Books to Read While the Algae Grow in Your Fur, December 2012

February 7, 2013

(This article was originally published at Three-Toed Sloth , and syndicated at StatsBlogs.)

Attention conservation notice: I have no taste.

Tamim Ansary, Games Without Rules: The Often-Interrupted History of Afghanistan
A history of the country from the beginning of the Durrani state in the 1700s to the present day, emphasizing the internal dynamics of the country, and how these have interacted with the designs of rival great powers. In particular, the rulers of the Afghan state have always had to be something of an interface between the largely-rural mass of the country (and the rural elites), and those same great powers. Ansary is very good at giving the Afghan point of view in a manner accessible to American readers — or rather, as he makes plain, an Afghan point of view, that of urbanized and westernized educated professionals, who still nonetheless have ties to the old world of the "village republics". His writing is clear and fast-paced, mingles sympathy with irony (describing the future Amir Abdurrahman when a boy of twelve or thirteen: "any fellow who can kill a man just to see if his gun works is going to prove useful to someone"), free of cant, and informed by good scholarship without being scholastic. It's the best introduction to Afghan history for a general American audience I've seen, and I hope it's widely read.
Disclaimer: My grandfather and Ansary's father were friends, so he's a family connection.
An essay by Ansary about writing the book.
Brian K. Vaughan and Fiona Staples, Saga
Mind candy.
Suzanne Mettler, The Submerged State: How Invisible Government Policies Undermine American Democracy
A short, repetitious, impassioned book with an important point. Over roughly the last 30 years, the US has seen a huge expansion of what Mettler calls "the submerged state", whereby the government pays people, or non-governmental actors (e.g., private firms or charities), to do things, often to do things that in other places or other times would be directly performed by the government. Sometimes the payment is direct, and sometimes it takes the form of tax breaks. (From the viewpoint of the public budget, money not received because of such tax breaks has to be made up somehow, so such "tax expenditures" have to count as spending.) What follows from this?
  1. The extra layers of indirection make it hard for citizens to know what, exactly, their government is doing, which is bad on just about any view of democracy. Whether you want to think of democracy as being about public accountability, or popular sovereignty, or problem-solving, deliberately obscuring what the government is doing is not good. (It's cutting feedback channels, or at least filling them with noise and distortion.)
  2. The indirection hides just how much the government intervenes in many sectors, such as health care (subsidized through not taxing employer-provided medical insurance), real estate (through not taxing home mortgage interest), and even the arts (through not taxing charitable donations). Many beneficiaries of the submerged state do not think of themselves as engaging with a government program at all*. Even if one thinks that the state should play a minimal role in the economy, this is not really getting there.
  3. The benefits of the "submerged state" tend to be much more tilted towards the already rich and powerful than are direct forms of social provision, especially through the use of tax breaks. Experimentally, as Mettler shows, when you tell Americans about who benefits from things like the mortgage interest deduction, support for them goes way down.
  4. Indirect provision tends to create interest groups of middle-men, for whom the particular organs of the submerged state are not remote and hazy, but real and vivid parts of their livelihood. Any change to their particular tax break or provision is accordingly watched, and lobbied, with great concern.
Put this together, and you have a recipe for using the government as an agency for upwards redistribution, especially towards particular interest groups which will entrench themselves, in ways which will make it very hard for citizens to notice, or for the consequences to be changed. Though I didn't notice Mettler referencing any of the "public choice" literature, or even Mancur Olson, it's very much an application of similar ideas to criticizing US policy from the (very slightly) left, as opposed to the usual public choice school, which is somewhat to the right of Attila the Hun.
As I said, this is a short book, but still repetitious, and the writing is not going to set anyone on fire**. But the message is important and sound, and "the submerged state" is a handy phrase.
— After writing this up, I remembered Henry Farrell post on this, which points to Mettler's own summary in The Washington Monthly.
*: I reproduce in modified form Mettler's table 2.1, which gives "Percentage of Beneficiaries of Specific Programs Who Report that They 'Have Not Used a Government Social Program' ". The first, italicized programs are among her prime examples of "the submerged state".
Program "No, Have Not Used a Government Social Program" (%)
529 or Coverdell Tax-Deferred Savings 64
Home Mortgage Interest Deduction 60
HOPE or Lifetime Learning Tax Credit 60
Student Loans 53
Child and Dependent Care Tax Credit 52
Earned Income Tax Credit 47
Social Security — Retirement and Survivors 44
Pell Grants 43
Unemployment Insurance 43
Veterans Benefits (other than G.I. Bill) 42
G.I. Bill 40
Medicare 40
Head Start 37
Social Security Disability 29
SSI—Supplemental Security Income 28
Medicaid 28
Welfare/Public Assistance 27
Government Subsidized Housing 27
Food Stamps 25
(Mettler reports to tenths of a percent. Since the total sample size was only 1370, I don't believe that third digit, and have rounded accordingly, but kept her ordering.) — While something clearly very odd is going on when a quarter of those who say they have used food stamps also say they've never used a government social program, it is a different odd thing than 60% of the beneficiaries of the home mortgage deduction saying they haven't.
**: Though most of it avoids the spectacular mixture of metaphors on p. 4, where, within the space of a paragraph, Obama "steers" his policy agenda into the "looming precipice of the submerged state", submerged precipice is simultaneously a "thicket" and some sort of machine. It's the closest to "the Fascist octopus has sung its swan-song" that I have ever seen from a non-student. — Update: Since people are writing in with much worse manglings of metaphors from David Brooks and even the the Moustache of Understanding himself, Thomas Friedman, which I had mercifully forgotten, I withdraw my remarks.
Richard S. Westfall, Never at Rest: A Biography of Isaac Newton
Portrait of the transcendent genius as a deeply unpleasant man. (Hooke, Leibniz, and Bernoulli do not end up looking much better.) Reading the key chapters, on what Newton actually did to make himself immortal, requires some grasp of analytic and Euclidean geometry, calculus, and classical mechanics. Also some knowledge of what alchemy was about, and for that matter the religious disputes of NW Europe circa 1700, would be helpful. With that, it's a great picture of a truly incredible mind.
Remark 1: Westfall is convincing that Newton, after an early infatuation with "the mechanical philosophy" of Descartes et al., tried very hard indeed to create a quantitative and experimental but non-mechanical natural philosophy. The whole point of his saying "I frame no hypotheses" about where the inverse square law comes from was to not present a picture of a clockwork universe. The mechanical philosophers got this once the Principia came out, and objected accordingly.
Remark 2: Westfall is also good at how novel it was for Newton to try to account for phenomena in quantitative detail — e.g., to use universal gravitation to explain not just why planets orbit the Sun, nor even why they do so in ellipses obeying Kepler's laws, but also to account for deviations from those laws. This was a new ideal of explanatory precision and unification, and in several places in the Principia Newton produced its appearance only by sharp practice.
Remark 3: The dominant impression I take away from Westfall's picture of Cambridge University in Newton's time is that of a sheer waste of financial and human resources. This wasn't the waste of trying many new things, most of which fail, nor was the intellectual deadness of most of the fellows exhaustion after having spent productive years competing for positions (as it sometimes is in modern academia). It was just a corrupt seeking after sinecures. It's true that this corruption kept paying Newtown while he worked, even though his science had nothing to do with the university's mission (and he seems to have been an awful and negligent teacher, and of course a heretic). One could argue that one Newton was worth the cost of both English universities, even if they produced nothing else of value. This feels a little like excusing my breaking thousands of eggs because somebody took a few of them to make an omelet while I wasn't looking.
Remark 4: The Newton-Leibniz priority dispute, in this telling, seems like nothing so much as a flame-war conducted with all the speed which communications c. 1700 allowed, complete with ever-lengthening replies to replies, and Leibniz using a sock-puppet to pronounce that lurkers support him in e-mail. Something about communicating through print, with insecure status hierarchies, perhaps?
Dennis Tedlock, 2000 Years of Maya Literature
This anthology runs, as promised, from ancient inscriptions (reproduced) through astronomical codices to post-Conquest mythic poetry and dramas. Each chapter begins with an introduction and outline of the piece translates, often quite lengthy, followed by the text and a facing translation, and then commentary. Tedlock's text is clear, and while I can't judge the accuracy of his translations, he seems to be trying very hard; certainly he goes out of his way to make everything accessible to those with no previous knowledge of Maya civilization. It's also a gorgeously produced book.
I have to confess that I was not much moved by the translated writing, which may have to do with the translator (Tedlock has read a lot in the wrong sort of literary theorist), or the sheer difficulty of making poetry work in translation, or perhaps the subject-matter. (Monarchical inscriptions, astronomical calculations, etc., from ancient times, or ritual drama from the colonial period, are intrinsically hard sells.) But I learned a lot, and I'm glad I read it.
G. E. R. Lloyd, Demystifying Mentalities
This is a very rich little book which I won't do justice to.
Fundamentally, Lloyd is attacking the notion that different cultures have different "mentalities" — that their members think in fundamentally different ways. The roots of this notion go back to Levy-Bruhl's ideas, around the beginning of the 20th century, of a pre-logical "primitive" mentality, which didn't care about contradictions and thought of related things or ideas somehow "participating" in each other. From there, the idea of multiple distinct mentalities for multiple societies spread to other anthropologists, to historians, and even to sociologists. (As Lloyd notes, Levy-Bruhl had some second thoughts about this.)
The idea of distinct mentalities tends to get invoked when members of some ancient or distant society say (or more rarely do) things which an observer finds extremely puzzling — the classic "apparently irrational belief" of otherwise-rational people. What Lloyd hammers home is that invoking distinct and divergent mentalities actually does nothing to explain such phenomena. What might help is to look at the explicit concepts available to people — e.g., do they have a clear distinction between literal and metaphorical statements? — and at the communicative contexts these statements get made in.
Lloyd's claim, then, is that members of different societies don't have different kinds of minds, but rather that different societies give their members more or less opportunity to employ different, but universally-present, modes of thought. Social and political contexts, and patterns of communication, not only preferentially evoke different modes of thought, they also create opportunities to articulate them, to refine the conceptual toolkit that goes along with them. In every society, there are opportunities for arguing and giving reasons, but some societies provide (some of) their members with more opportunity for this. Some social contexts encourage doing this adversarially, some more cooperatively; in some there is a broad audience to persuade, in others there's just one decision-maker to win over, etc. This will lead to more or less refinement of different modes of thought, and to more or less self-consciousness about what is going on.
To be a little more concrete, Lloyd traces the ancient Greek concerns with proof and foundations not just to the generally-adversarial aspects of Greek culture, but also to the wide-spread participation in legal juries and political assemblies, along with deep ideological divisions over how to run a polis. The audience of the philosophers and the sophists were citizens who were very experienced with weighing arguments, who had to be won over in the mass (and so impersonally), and who had some experience of quite fundamental and deep disagreements — not to say one-up-man-ship*. (As an exaggerated slogan: Logic was the child of democracy.) The context for Chinese philosophy was quite different, and while rhetoric and logic were by no means completely neglected, the fact that there the goal was always to win over a monarch channeled energies in very different directions. I find this argument extremely plausible, but wish he had been able to add ancient India into the comparison...
The social psychologists speak of the "fundamental attribution error", the tendency to over-estimate how much of what someone does stems from the kind of person they are, and to neglect the kind of situation they find themselves in. Though Lloyd doesn't put it this way, he's basically saying that mentalities only seem plausible because we apply the fundamental attribution error to whole societies.
*: Lloyd in fact traces explicit distinctions between myth and history, or myth and reason, and between literal truth and metaphors, to such one-up-man-ship on the part of ancient rationalists. This is part of a fascinating discussion on just why so many ancient rationalists, such as the Hippocratic author of On the Sacred Disease, were so much more cogent when attacking predecessors than supporting their own positions. This is one aspect of the book I won't do justice to.
David A. Harris, Failed Evidence: Why Law Enforcement Resists Science
DNA testing very much excepted, most "forensic science" lacks any basis in actual science, or even any demonstrated empirical reliability. (This does not stop, e.g., finger-print examiners at the FBI from saying that finger-print identification is infallible.) Beyond this, psychologists have spent a lot of time identfying the ways in which eye-witness testimony can be unreliable, how the traditional identification line-up encourages false positives, and how innocent people will confess to serious crimes. Perhaps unsurprisingly, there are now about 250 cases in the US of people who have been convicted of crimes, and sentenced to execution or life imprisonment, whose innocence has been established by DNA testing. It doesn't seem possible to extrapolate from this to a general rate of false convictions, but it seems insane not to try to improve the standards of evidence in the criminal justice system accordingly. Yet, with a few exceptions (that Harris celebrates), reforms are resisted fiercely by police and prosecutors. Mostly, this book is devoted to documenting, and elaborating on, these themes, along with some suggestions for reform which could be summarized thus: point to the places where things have been improved and the sky hasn't fallen; find right-wing sponsors for reforms; and, bang on the fact that locking up the wrong people leaves the actual criminals free. This seems weak, and I am not optimistic.
— Harris elaborates greatly on ideas like "loss aversion" to explain why police and prosecutors are not eager to give up on unreliable forensic techniques, switch to blinded sequential line-ups, etc. In this he seems to be far more subtle than there is any call for. Let me mention three big things he neglects. (1) Saying that police and prosecutors have been relying on — have been swearing to — crappy evidence is going to be felt as an attack on the institutions with which they strongly identify, and often on them as individuals. Resistance and rejection are very natural defensive responses. They are not so much intellectual acts (as cognitive dissonance would suggest) as ways of saying "No, asshole, fuck you!". (2) Admitting the problem will undermine trust in law enforcement, imperil the livelihood of the forensic technicians, and open the way for an immense number of law-suits and claims for re-trials, including many on the part of people who are actually guilty. (Harris mentions the re-trials issue, but doesn't give it anywhere near enough weight, it seems to me.) Law enforcement has a massive direct interest in denying that anything needs to change. (3) At least some people in law enforcement — I honestly have no idea of how many — feel that that nobody gets charged with serious crimes unless they're bad actors. Even if they're not guilty of the particular crime they're charged with, they're guilty of something, and the rest of us are better off with them in jail*. Matching crimes to criminals is a legal fiction, the important thing being punishing undesirables.
— A further quibble. Harris really dislikes forensic techniques which rely on the technician making a judgment about matches or comparisons; he complains they're unobjective and unscientific. Now, I was never very good at experiments, but even so I put in enough time in physics and chemistry labs to appreciate how a lot of the skills needed to make scientific apparatus and procedures work are matters of muscular and perceptual skill, very hard to articulate, or to pass on other than through demonstration. (It is no accident that the idea of "tacit knowledge" came from an eminent chemist.) It doesn't bother me that (say) finger-print examiners can't put in words (or math or code) what makes them say that a match is good. What bothers me is that they can't show that they aren't making tons and tons of errors when they say that a match is good, or for that matter when they rule someone out as not matching. What bothers me even more is that they resist elementary steps to control error, like having identifications checked by a second examiner who doesn't know what the first examiner concluded.
*: I forget who said "spank your children every day; if you don't know why, they do", and Google is unhelpful.

Books to Read While the Algae Grow in Your Fur; Writing for Antiquity; Commit a Social Science; The Great Transformation; Physics; Mathematics; Psychoceramica; The Beloved Republic; The Collective Use and Evolution of Concepts

Please comment on the article here: Three-Toed Sloth