<< Just Possibly Like So but Maybe Not So Much Stories | Front Page | Visions in the Yamuna: Nirmal Verma >>
Wednesday, October 26, 2005
PSA/A New Review of A New Kind of Science
Many of you may remember Prof. Synecdoche. He seems to have deleted his blog. Unfortunately, it’s been replaced by a porn-spam blog. So, I’d advise everyone to de-link.
Polymath Cosma Shalizi has an entertaining review of Stephen Wolfram’s A New Kind of Science. I have a paper in various stages of revision on the rhetoric of Wolfram’s book, and Shalizi’s discussion of Wolfram and the taxonomy of crankishness is very apt there. In fact, I invoked his guano comparison in the version I read at a conference.
I have to register disagreement in a few places, however. I suspect that there have to be correlations between any useful version of “complexity” and what is visually interesting to the cortex of an East African plains ape. Wolfram is indeed vague on that point, and I appreciate quantitative measures of complexity as an abstract principle, but I’m not convinced that it is as arbitrary as Shalizi thinks it is. I’ve read Investigations and am intrigued to learn of the apparent existence of a cult devoted to it (shape spacers?), though I very much appreciate that Shalizi has also been annoyed by Lakoff’s definition of cognitive science (the worst display of which I’ve encountered is in Philosophy in the Flesh).
Of Grammatology‘s inclusion in the list of crank works at the end is unfortunate, however. Putting aside the question of whether you think it is a major philosophical work or one filled with elementary misreadings (as Chomsky has said), it presents a trial for its readers. The other works listed there (and Wolfram’s) all seek to explain complex matters very simply.
Comments
Of Grammatology‘s inclusion in the list of crank works at the end is unfortunate, however. Putting aside the question of whether you think it is a major philosophical work or one filled with elementary misreadings (as Chomsky has said), it presents a trial for its readers.
You’re right to insist that there’s a difference between a crank and a fraud.
(runs and hides)
Jonathan,
It seems to me odd to contrast ‘presents a trial for its readers’ and ‘seeks to explain complex matters very simply’. Much of the history of philosophy (and science) is a matter of readers (and future investigators) trying to figure out whether simple explanations and reductions have any merit, if so how much. You seem to be conflating the vice of oversimplification (which is really a couple vices in one) with more rigorous applications of Occam’s Razor.
And not to get all Bloomian, but it is quite plausible that most major philosophical works are filled with elementary misreadings, as major philosophers strive too hard to relegate their predecessors to the dustheap. Misreadings are corrected by careful scholars and exegetes, who tend not to produce major works of philosophy. I take it to be pretty obvious that Derrida IS full of elementary misreadings - of Plato and Saussure and many others. The only question is whether they are ‘strong’.
Misreading is a tricky concept, unless you’re a nineteenth-century German philologist, which would be very interesting if true.
Cosma put the Derrida after two new agers, which I thought was a fairly gratuitous insult. Simple and intuitive are distinct. I think the New Age stuff tends to rely on the latter, whereas Derrida is neither simple nor intuitive. I suppose the point of contact is supposed to be their followings, but still.
I’ve read big hunks of Wolfram’s book and found it, well, intriguing. I think the notion of computational irreducibility is a useful one—or at least, I’ve cited it in an article or two.
The question of whether Wolfram is merely a self-regarding crank, or a genuine self-regarding crank genius is not something on which I am even qualified to have much of an opinion as I’m in no position to assess his contributions to “the field,” whatever that field. But I does seem to me that Shalizi—I’ve read many of his reviews, many of this interesting too—may be as self-regarding as Wolfram is. And that leads me to take his judgment of Wolfram with a grain of salt.
That strikes me as a highly dubious claim, Bill. Have anything in particular in mind?
Which claim are you talking about? My claim about computational irreducibility of about Shalizi’s self regard?
On the latter, frankly, that seems obvious to me simply from the wide variety of topics on which he publishes pronouncements. It’s one think to read broadly; it’s something else again to publish reviews on a broad range of subjects. His confident dismissal of Derrida is a case in point. Now maybe he isn’t in Wolfram’s league, but he’s no piker either.
If you restrict yourself to the reviews per se, I’m pretty sure that he’s got a sure background in his subjects. The Derrida thing bothers me a bit because of his tacit assumption that his readers will agree with him that it’s apparent that Derrida is fraud, etc. in the same manner he argues that Wolfram is. I think that, independent of whether that was true (and I don’t think it is), they’d have to be classified very differently.
"If you restrict yourself to the reviews per se, I’m pretty sure that he’s got a sure background in his subjects.”
Maybe so, but it’s an awefully wide range of subjects. Here’s a list of his recent reviews:
http://cscs.umich.edu/~crshalizi/reviews/
Yes. If you read all of those plus his notebooks, you’ll see that he’s a national treasure.
But there’s no contradiction between being a national treasure and being self-regarding. None at all. (I’m thinking e.g. of all the musicians from Bix Beiderbecke through Michael Jackson who have self-imploded because they were unable get beyond the vortex of their own creativity and success.)
And one thing that self-regarding national treasures do is put other national treasures in their place. In that game, Wolfram’s been at it longer and has a much bigger platform than Shalizi. As far as I can tell, he’s also got a more substantial record of accomplishment.
So we’ve got an up and coming treasure taking aim at an established treasure. What could be more banal?
If Cosma trumpeted himself and his work the Wolfram does, your comparison might be valid. But that’s the key distinction.
The key distinction with respect to what, exactly?
I think it takes a pretty high order of chutzpah to put your personal notes out on the web, as through your every thought were worth the attention of others.
The key distinction is that Wolfram considers himself an isolate genius. He has not published papers in peer-reviewed forums in quite some time. His work steadfastly ignores or minimizes the work of others.
Rather than chutzpah, I see a valuable civic service. I have learned much from his notes, as would if you read them, and I wish more people would do it. In fact, it would be a good thing if more people could have his range of reference. That they generally don’t is a symptom of the pressure towards overspecialization in our educational system.
Why do you think I haven’t read his notes and reviews?
Two reasons, I suppose: quantity first of all and quality inconsistent with your previous remark. Now if your own personal erudition is such that you learned nothing new, then you have a moral obligation to post your own notes somewhere. Unless you’re hoarding it all like the sea in a Wolframian manner, of course.
I don’t follow.
Yes, he’s written lots of reviews and posted lots of notes. I’ve read around in that stuff and have copied at least one of his reviews (of Moretti’s Atlas) into my notes. As for quality, how can you judge the quality of his notes and reviews if you don’t yourself know the material he’s dealing with? Sure, he’s brilliant. That’s not at issue, at least not for me.
Wolfram is brilliant as well; he’s also ill-mannered. That in itself doesn’t make his ideas wrong. Why doesn’t Shalizi rest his case on technical issues? Why does he go to the trouble of citing Gardiner on intellectual crankism and then showing, clause by clause, how Wolfram meets the definition?
That Wolfram’s behavior is distinctly odd and flaunts the mores of institutionalized intellectual life, that is obvious on the face of it and need not be belabored. So why belabor it?
* * * * * *
As for my own erudition, I’ve got a bunch of papers here and there on the web, with the biggest chunk located here:
You’ll find three of my recent papers on literature in an online journal called PsyArt. And I contribute to a number of listserves—Ian Pitchford’s evolutionary psychology list, cognitive linguistcs, biopoetics, and one or two others.
I have no intention of posting my notes to the web, though I share them with correspondents and so forth.
I think Wolfram’s behavior towards Matthew Cook goes a bit beyond “ill-mannered.” In addition to credit-hoarding, self-publication and consequent withdrawal from the normal process of scientific review invite the discussion of crankishness. You can’t thoroughly discuss the book’s merits without addressing why it was published in this form. Your earlier suggestion that Wolfram’s behavior is comparable to publishing research notes and reviews of various subjects on your personal website retains a puzzling, tu-quoqueish aspect.
Perhaps a more useful subject of conversation would be what, if any, technical aspects of Shalizi’s review do you disagree with and why?
Yes, Wolfram’s treatment of Cook was deplorable. And, no, I don’t Shalizi has done anything so bad. And, sure, he can publish whatever he wants on his website, as can anyone else. And if someone else decides to put their notes on their website, I’m likely to think they’re self-regarding as well. I know of one case, Howard Bloom, and he’s certainly an intellectual megalomaniac, brilliant, but also undisciplined.
There is no valid comparison between placing well-sourced academic notes on a website and the behavior of either Wolfram or Bloom*. It’s not undisciplined. It’s not megalomaniacal. It’s practical and useful to yourself and others.
*I’ve never heard of this man and just looked at his web site, where I read a fever dream about Pakistan having access to submarine technology beyond the ability of other world navies to detect and which Bin Laden is going to commandeer and use to nuke a few U.S. cities. If this happens, and we both survive, I pledge here now to send a copy of “King of the World” on tape to Bloom via Kevin Costner.
Yes, I think that Shalizi’s placing his raw unedited notes—he says first drafts, he never revises—on the web is undisciplined. If you don’t think so, then you don’t think so. We disagree.
The most charitable interpretation I can come up with for why you would hold this altogether baffling opinion is that you think that the act of publishing notes on a web site is intended to convey finality and authority on the subject in the sense that a peer-reviewed article would, perhaps, rather than simply being notes on a web site. Someone might indeed frame such writings that way, but Shalizi clearly does not. He even has a FAQ about it.
Also, Fredkin is a better source about computational irreducibility.
I think putting your raw notes where everyone can see them is an act of vanity. In so doing, Shalizi is inviting others to place a value on them which he then denies in his FAQ. And so, in writing that FAQ, he has now declared his superiority to those poor benighted web surfers who think too highly of his humble scribblings. Gimme a break!
Before the web, such vanity was expensive, and so not often done. Now it is cheap.
Maybe my age is showing. I’m still living in a world where the notes and letters of major thinkers are gathered together and placed in liberaries and, in some cases, even published, though not so much for the ideas they contain as a source of insight into the mind of their author. Perhaps the ease of self-publication simply makes that notion obsolete. In any event, Shalilzi has not yet been declared a major thinker and I do find that his note-posting is an encroachment on that territory.
I keep thinking of a passage I read in a biography of Richard Feynman—written, I believe, by James Gleich. Someone was visiting Feynman at his office on an intellectual matter and Feynman pulled something out of his files on that same topic. The visitor remarked that the notes Feynman regarded as unpublishable were better than papers in the refereed literature. I don’t think Feynman would have published those ideas on the web, no matter how cheap it may be to do so.
In any event, this conversation is getting more interesting.
Just what is the proper boundary between public and private in the internet age? How is that sort of issue negotiated?
Let me give you a specific example of why I think your argument here teeters on the edge of absurdity:
Here is a “raw note” entitled Markov Models.
It’s a somewhat annotated bibliography, with many links, on the subject. You are suggesting, stupendously, that putting this and others like it on the internet is an act of vanity. That it has some undetermined negative effect on his published and to-be published scholarship.
Scholars help organize a chaotic world. Sharing the results of their research, even in fragmentary form, is a good and necessary thing. These notebooks and reviews are precisely not empty. The internet would be a much richer place if everyone with the necessary resources had followed Shalizi’s lead in 1994, and it will be so ten years from now if more people emulate him. In fact, I think what he’s done presents one of the best models for the increased utility and relevance of academic blogging. Academic departments should uniformly credit this kind of activity as service.
Come on, how’s that list significantly better than going to Google Scholar and doing a search of “markov models”? I put myself through graduate school writing abstracts for what was then the Amereican Journal of Computational Linguistics (now I believe it’s just the Journal of Computational LInguistics). None of Shalizi’s annotations are up to the standard I had to meet to get a pay check.
That list is certainly better than nothing if you’re going in search of stuff on Markov models. But “nothing” isn’t the relevant comparison. That Google Scholar search is one of those comparisons. Another would be the bibliography that shows up in some of the online papers turned up in the Google search.
Is Shalizi a smart guy? Yes. is there some interesting stuff in his notes and reviews? Yes. Can you take any of it at face value?
One tenth the overall character count at a higher quality would be more of a contribution.
There are a couple of points here: first, he’s been doing this before Google Scholar and indeed Google existed.
Second, remember that you were extending your vanity-claim to include his book reviews, some of which have been published in the Santa Fe house journal and Quantitative Finance. You’re now stating that some of the same material could, most likely, be found in recently available resources; and that a highly specialized journal you once worked for required more detailed abstracts of its articles.
As for taking it at face value, when you look at someone’s vita and see that he publishes articles in Physical Review E, Physica D, BBS, and the Journal of Statistical Physics, inter alia, that’s a good indicator of relevant expertise. If he also has bibliographic notes about other subjects, so much the better.
It’s increasingly difficult to avoid the impression that you’re flying the tu quoque plane until it crashes into the mountain here. I am very curious to know if anyone else has ever come across Cosma’s site, read through some of the material there, and come away thinking “what chutzpah he must have for cluttering the internet?"
I’m with Jonathan on this one. (Disclosure: Cosma is someone with whom I’ve exchanged occasional friendly email, lo the last 2+ years.) I think he comes across as rather modest, even wryly self-deprecating actually. His eclecticism is endearing, in a bloggy way, even while his expertise is impressively extensive. I think Jonathan is right that it isn’t fair to compare him to Google scholar and such, given how many years earlier he started this whole thing. He’s a pioneer as well as a treasure.
I find Shalizi self-regarding. Is he in Wolfram’s league on that? No, on that I over-stepped. But there is plenty of room for smug self-satisfaction without being in Wolfram’s league.
That he is intelligent, witty, hard-working, is not at all evidence against that judgment. Nor is even a bit of self-deprecation, especially of the wry kind. He’s put a lot of stuff out on the web. As far as I can tell, it’s of widely varying quality. I’m not so impressed with long reading lists—of things read & good, read & bad, to be read, annotated and not—as you are.
What particularly set me off is the tone of his review of Moretti’s Atlas of the European Novel, a book I’ve notl read. Here’s his review:
http://cscs.umich.edu/~crshalizi/reviews/atlas-of-the-european-novel/
And here’s an anonymous review at a site billing itself as the Complete Review:
http://www.complete-review.com/reviews/morettif/atlasen.htm
Both reviews are critical of Moretti, the Complete Review is, if anything, more critical. But it doesn’t call attention to itself, it’s not cute, and it’s not patronizing. Shalizi’s review is all those things, and for no good reason. It’s that difference in tone that leads me to say that Shalizi is self-regarding.
At another point
http://cscs.umich.edu/~crshalizi/notebooks/lit-crit.html
he says this: “At some point I should use this space to record some thoughts about what a natural history of literature would look like, and how it would differ from hitherto-existing literary criticism; but really I should be working now.”
Given what I’ve read of his—some 10s (but not 100s) of entries of all sorts—iIt’s not clear to me that he’s likely to have anything interesting to say on that issue, which interests me a great deal. If he gets serious, then he might have something to say. And perhaps he will get serious.