find an appointment in the sociology department", and "I don't want to criticize you, but this is the way superstring people think".

I was never sure which was supposed to be worse, but now I know. Sociologists have many faults, but they do know better than to try explaining a variable with a constant, while string theorists evidently do not. (Via CapitalistImperialistPig, who has better things to write about.)

The fact that Prof. Motl reasons so badly here that he'd fail my freshman stats class is, of course, infinitely less offensive than fact that he's a bigot (of the "we must squarely face the harsh light of my pseudo-scientific prejudices" variety). But I can't help feeling -- hoping, even -- that the two sorts of idiocy are linked.

Comments

[Gonerill]

That would be Cosma Shalizi, statistical physicist. Lubos, on the other hand, is the cringing embarrassment of the Harvard Physics department.

[Aaron Bergman]

Please don't think of Lubos as representative of string theorists.

[kid bitzer]

goneril--just for the record:

is it Lubos who cringes, or the department?

And if the second, wouldn't that make him, e.g., the "cringe-inducing" embarrassment etc.?

[Gonerill]

The latter, and yes. I don't imagine Lubos cringes very often. With him, it's more whinge than cringe.

[Robin]

I know that old Lubos if of the post-Soviet keep the Manicheanism, switch the labels type, but does his prose style remind anyone else of Lenin?

[CapitalistImperialistPig]

Aaron - I don't want to think of Lubos as the paradigmatic string theorist, but very few ST types have been willing to challenge him or distance themselves from his extremism, and Leonard Susskind expresses views (on non-stringy physics) rather similar to those of Lubos.

[anne]

http://www.nytimes.com/2006/01/15/books/review/15powell.html?ex=1294981200&en=26d7d18e0f334fc9&ei=5090&partner=rssuserland&emc=rss

January 15, 2006

Across the Megaverse
By COREY S. POWELL

Physicists are not like ordinary people, and string theorists are not like ordinary physicists. Even compared with their peers, crafters of the arcane model of reality that is string theory think in terms of sweeping explanations of nature's design. Leonard Susskind, a founder of the theory and one of its leading practitioners, brazenly lays out this no-boundaries attitude on the first page of his new book. His research, he declares, "touches not only on current paradigm shifts in physics and cosmology, but also on the profound cultural questions that are rocking our social and political landscape: can science explain the extraordinary fact that the universe appears to be uncannily, nay, spectacularly, well designed for our own existence?"

What troubles Susskind is an intelligent design argument considerably more vexing than the anti-evolution grumblings recently on trial in Dover, Pa. Biologists can point to unambiguous evidence that evolution truly does happen and that it can account for many otherwise inexplicable aspects of how organisms function. For those who take a more cosmic perspective, however, the appearance of design is not so simply refuted. If gravity were slightly stronger than it is, for instance, stars would burn out quickly and collapse into black holes; if gravity were a touch weaker, stars would never have formed in the first place. The same holds true for pretty much every fundamental property of the forces and particles that make up the universe. Change any one of them and life would not be possible. To the creationist, this cosmic comity is evidence of the glory of God. To the scientist, it is an embarrassing reminder of our ignorance about the origin of physical law.

Until recently, most physicists took it on faith that as they refined their theories and upgraded their experiments they would eventually expose a set of underlying rules requiring the universe to be this way and this way only. In "A Brief History of Time," Stephen Hawking recalled Albert Einstein's question "How much choice did God have in constructing the universe?" before replying that, judging from the latest ideas in physics, God "had no freedom at all." Like many leading physicists at the time, Hawking believed that scientists were closing in on nature's essential rules - the ones that even God must obey - and that string theory was leading them on a likely path to enlightenment....

[hack]

Lubos is sort of the canary in the coal mine, showing us the kind of mental illnesses that can develop when theoretical physicists are deprived of experimental data for too long. Most are stronger than Lubos, but they can only hold out for so long.

[Walt]

Isn't Cosma a physicist turned mathematician/statistician?

[Mahndisa S. Rigmaiden]

08 19 06

Cosma is a statistical physicist. There is a big branch in physics called statistical mechanics. The formalism (coupled to quantum mechanics) is so powerful that it applies to almost every system in nature, even economics. Lubos is an elitist, a racist, a sexist and a fool. One should probably not spend too much time giving him attention; you may invoke him. Ick!

[Barkley Rosser]

String theory is running into problems. Has
not delivered on the good empirically in the scattered places where it might have.

Check out competing theory of Rafalel Sorkin.

[plum]

I first encountered Lubos on a climate change board in usenet. I have never seen anybody so stubbornly misrepresent other people's views, pump up his own sense of superiority, claim 'victories' over minor matters of protocol, and trumpet the flimsiest evidence to support his standpoint while waving away mountains of evidence that contradicts that standpoint.

And all while never admitting his own failures of logic.

Do I need to spell out what position he takes on global warming? I didn't think so.

[Walt]

I'm pretty sure Cosma is a skeptic on the idea that statistical mechanics is so powerful that it applies to every system, even economics. When he was guest-posting to Crooked Timber, he suggested that econophysics (which is in large part the application of statistical mechanics to economics, especially financial markets) was doomed to failure, though he never elaborated.

[A. Zarkov]

Three-Toed Sloth or Three-Card Monte?

Which card has is the sociologist? Cosma Shalizi is a statistical physicist turned statistician focusing on high dimensional complex systems. His web page presents a picture of Leon Trotsky, as his personal photograph. (does this tell us something?) He is the author of CSSR, an algorithm for building Markov models from observed data. I worked on this problem myself, but took a somewhat different approach. He seems like a smart guy, so I’m inclined to take him seriously on matters statistical. Lubos Motl claims status as a theoretical physicist, and assistant professor of string theory. Now Shalizi attacks Motl calling him a “bigot,” and saying: “… he'd fail my freshman stats class …” Strong stuff. What did Motl do to warrant such opprobrium? He “[tried] to explain a variable with a constant …” Evidently the variable (or constant) is difference in mean IQ test scores of blacks and non-Hispanic whites expressed as a standard deviation of the scores in one of the groups (both are about equal). Motl compounds the felony by using this difference to explain the low incidence of black students at New York City’s specialized high schools.

I don’t know what Shalizi means by “variable.” Perhaps he means the “Flynn Effect” where the IQ in a few countries has increased by three points per decade during the 20th Century. This subject too invokes much controversy, and I don’t know if Shalizi thinks this effect explains the difference between ethnic groups. It would be nice if Shalizi had discussed this subject in a more dispassionate manner and avoided name-calling. I could then take him more seriously. Otherwise it seems to me that Motl’s real sin is to be discussing racial differences at all. This brings us back to Three-Card Monte. It’s a game the mark can never win because even if he picks the right card, the operator’s shills will start a ruckus and the game folds.

[Henry]

Walt, Cosma is still promising his econophysics post one of these days. When it comes, I'll probably post it to CT with his permission (it was part of the grand masterplan for his guestblogging stint).

[Cosma]

Zarkov - The variable Lubos was trying to explain was the _decline_ in the proportion of black students at elite NY high schools over the last decade. This cannot be accounted for by an _invariant_ difference in intrinsic intelligence, supposing that to exist. I've added an update to clarify this.

[stewart]

Ethnic differences in intelligence, academic performance, educational attainment are all very interesting, and everyone has a theory. So what? Not all opinions are equal, some are merely foolish despite the owner's qualifications in any other area. If you don't know the research and the data, it's just a barroom conversation, and not an enlightening one. The data indicates, for example, that black/white differences in IQ are slowly decreasing in the US (slightly faster the differences between men's and women's wages), larger than anywhere else, and that Asians attend enriched classes/further education at a much higher rate than predicted by IQ. Perhaps something's going on that doesn't have to do with IQ scores (a combination of knowledge and reasoning, as we usually forget). As an aside - I'm not aware of any countries which haven't displayed the Flynn effect, although it may be stagnating in Norway over the past few years.

[rootless]

His web page presents a picture of Leon Trotsky, as his personal photograph. (does this tell us something?)

Well, maybe, maybe not. His mention of the town of Madison Wisconsin links to a Hieronymus Bosch painting and his image of "my room" links to a picture of St. Jerome, whose Latin name, now I think of it, is "Hieronymus."

AHA!

[wcw]

Zarkov -

To spell out the critique, it is of Motl's parroting the nakedly racist 'fundamental constant of sociology' trope. For a man with a Jewish name to do so would seem richly ironic, since US Jews are the canonical object lesson. Their test scores were once a standard deviation low, too. Less than a century later, their scores are slightly higher than the median.

As for your insatiable appetite for this kind of comment, I wish our host hadn't deleted that erstwhile ad-hominem I posted last time I noticed you racebaiting. While its language was frangrant unto pungency, I do think it encapsulated accurately the fundamental deceit underlying your posts on this subject. To echo your deceitfully diplomatic language, I believe it is high time you examined your odd proclivity for posting comments defending people like Motl.

Go ahead and delete this one, too, Brad. I am comfortable watching my opinion of you drop.

[A. Zarkov]

Cosma:

Thanks very much for the clarification. There are at least three things going on. First is a change in admissions policy to a strictly test-based system, and it’s not clear exactly what year that occurred unless I missed it. Then there is a special institute to tutor minorities for the admissions test, and finally the nature of the applicant pool. I wish the Times had provided a table along with the graphs, and freshman enrollments instead of total enrollments. If they had done that we could see the effects of the policy change more clearly. (Note to readers you need to click on the graphs to get time labels.) It takes four years before the total enrollment will fully reflect the policy change. I suspect the difference in endpoint enrollments simply reflects a change in the applicant pool. I would very surprised if institutional racism were the driving force. Not in New York of all places. Statements like “The statistics clearly show that black New Yorkers are being shut out…” are completely unwarranted. In any case you can’t tell much from the data presented in the Times article other than the world changes. So I agree that Motl is confused in his modeling. But I disagree that his posting alone makes him a bigot. In the UK he could sue for libel and likely win absent additional evidence because the defendant carries the burden of proof.

wcw:

Racists do not design the IQ tests we use today. If you have evidence to the contrary share it. If the Jews really did score below average (reference please), they took a vastly different kind of test. Their scores today are not “slightly above the median,” at least not for Ashkenazi Jews, it’s more like one standard deviation. See http://en.wikipedia.org/wiki/Ashkenazi_Jews and the references therein.

You accuse me of race baiting. Show me where. It’s a shame that you have trouble expressing yourself without recourse to personal attacks. Isn’t that a trait of real bigots?

[wcw]

"Perhaps the most dramatic changes were those in the mental test performances of Jews in the United States. The results of World War I mental tests conducted among American soldiers born in Russia--the great majority of whom were Jews--showed such low scores as to cause Carl Brigham, creator of the Scholastic Aptitude Test, to declare that these results "disprove the popular belief that the Jew is highly intelligent." Within a decade, however, Jews in the United States were scoring above the national average on mental tests, and the data in The Bell Curve indicate that they are now far above the national average in IQ."

Thomas Sowell, Vol. 28, American Spectator, 02-01-1995, pp 32.
http://www.mugu.com/cgi-bin/Upstream/Issues/bell-curve/sowell.html

"Now imagine if Lay was not some rich white guy, but a poor black one accused of a violent crime and the prosecution used similar tactics. We would never hear the end of it. European governments would protest, he would get nominated for the Nobel peace prize and Al Sharpton would go on yet another nutritious hunger strike."

A. Zarkov, July 07, 2006 at 11:11 AM
http://delong.typepad.com/sdj/2006/07/i_think_joe_noc.html#comment-19422522

The exegetical approach of fitting data to orthodoxy has a weakness.

It is easily refuted by fact.

[lise]

WCW - there are protests that are needed. Thanks for protesting.

[A. Zarkov]

“The results of World War I mental tests conducted among American soldiers born in Russia--the great majority of whom were Jews--showed such low scores …”

Those Russian soldiers were relatively recent immigrants. As such they had poor skills in English and couldn’t read the test very well. Ten years later their children had good English language skills and could score consistent with their innate intelligence.

Thomas Sowell and the American Spectator as authoritative references—I'll remember that.

[Matt McIrvin]

Somehow I knew which physicist you were referring to before I started reading the post.

But, yes, judging from his papers, as far as I can tell Shalizi is himself a physicist or at least highly trained in physics.

[wcw]

A note to our host about my third, deleted comment may be viewed at http://www.bignose.org/~wcw/delong_heart_zarkov.txt

Don't click through, though -- you're not missing much, aside from my general befuddlement.

[Stewart]

Yes, the 1910-1920's data were often for early immigrants, not just 'Galician' but also Italian, Slavic, Portugese, and other low-scoring groups. (Sowell's comments were right, something I seldom say). However, the conclusions were based on 'sound science', or so they thought at the time. After all, intelligence is something innate, no? Well, also reflecting opportunity, family values, etc. Racists don't usually design these tests, but all tests use a combination of experience, reasoning, and speeded performance, which aren't equally valued in each setting and can differ a lot between families. The strongest predictor of a child's IQ is parental expectations of school performance, far stronger than parental income, education, number of books in the house, etc. I don't want to credit my kids school performance to their Jewish thrice-great grandmother. Again, why are folks of Chinese/Korean/Japanese background so over-represented in these programs, far beyond expectations from IQ. Similarly, sex differences in educational numbers are far beyond anything reasonably predicted by any differences in academic profile. As well, there can be differences in subgroup inclusion based on whether academic skills (writing, math, etc.) are used to admit, or IQ measures.
For a decent rehash of these issues in the past, I recommend Leila Zenderland or Franz Samelson, who had articles on "The Bell Curve" in the context of psychological history, in Journal of the History of the Behavioral Sciences, 1997, vol 33(2).

[Biff Spaceman]

No mention of institutional racism? Or the gross biases against all manner of European and other immigrants when they were allowed in to the US at all? And the Bell Curve? Ack.

[Biff Spaceman]

No mention of institutional racism? Or the gross biases against all manner of European and other immigrants when they were allowed in to the US at all? And the Bell Curve? Ack.

[eweininger]

Thanks for joining us folks. That concludes another episode of the Larry Summers Power Hour. See you next week.

[Stewart]

Remind me again: What's the name for the kind of courage it takes to consider a small aspect fo a large and complicated question, fearlessly avoid discomforting questions and history, and confirm that others are inferior to you? I know it's not science.

[mahndisa]

08 24 06

I do think that the tools in stat mech can apply to any physical system. The mathematical machinery that is used in time series analysis in economic trends etc is very much the same that is used in Fourier analysis, in principle. I think that physics is a fundamental of everything, if you choose to see the connexions.

[Stewart]

Where's Anne when you need her? There's a pertinent article in the August 9 NY times about family/home influences on academic performance, and our foolishness in believing that schools can make up the difference. (on the current website, in the education section)