October 31, 2007

Books to Read While the Algae Grow in Your Fur, October 2007

John Dewey, The Public and Its Problems: An Essay in Political Inquiry
I finally read this as part of writing my social media paper; I'm very glad I did, and wish I had done so earlier.
"The public" exists, potentially, whenever there are serious and persistent externalities; it consists of those who are on the receiving end of the resulting market failures. The public organizes itself to regulate those externalities; these specialized organs and officers constitute government, or the state. (Cf. ibn Khaldun.) The implementation of all this, and the monitoring of those officers, raises problems of collective action; but prior to this is a problem of collective cognition, of recognizing that these externalities exist and deciding intelligently, that is, with regard to concrete consequences, what to do about them. The great problem of the public is finding modes of organizing itself, and its inquiries into what should be done, which are consonant with the modern scope of externalities and interdependence brought about by industrialization. (I have modernized Dewey's terminology, but not, I think, distorted his meanings.)
Obviously, I think this is all very, very good, and would love to see what he would have thought of our modern technologies of communication. (I suspect that there would have been a certain amount of weeping and wailing and gnashing of teeth involved.) Extremely strongly recommended to anyone interested in democracy, general social theory, or social media.
Liz Ball, Month-By-Month Gardening in Pennsylvania and Pennsylvania Gardener's Guide
Rose Marie Nichols McGee and Maggie Stuckey, The Bountiful Container: A Container Garden of Vegetables, Herbs, Fruits and Edible Flowers
Having now gone through a full agricultural year, I think I can safely say that these are the best of the gardening books I've used. McGee and Stuckey in particular was very handy and had lots of useful ideas (for instance, Malabar spinach).
Ian Kershaw, Hitler, 1936--1945: Nemesis
The conclusion of the story of, probably, the worst person who ever lived: a man who in a better world would have been an unusually boring, cranky and mean coffee-house loafer. The fascination lies in seeing just how this man was able to cause so much damage and pain, which means that a large part of Kershaw's 887 pages (plus notes) is about the German state, German society, and — more and more as the story goes on — the war and the genocides. This is appropriate, because, as Kershaw makes plain, these horrible events were not just the products of Hitler's crazed beliefs and wicked desires, but also of the fit between those and what other members of German society, especially its most powerful members, wanted and believed, and the choices they made.
Let me just quote two paragraphs (from p. 841), which sum things up and give a sense of Kershaw's style:
Never in history has such ruination — physical and moral — been associated with the name of one man. That the ruination had far deeper roots and far more profound causes than the aims and actions of this one man has been evident in the preceding chapters. That the previously unprobed depths of inhumanity plumbed by the Nazi regime could draw upon wide-ranging complicity at all levels of society has been equally apparent. But Hitler's name justifiably stands for all time as that of the chief instigator of the most profound collapse of civilization in modern times. The extreme form of personal rule which an ill-educated beerhall demagogue and racist bigot, a narcisstic, megalomaniac, self-styled national saviour was allowed to acquire and exercise in a modern, economically advanced, and cultured land known for its philosophers and poets, was absolutely decisive in the terrible unfolding of events in those fateful twelve years.

Hitler was the main author of a war leaving over 50 million dead and millions more grieving their lost ones and trying to put their shattered lives together again. Hitler was the chief inspiration of a genocide the like of which the world had never known, rightly to be viewed in coming times as a defining episode of the twentieth century. The Reich whose glory he had sought lay at the end wrecked, its remnants to be divided among the victorious and occupying powers. The arch-enemy, Bolshevism, stood in the Reich capital itself and presided over half of Europe. Even the German people, whose survival he had said was the very reason for his political fight, had proved ultimately dispensable to him.

(A quote which should not be inflammatory, but under the present circumstances, is, from p. 779: "[General] Guderian recalled Hitler stating: 'The soldiers on the eastern front fight far better. The reason they give in so easily in the west is simply the fault of that stupid Geneva convention which promises them good treatment as prisoners. We must scrap this idiotic convention.' ")
Donald MacKenzie, An Engine, Not a Camera: How Financial Models Shape Markets
For the most part, this is a recounting of the history of the development of modern financial economics, circa 1950 to circa 1980 — the random character of securities prices, the Modigliani-Miller propositions about the irrelevance of capital structure, portfolio selection theory, the capital asset pricing model, and most of all the derivative-pricing problem, culminating in the work of Black, Scholes and Robert C. Merton, and the rise of the no-arbitrage condition and martingale methods as central organizing ideas in financial theory. This is also a history of the development of markets in derivative securities, covering that same period down to roughly 1990, and a peak into some of the things going on in the late 1990s, most notably a new explanation of how Long-Term Capital Management got in trouble. (To summarize: it was so successful that it was approximately imitated by many other funds, so they were all participating more or less strongly in a common "super-portfolio", and this itself created additional correlations among the assets in that portfolio.) These stories are all told really excellently; I don't think I've ever seen a better non-mathematical explanation of any of these matters, and the book is worth reading for them alone.
The frame-tale, however, is given by the subtitle: that the book is an investigation of how, and to what extent, financial theory is "performative". That is, MacKenzie wants to know not just whether participants in the financial markets talk about financial models (they do), or use them practically (they do), but whether that use caues the markets to change in ways making the models more accurate ("Barnesian performativity", which he admits slights the much greater sociologist Robert K. Merton), or indeed in ways which make the models less accurate ("counterperformativity"). This is an interesting idea, but to claim to have an example of either kind of performativity is to advance an extremely complex hypothesis about social causation, one which is in the nature of things very difficult to establish. MacKenzie (cough unlike some of his colleagues in science studies cough) grasps this, and is correspondingly cautious in his claims.
The best example of possible "Barnesian" performativity he provides concerns the Black-Scholes option-pricing formula. If one compares the prices it predicts for options to the (admittedly very limited) historical data on option pricing before the formula was widely used, the fit is not horrible but not outstanding. There then follows a period from the mid-1970s through 1987, when organized stock-option markets came into existence and flourished, and many of the participants deliberately used the formula as a guide to pricing. During this period, the fit of the model to the data is excellent. In particular, it implies a certain relation holds betwen the price of a stock-option contract, the strike-price of the contract, the expiration date of the contract, and the underlying stock's current price and the volatility of that price. All of these, except the volatility, are observable, so observed option prices can be used to solve for the implied volatility. This should be the same for all option contracts on the same underlying asset, which, during this period, they were, to a very close approximation. Since 1987, however, this nice constancy has disappeared, and implied volatility has varied systematically with the strike-price of the option. MacKenzie's interpretation --- which he supports in ways I will not go into here --- is that this systematic "volatility skew" is due to the great crash of 1987, and the caution, not to say fear, it continues to inspire about how certain kinds of trades can go wrong, thereby altering the prices people are willing to trade at. Thus an era when the theory was very strongly performative has, he says, been followed by one where it is not, though it continues to be used in other ways. It should be possible to model this through evolutionary game theory: Black-Scholes pricing strategies are able to invade at the expense of their predecessors, and they form a Nash equilibrium, but not an evolutionarily stable strategy.
Minor disappointments: the idea that markets are systems for collective calculation is not a recent invention of sociologists of science, but goes back to the participants in the "socialist calculation" debate of the 1930s — to von Mises, to Lange, and especially to Hayek. Speaking of the present epoch as "high" or "late" modernity carries connotation of "looking backwards" and of historical prophecy which is thoroughly unjustified.
MacKenzie says that his goal is to help improve the "conversation" about markets. That this book will improve the academic discussions of financial markets, economics, and the social life of the mind I have no doubt. But MacKenzie also says that he wants to contribute to the broader, non-academic discussion of marekts, and to to help people come up with positions more nuanced, and more useful, than "holy, holy holy is the invisible hand almighty" and "greed kills". (The caricatures are mine, not his.) This suggests an aspiration to political relevance which his findings, I think, lack. The basically political questions about financial markets have to do with details: what are the consequences of having this market in these securities, organized in this way, are those consequences good or bad, and what are the alternatives? MacKenzie contributes here to answering such questions only by forcefully demonstrating that financial markets are, indeed, human institutions, created, designed, re-designed and sustained by all the usual social processes. Some people will need this reassurance that the markets are not enchanted, that their current organization is not inscribed in the foundations of objective reality, before contemplating what, if anything, we should do about them. In this respect An Engine, not a Camera is itself more of a camera than an engine: but what a camera!
Jenny Davidson, Heredity
I hardly know how to even begin categorizing this one; a uniquely perverse fusion of the literary-investigation "secret history" novel with the 18th century London underworld and early 21st century biomedical schemes, slathered in great heaping doses of morbid self-abuse. But in the fun way one expects from the author of the Light Reading blog. I find it astonishing that this is a first novel.
Neil Mercer, Words and Minds: How We Use Language to Think Together
A social psychologist's investigation into "inter-thinking", and the detailed mechanics of conversation and collective cognition, including how different kinds of conversation lead to different kinds of "inter-thinking", which may be more or less productive. He particularly draws attention to the differences between "disputational" talk (basically, arguing with each other and self-defense), "cumulative" talk (building rapport and solidarity without concern for quality or accuracy) and "exploratory" talk, when we "engage critically but constructively with each other's ideas". The examples are largely but not entirely drawn from classrooms, which is understandable, though I'd like to see how well they generalize to other settings --- pretty well, I'd guess. My biggest reservation is that, like many people (rightly) impressed with Vygotsky, Mercer has an unduly negative view of computational models of cognition. I on the other hand think it would be fascinating, and important, to try to model his phenomena algorithmically.
This requires no background in linguistics or cognitive science to follow; it's almost a popular science book. Recommended very strongly if any of this sounds the least bit interesting.
Ivan Strenski, Four Theories of Myth in Twentieth-Century History: Cassirer, Eliade, Lévi-Strauss, and Malinowski
Interesting "ideas-in-context" intellectual history of the four theorists-of-myth named in the subtitle, trying to answer the question of why they bothered coming up with theories of myth in the first place, how the kind of theories they gave fitted in to the larger intellectual and social scnes they lived in, and why it is that their theories are so totally different it is very hard to believe that they are even talking about the same thing. The answer to that last, Strenski says, is that they aren't, that there is no well-defined subject of myths about which one might profitably theorize. One can appreciate his intellectual history without necessarily accepting that last point (with which I must say I am somewhat sympathetic). However, the discussion of Eliade is a little superseded by evidence which has since come to light about just what he was up to in the '30s (not good), and his connections to "traditionalism" in a rather different sense than Strenski uses the word (see here).

Books to Read While the Algae Grow in Your Fur; The Collective Use and Evolution of Concepts; The Dismal Science; Writing for Antiquity; Philosopy

Posted by crshalizi at October 31, 2007 23:59 | permanent link

Three-Toed Sloth:   Hosted, but not endorsed, by the Center for the Study of Complex Systems