Three-Toed Sloth

Books to Read While the Algae Grow in Your Fur, December 2008

Joe Hill and Gabriel Rodriguez, Locke and Key, vol. 1: Welcome to Lovecraft

Psycho killers, life in a haunted house, and realistic grief. No Cthulhiana (yet), despite the subtitle.

Pat Lewis, The Claws Come Out

An affectionate send-up of various staples of the horror genre; effectively a short-story collection in comic-book form. I was tipped into buying my copy by the fact that Lewis is a local author, but worth it without that.

Fall of Cthulhu, vol. 3: The Gray Man

More Lovecraftian apocalypse-fiction. Only worthwhile if you've read the previous installments; but good if you have.

John Dewey, Freedom and Culture

A defense of liberal democracy against its 1939-vintage rivals, along with some rebukes (especially to Americans) about the gap between statements of democratic faith and actual conditions. Plus pleading for the scientific attitude as a natural accompaniment of democracy, with support going in both directions. (One wonders what Dewey would have made of the Lysenko affair.) No specifically pragmatist or instrumentalist doctrines are in play. — A systematic comparison of this with say Popper's The Open Society and Its Enemies would be interesting, but beyond me right now.

Overall, I'd say that you're better off reading The Public and Its Problems, unless you have a special interest in Dewey or mid-20th-century thought.

Andrew M. Fraser, Hidden Markov Models and Dynamical Systems

Full-length review: Statistics of Moving Shadows

Wen Fong (ed.), The Great Bronze Age of China

Catalog of an exhibit of artifacts loaned by the People's Republic to the Metropolitan Museum of Art in 1979. Mostly an collection of bronze vessels, covering a span of over 2000 years. Also some contemporaneous jade work, and some of the then-newly-found sculptures from the tomb of the First Emperor.

The Shang dynasty bronzes are (in my supremely unqualified opinion) some of the most beautiful things in the world. This is a great source of high-quality pictures, explanations of the casting process, and descriptions of both the internal artistic development and the role the object played in ancient Chinese societies. (Also some historical oddities, like Fu Hao.)

Objectively considered, that role was to display the power and wealth of aristocrats in maximally-impressive forms, charged with superstitious awe, and the art dwindled and died out when controlling bronze-casting no longer meant controlling the leading technology of the age. But most conspicuous embodiments of power and wealth are ugly, not beautiful...

John Kenneth Galbraith, Money: Whence It Came, Where It Went

Despite being 33 years old, this remains crystal-clear, funny, and on all important points correct. The main failure comes at the end: Galbraith was not pessimistic enough to anticipate that inflation would be broken by deliberately inducing a huge recession, together with a general policy of crushing the labor movement and upwards redistribution. (Actually, it's a nice question whether developments since c. 1982 haven't transferred inflation to asset prices.) We could do with a new edition of this book, perhaps with an epilogue by James K. Galbraith on what's happened since 1975.

Joe Abercrombie, The First Law: The Blade Itself, Before They Are Hanged and Last Argument of Kings

Witty, yet serious, epic fantasy trilogy. Warning: there are rather a lot of fairly detailed torture scenes. (One of the main characters, and it must be said far from the least sympathetic, is an inquisitor.) In a way, this is a reworking of the core epic fantasy material inherited from Tolkien, under the influence of a rather cynical view of human nature, and especially of people who end up seeking and holding power. (Spoilers: It's clear from the start that Bajaz, the Gandalf figure, is not a nice man, but still I was rather surprised to see at the end just how much of a manipulative and hypocritical tyrant he was. Indeed, everything the evil wizard Khalul does, he does too, though perhaps not quite at the same scale, along with contravening the First Law. (Which leaves me wondering who did in fact kill Juvens.) But, and I think this might be Abercrombie's point, if you imagine a near-immortal man with wizardly powers and an interest in the affairs of mortals, who likely is it that he will be benevolently disposed towards us, rather than treating us like insects? Would the relationship between such a wizard and a man he plucks from obscurity for kingship be that between Gandalf and Aragorn, or that between Bayaz and Jezal? And of course a near-immortal would have a deep appreciation of the power of compound interest...)

Update: Jonathan Goodwin has reasons for not liking these books.

Peter Whittle, Networks: Optimisation and Evolution

This is an unusual and interesting book, which provides an interesting perspective on some, but by no means all, of the mathematics of networks. The first part, which I found the most interesting and learned the most from, concerns distributional networks (e.g., pipes, roads), including material structures whose goal is to distribute stress optimally. The part on artificial neural networks following was alright, as far as it went, but did not go very deeply into the capacities and limitations of neural networks, nor into learning theory, nor into their evolution (as opposed to incremental adaptive weight changes). The final parts concern queueing networks, including their optimal control, and a look at communications networks such as telephone exchanges, Internet routing protocols, and the growth of the web. (Most of that last is in fact devoted to a model Whittle proposed some time ago of polymer growth; regarded as a network model it appears to be a special case of the more general class of exponential families of random graphs.)

The implied reader is someone who is very well-versed in Lagrangian optimization theory and reasonably familiar with physics, though not so much with statistics. Very little is said about real-world phenomena, nothing systematically. The only probabilistic models of network formation discussed are the Erdos-Renyi model and Whittle's own. Social and biological-but-not-neural networks are omitted. In general, Whittle is at his best when discussing designed networks, especially ones whose design admits of a clear objective function. He is less illuminating on ones which have grown. In short, I found Part I immensely better than the remainder, but that part was extremely good, and I'm happy to have read the rest for the sake of that.

Thanks to Cambridge University Press for sending me a review copy of this book.

(Erratum: The last paragraph of section 7.4, on p. 105, looks like it was cut-off, since the "very simple result, due to Clerk Maxwell" is never descrbed.)

Caitlín R. Kiernan, Daughter of Hounds

Ghouls, a la Lovecraft, and the human children they kidnap, in the modern world. There are repeat appearances by characters from Kiernan's earlier novel Threshold, and apparently from other books as well, but this stands alone.

Books to Read While the Algae Grow in Your Fur; Scientifiction and Fantastica; The Progressive Forces; Networks; Mathematics; Philosophy; The Dismal Science; Enigmas of Chance; Writing for Antiquity

36-350, Data-Mining: Self-Evaluation and Lessons Learned

Attention conservation notice: > 600 words on how I'd teach this semester's course differently next time.

So; grades are done, and, a decent interval after submitting the grades, I got the (anonymized) student evaluations. (Five of the eighteen students bothered to fill them out.) This seems like a good time to take a look at how things went.

Overall, I'm pleased with the semester. Their grades were quite good, and actual performance on the final exam was even better than I'd hoped — several students who'd done poorly on the homework pulled off really good exams, and nobody did much worse on the exam than on the homework. Most importantly, judging by what people wrote for the final, lots of them actually understood what I was trying to say. (Of course, I didn't give them a version of the final exam at the start of the class, so maybe they all knew it already.) I'm also reasonably satisfied with the choice of materials, and definitely think that replacing the weekly lab sessions with an extra lecture was the right thing to do.

Of course it wasn't all good. While linear algebra is not a pre-req for the class, I was still surprised at how unfamiliar many of the students were with it. The difficulty being that it is very, very hard to say anything about high-dimensional data without linear algebra. Some of them of course had no problem; perhaps I need a pre-test at the start, with catch-up reading for those without the background. (Making linear algebra an official pre-req doesn't seem like an option.)

The big issue, both from my point of view and according to three of the five students who bothered to write evaluations, were the programming assignments. These were much harder for them, especially for the bottom half of the class, than I had anticipated. In fact they kept being harder than I anticipated, so I really need to dial down the initial programming expectations, and include more programming instruction. (See previous post.) I am not sure what to cut to make room for this; the best approach might be to integrate demos and code walk-throughts with some lectures. Teaching them data-mining without getting their hands dirty, however, seems like a travesty.

Student participation also needs work. Out of eighteen students, there were, to first order, three who spoke up in lecture. (To second order, maybe six.) This was not a problem with them, but rather I should have done more to encourage the others to talk. Likewise, only three students came to office hours.

Some more specific things to work on, in no particular order:

• The "waste, fraud and abuse" theme never really materialized. I can't see how the class was any the worse for this, though.
• A bunch of the hand-outs are still heavily based on Tom Minka's old notes; re-write. (There were actually complaints about the style clash!)
• The information-theory lectures were very confusing to the students. Re-work? Shorten? Drop?
• Systematially use the np package for nonparametric regression and density estimation (and, with discrete response variables, classification). Abandon ksmooth, lowess, etc., for anything except decorating plots.
• Cut neural networks?
• I definitely need to cover mixture models and density estimation.
• Additive models should probably be a lecture (or two?) of their own, not just a (much-kvetched-about) homework assignment.
• Consider specifically talking about successive-approximation algorithms and their convergence.
• Actually give the causal-inference lectures! (And see if someone from downstairs can't be persuaded to demo Tetrad. [Better yet, bug them some more about making it an R package.])
• Rigorously enforce the rule against just turning in printed-out R sessions in homework.
• Prepare a handout on "Things I expect you to pretend to remember from intro stats"? (What should be on it?)
• More real-data examples? On the one hand, they're convincing; on the other hand, real real data is big, messy, ambiguous, and not-infrequently expensive. But on the third hand, perhaps that is a lesson which should be driven home.
• Replace the newsgroups example with text which doesn't take so much back-story to explain. (Also, a larger corpus.) Update, 14 Feb. 2009: K. suggests using Wikipedia, which is available for download. Probably still needs some pre-processing.

Update, 16 March 2009: A nice sequence might be: PCA (subtracting off successive principal components), to the coordinate-descent/back-fitting approach to linear regression, to the coordinate descent lasso, to additive models, to SpAM. But this will need a lot of linear algebra, and the middle steps are impractical.

Corrupting the Young; Enigmas of Chance; Self-Centered

Statistics 36-350: Data Mining (Fall 2008)

Since class begins Monday, this is a good time for the public website to make its appearance. As before, lecture notes will also be posted here; you can use the RSS feed for this entry to keep track of them.

0. Introduction to the course (25 August)
1. Information retrieval and similarity searching (25 August)
2. Multidimensional scaling and a first glance and classification (27 August)
3. A little about page-rank (29 August)
   Homework #1, due 8 September: assignment, R, newsgroups.tgz data file
   Solutions
4. Image search, abstraction and invariance; the accompanying slides (8 September)
5. Finding informative features (10 September)
   Additional reading: David Feldman, "Introduction to Information Theory", chapter 1
6. Information and interaction among features (12 September)
   Additional reading: Aleks Jakulin and Ivan Bratko, "Quantifying and Visualizing Attribute Interactions", arxiv:cs.AI/0308002
   Homework #2, due 22 September: assignment
   solutions, solutions code
   Note: Information theory, axiomatic foundations, connections to statistics — elaboration on some points raised in lecture (12 September)
7. Categorization: types of categorization, basic classifiers and finding simple clusters in data (15 September)
8. Hierarchical clustering; how many clusters? (17 September)
9. Yet more clustering (19 September; slides)
10. Making better features: transformations, principal components (22 September)
11. Mathematics of principal components analysis; interpretations and limitations of PCA (24 September)
12. Yet more on linear dimensionality reduction: PCA + information retrieval = Latent semantic indexing. Factor analysis: motivations, historical roots, preliminaries to estimation (26 September)
    Optional reading: Deerwester et al., "Indexing by Latent Semantic Analysis" [PDF]
    Optional reading: Landauer and Dumais, "A Solution to Plato's Problem: The Latent Semantic Analysis Theory of Acquisition, Induction, and Representation of Knowledge" [PDF]
    Optional reading: Thurstone, "The Vectors of Mind"
    Home #3, due 3 October: assignment
13. More on factor analysis: estimation and the rotation problem (29 September)
14. Principal Components versus Factor Analysis: worked examples, basic goodness-of-fit testing for factor analysis; R code for lecture (1 October)
15. The truth about principal components and factor analysis: strengths, limitations, factor models as graphical models, factor models and mixture models, Thomson's sampling model; R code for Thomson's model (3 October)
    Homework #4, due Friday, 10 October: assignment, nci.kmeans, nci.pca2.kmeans
16. Regression: predicting quantiative features: point prediction; expectations and mean-square optimality; regression functions; regression as smoothing; linear regression as linear smoothing; other kinds of linear smoothers; nearest-neighbor regression; kernel regression. R code for figures, data for running example (6 October)
17. The truth about linear regression: optimal linear prediction; shifting distributions and omitted variables; rights and obligations of probabilistic assumptions; abuses of linear regression; how to hurt angels (8 October)
18. Extending linear regression: weighted least-squares, heteroskedasticity, local linear regression. R code for figures, data for running example (10 October)
19. Mid-term review (13 October; no hand-out)
20. Mid-term: exam, solutions (15 October)
21. Evaluating preditive models: in-sample and generalization error; over-fitting and under-fitting; model selection, capacity control, cross-validation. R for figures. (20 October)
22. Using cross-validation: mechanics and examples (22 October; notes forthcoming)
23. Using non-parametric smoothing: adaptive smoothing, testing parametric forms (24 October; notes forthcoming)
    Homework #5, due Friday, 31 October: assignment; solutions
24. Prediction trees 1: mostly regression trees, plus a "classification tree we can believe in" (27 October)
25. Prediction trees 2: classification trees (29 October and 3 November)
26. Bootstrapping, Bagging, and Random Forests (5 November)
27. Combining Predictive Models and the Power of Diversity (7 November)
28. Linear Classifiers and the Perceptron Algorithm (10 November)
29. Logistic Regression and Newton's Method (12 November)
    Homework #7, due Friday, 21 November: assignment; solutions
30. Neural Networks: The Mathematical Reality (14 November)
31. Neural Networks: The Biological Myth (17 November)
32. Support Vector Machines (19 November)
33. Support vector machines continued (21 November; same handout as previous)
    Homework #8, due Monday, 1 December: assignment; solutions
34. The Lecture Full of Fail: The wrong data, lying data, covariate shift, low base-rates and overwhelming false positives, response Waste, fraud and abuse (24 November)
    Homework #9, due 15 December: assignment; solutions

Corrupting the Young; Enigmas of Chance

Minimal Advice to Undergrads on Programming

I seem to be mostly teaching classes with a big computational component. After being hit over the head a few times by the very, very wide range of programming skill among the students, I decided to write out some advice on how to program, with a bit of special reference to R. This is not advice on how to become a brilliant programmer, because I can't give such advice; I am at best adequate for a scientist. But that's all I ask.

Corrections and suggestions are appreciated.

Update, 28 December: What follows is an updated version, incorporating the useful suggestions of Geet Duggal, Derek M. Jones, Thomas Lumley and Chris Wiggins. The original, if anyone cares, is archived here.

---

In roughly decreasing order of importance:

Take a real programming class

Learning enough syntax for some language to make things run without crashing is not the same as actually learning how to think computationally. One of the most valuable classes I ever took was CS 60A at Berkeley, which was an introduction to programming, and so to a whole way of thinking. (The textbook was The Structure and Interpretation of Computer Programs.) If at all possible, take a real programming class; if not possible, try to read a real programming book.

Of course by the time you are taking my class it is generally too late to follow this advice; hence the rest of the list.

(Actual software engineering is another discipline, over and above basic computational thinking; that's why we have a software engineering institute. There is a big difference between the kind of programming I am expecting you to do, and the kind of programming that software engineers can do.)

Comment your code

Comments lengthen your file, but they make it immensely easier for other people to understand. ("Other people" includes your future self; there are few experiences more frustrating than coming back to a program after a break only to wonder what you were thinking.) Comments should say what each part of the code does, and how it does it. The "what" is more important; you can change the "how" more often and more easily.

Every function (or subroutine, etc.) should have comments at the beginning saying:

1. what it does;
2. what all its inputs are (in order);
3. what it requires of the inputs and the state of the system ("presumes")'
4. what side-effects it may have (e.g., "plots histogram of residuals");
5. what all its outputs are (in order)

Listing what other functions or routines the function calls ("dependencies") is optional; this can be useful, but it's easy to let it get out of date.

You should treat "Thou shalt comment thy code" as a commandment which Moses brought down from Mt. Sinai, written on stone by a fiery Hand. I will treat it so when I grade you.

RTFM

If a function isn't doing what you think it should be doing, read the manual. R in particular is pretty thoroughly documented. (I say this as someone whose job used to involve programming a piece of special-purpose hardware in a largely undocumented non-standard dialect of Forth.) Look at (and try) the examples. Follow the cross-references. There are lots of utility functions built into R; familiarize yourself with them.

The utility functions I keep using: apply and its variants; sort, order; aggregate; table; rbind and cbind; paste.

Start from the beginning and break it down

Start by thinking about what you want your program to do. Then figure out a set of slightly smaller steps which, put together, would accomplish that. Then take each of those steps and break them down into yet smaller ones. Keep going until the pieces you're left with are so small that you can see how to do each of them with only a few lines of code. Then write the code for the smallest bits, check it, once it works write the code for the next larger bits, and so on.

In slogan form:

1. Think before you write.
2. What first, then how.
3. Design from the top down, code from the bottom up.

(Not everyone likes to design code this way, and it's not in the written-in-stone-atop-Sinai category, but there are many much worse ways to start.)

Break your code into many short, meaningful functions

Since you have broken your programming problem into many small pieces, try to make each piece a short function. (In other languages you might make them subroutines or methods, but in R they should be functions.)

Each function should achieve a single coherent task — its function, if you will. The division of code into functions should respect this division of the problem into sub-problems. More exactly, the way you break your code into functions is how you have divided your problem.

Each function should be short, generally less than a page of print-out. The function should do one single meaningful thing. (Do not just break the calculation into arbitrary thirty-line chunks and call each one a function.) These functions should generally be separate, not nested one inside the other.

Using functions has many advantages:

• you can re-use the same code many times, either at different places in this program or in other programs
• the rest of your code only has to care about the inputs and outputs to the function (its interfaces), not about the internal machinery that turns inputs into outputs. This makes it easier to design the rest of the program, and it means you can change that machinery without having to re-design the rest of the program.
• it makes your code easier to test (see below), to debug, and to understand.

Of course, every function should be commented, as described above.

Never do the same thing twice

Many programs involve doing the same thing multiple times, either as iteration, or to slightly different pieces of data, or with some parameters adjusted, etc. Never write two pieces of code to do the same job. Never copy the same piece of code into two places in your program. Instead, write one piece of code (generally a function; see above) and call it twice.

Doing this means that there is only one place to make a mistake, rather than many. It also means that when you fix your mistake, you only have one piece of code to correct, rather than many. (Even if you don't make a mistake, you can always make improvements, and then there's only one piece of code you have to work on.) It also leads to shorter, more comprehensible and more adaptable code.

Use meaningful names

Unlike some older languages, R lets you give variables and functions names of essentially arbitrary length and form. So give them meaningful names. Writing loglikelihood, or even loglike, instead of L makes your code a little longer, but generally a lot clearer, and it runs just the same.

This rule is lower down in the list because there are exceptions and qualifications. If your code is tightly associated to a mathematical paper, or to a field where certain symbols are conventionally bound to certain variables, you may as well use those names (e.g., call the probability of success in a binomial p). You should, however, explain what those symbols are in your comments. In fact, since what you regard as a meaningful name may be obscure to others (e.g., me, when I am grading your work), you should use comments to explain variables in any case. Finally, it's OK to use single-letter variable names for counters in loops (but see the advice on iteration below).

Check whether your program works

It's not a enough --- in fact it's very little --- to have a program which runs and gives you some output. It needs to be the right output. You should therefore construct tests, which are things that the correct program should be able to do, but an incorrect program should not. This means that:

• you need to be able to check whether the output is right;
• you should program the test, so it checks whether the output is right (and you can easily repeat the test as many times as you need);
• your tests should be reasonably severe, so that it's hard for an incorrect program to pass them;
• your tests should help you figure out what isn't working.

Try to write tests for the component functions, as well as the program as a whole. That way you can see where failures are. Also, it's easier to figure out what the right answers should be for small parts of the problem than the whole.

Try to write tests as very small function which call the component you're testing with controlled input values. For instance, a test for a function which supposedly calculates derivatives might check whether it gets the derivative of x² or 7e^-5x right at ten randomly-chosen points. The testing function should warn you if the computed derivatives differ by more than a tolerance you specify from the actual derivatives. (That's why you're using such simple functions.)

With statistical procedures, tests can look at average or distributional results. For example, I once wrote a program to estimate some parameters by maximum likelihood; I could then use the fact that a likelihood ratio test should have a chi-squared distribution to check that the estimation part was working properly.

Of course, unless you are very clever, or the problem is very simple, a program could pass all your tests and still be wrong, but a program which fails your tests is definitely not right.

(Some people would actually advise writing your tests before writing any actual functions. They have their reasons but I think that's overkill for my courses.)

Don't give up; complain!

Sometimes you may be convinced that I have given you an impossible programming assignment, or may not be able to get some of the class code to work properly, etc. In these cases, do not just turn in nothing saying "I couldn't get the data file to load". Let me know. Most likely, either there is a trick which I forgot to mention, or I made a mistake in writing out the assignment. Either way, you are much better off telling me and getting help than you are turning in nothing.

When complaining, tell me what you tried, what you expected it to do, and what actually happened. The more specific you can make this, the better. If possible, attach the relevant R session log and workspace to your e-mail.

Of course, this presumes that you start the homework earlier than the night before it's due.

Avoid iteration

This one is very much specific to R. Explicit iteration in R is slow. (We could talk about the reasons for that sometime if you're interested.) In many languages, this would be a reasonable way of summing two vectors:

for (i in 1:length(a)) {
  c[i] = a[i] + b[i]
}

In R, this is stupid. R is designed to do all this in a single "vectorized" operation:

c = a + b

Since we need to add vectors all the time, this is an instance of using a single function repeatedly, rather than writing the same loop many times. (R just happens to call the function "+".) It is also orders of magnitude faster than the explicit loop, if the vectors are at all long.

Try to think about vectors as vectors, and, when you need to do something to them, manipulate all their elements at once, in parallel. R is designed to let you do this (especially through the apply function and its relatives), and the advantage of getting to write a+b, instead of the loop, is that it is shorter, harder to get wrong, and emphasizes the logic (adding vectors) over the implementation. (Sometimes this won't speed things up much, but even then it has advantages in clarity.)

I emphasize again, however, that the speed issue is highly specific to R, and the way it handles iteration. A good programming class (see above) will explain the virtues of iteration, and how to translate iteration into recursion and vice-versa.

Manual trackback: Stephen Kinsella; Quantum of Wantum; Hacker News; Uncertain Principles; The Shape of Code

Corrupting the Young

Books to Read While the Algae Grow in Your Fur, November 2008

Carrie Vaughn, Kitty and the Silver Bullet

Fourth volume in the continuing adventures of a werewolf named Kitty; here combining what amounts to an ugly gang war among supernatural denizens of Denver with only-too-recognizable family dramas. (Previous installments: 1, 2 and 3. But probably still fun without them.)

Chris Moriarty, Spin State

Mind-candy. Hard-ish SF about quantum teleportation and labor struggles in a mining settlement. (Only hard-ish because some important things turn on some fairly bogus points.) Plus Gibsonian cyberspace, which is always irritating. But quite good nonetheless, especially as a first novel. Thanks to Dan Marthaler for the recommendation.

A. Merritt, The Moon Pool

On the one hand: that was a lot better when I was 14. (To be fair, Merritt's white characters are just as much ethnic stereotypes as his non-white characters.) Plus, the world-building has huge gaps, even by adventure fiction standards. On the other hand: there was a reason I read my way through everything he wrote, and it was that he could in fact really tell a story.

There's a free electronic edition.

Nadia Gordon, Sharpshooter

Mystery: murder among the wine-and-foodies of Napa Valley. Fun. Followed by Death by the Glass.

Larry Gonick and Craig Criddle, The Cartoon Guide to Chemistry

A good high school/beginning college level introduction to chemistry, with the trademark Gonick wit and visual felicity. Strong on the physical basis (quantum mechanics, reaction kinetics and stat. mech./thermo.), and a nice view of organic chemistry, but no biochemistry. You'd have to ask a chemist whether it could really replace a textbook.

Warren Ellis and Gianluca Pagliarani, Aetheric Mechanics

A continuation of Ellis's semi-imaginary "Apparat" line, here mashing together all manner of circa-1900 genre fiction.

Eric Rauchway, The Great Depression and the New Deal: A Very Short Introduction

Short, as promised, but within that span wholy admirable. I especially liked the way he managed to keep in view both the global forces which gave us the Depression and sustained it, and the very real pain it produced in individual lives; the sense that every tick-mark in a graph is a hungry mouth, a down-cast pair of eyes. The one thing I would have liked to see more of was the labor movement, but probably everyone else would see other things that could be expanded; perhaps he should just write a very long book on the same topic.

Hopefully, this will not prove to be too topical.

Warren Ellis and J. H. Williams III, Desolation Jones

I like the conceit: Los Angeles is a great open-air reservation for used-up spooks; our protagonist is one of those used-up spooks, and is hired to retrieve (this is not a spoiler) a lost film canister containing Hitler's pornographic home movies, shot in the bunker. (Mercifully, not even Ellis attempts to actually show us these.) There follows a sordid yet compelling plot among sordid yet compelling characters, in which Things Are Not As They Seem.

Books to Read While the Algae Grow in Your Fur; Scientifiction and Fantastica; The Beloved Republic; Writing for Antiquity

Salins's SAT Sokaling

I blame Alan Sokal. The trick of showing up various publications by fooling them into publishing documents which seem impressively technical, but which are obviously nonsense to anyone minimally skilled in the field — well, I thought it was hilarious the first time, but inevitably there are imitators, and they never match the spirit of the first effort.

The latest epigone is one Peter D. Salins, a professor of political science at SUNY Stony Brook and former provost of the SUNY system, and his victim is the editorial page of the New York Times. He purports to offer evidence that the SAT score has some power to predict academic outcomes in college — specifically, whether students will graduate or not — over and above its relationship to high school grades:

In the 1990s, several SUNY campuses chose to raise their admissions standards by requiring higher SAT scores, while others opted to keep them unchanged. With respect to high school grades, all SUNY campuses consider applicants' grade-point averages in decisions, but among the total pool of applicants across the state system, those averages have remained fairly consistent over time.

Thus, by comparing graduation rates at SUNY campuses that raised the SAT admissions bar with those that didn't, we have a controlled experiment of sorts that can fairly conclusively tell us whether SAT scores were accurate predictors of whether a student would get a degree. ...

Among the campuses that raised selectivity, the average incoming student's SAT score increased 4.5 percent (at Cortland) to 13.3 percent (Old Westbury), while high school grade-point averages increased only 2.4 percent to 3.7 percent — a gain in grades almost identical to that at campuses that did not raise their SAT cutoff. Yet when we look at the graduation rates of those incoming classes, we find remarkable improvements at the increasingly selective campuses. These ranged from 10 percent (at Stony Brook, where the six-year graduation rate went to 59.2 percent from 53.8 percent) to 95 percent (at Old Westbury, which went to 35.9 percent from 18.4 percent). Most revealingly, graduation rates actually declined at the seven SUNY campuses that did not raise their cutoffs and whose entering students' SAT scores from 1997 to 2001 were stable or rose only modestly. Even at Binghamton, always the most selective of SUNY's research universities, the graduation rate declined by 2.8 percent.

I submit that Salins has Sokaled the Times, since there is no way someone with enough grasp of social-scientific methods to hold his position could make such huge howlers unintentionally.

Item: The question of interest is at the individual level: given otherwise similar students in the same academic environment, does a higher SAT score predict better academic outcomes, i.e., a higher likelihood of graduation. The data presented, however, are at the institutional level. At best they speak to whether more selective colleges have higher graduation rates, averaging over all students. This is compatible with nearly any relationship whatsoever between SAT scores and graduation rates at the individual level. (Likewise: In every state, rich people are more likely to vote for the Republican party, but richer states are less Republican.) Are we to suppose that Salins doesn't understand that there are different levels of aggregation here, that he has never heard of the ecological fallacy or Simpson's paradox?

Item: This was not a controlled experiment. There was neither actual control of variables other than SAT demands, nor effective control via randomization. The campuses which became more SAT-selective differ in many ways from the ones which didn't, and there is no control for that here. Even when he makes paired comparisons, there are huge differences *. Are we to believe that Salins doesn't know what the phrase "controlled experiment" means?

Item: By Salins's own account, many of the campuses which increased their selectivity, as measured by SAT scores, actually saw their graduation rates decline. The cases Salins mentions are Albany (SAT scores up 1.3%, graduation rate down 2.7%), Oswego (+3% and -1.9%, respectively) and Plattsburgh (+1.3% and -6.3%). Salins's resolution of this apparent contradiction is conspicuous by its absence. It's possible that there is some sort of threshold effect, so that (proportional) increases in SAT scores which fall below that threshold (around 3%, perhaps?) have no or even negative effects on graduation rates, but larger gains raise graduation rates, but this is hardly the case Salins says he is making.

Since we cannot believe that someone in Salins's position is actually writing seriously with so many mistakes and internal contradictions, we are forced to reject the idea that his actual meaning is his apparent meaning. It could be that Salins is engaging in esoteric writing (in the sense of Strauss), but it seems simpler to me to suppose that he was bored, and decided to see if he could get the times to believe that inconclusive noodling is a decisive and boldly contrarian finding.

— For the record, I would actually be a bit surprised if, ceteris paribus, higher SAT scores didn't predict higher likelihood of graduation. (Bad arguments do not become correct because their conclusions are true.) Also for the record, there are sensible ways of doing ecological inference, and of drawing causal inferences from observational data; but what Salins does isn't even close.

(Thanks to Kristina for pointing out the op-ed and discussing it with me.)

*: For instance, he pairs Albany with Stony Brook, because they are both research university campuses. This is true, but they are very different research universities. For one thing, and I say this with all due respect for my colleagues at Albany, Stony Brook has an immensely stronger scholarly reputation, e.g., three Nobel Prize winners on the faculty vs. zero. (Salins may, generously, be trying to reduce Stony Brook's advantage on this score.) For another, Stony Brook is a much nicer place to live. (When I went to Albany last year to give a talk, the campus was plastered with official posters warning students that "Walking alone at night makes you a target". This sort of thing tends to have a discouraging effect on prospective students and their parents.) Now, since we are interested in explaining changes in graduate rates, if these differences between campuses were stable over the period, it'd be harder to see them accounting for that change. (But not impossible; they might modulate how the graduation rate responded to some other factor which did change over the period, e.g., the perceived extra value of attending a higher-prestiege school.) But there's no reason to think that the relevant differences were stable.

Manual track: Vlorbik; Quantum of Wantum; The Greenbelt

Engimas of Chance; Learned Folly; Anticontrarianism

Return of the Cartogram

Four long, long years ago, I helped Michael Gastner and Mark Newman make some cartograms of the last Presidential election, showing areas on the electoral map in proportion to population, not to land-mass. People liked them (though apparently professional cartographers hate them), and I am very happy to relate that Mark has made a new set of cartograms with last night's results:

This map makes me happy, though not nearly as much as the reality.

The Beloved Republic; Engimas of Chance

Election Night

I am fucking crying uncontrollably listening to a speech by a professional politician. And constantly expecting to hear a gunshot.

The Beloved Republic

Books to Read While the Algae Grow in Your Fur, October 2008

Warren Ellis and Juan Jose Ryp, Black Summer

A typically Ellisian sermon on the text "With great power comes great responsibility."

Riccardo Rebonato, Plight of the Fortune Tellers: Why We Need to Manage Financial Risk Differently

This book needs a full review. In lieu of that, while I largely agree with Rebonato's recommendations (to wit, it's crazy to think that anyone can measure financial risk as precisely as current practice thinks it can), his framing of them as somehow deriving from Bayesian subjective probability is wrong, wrong, wrong. This reveals a double failure on the part of the statistical profession: the big failure is that we have allowed, and even encouraged and participated, in the growth of a huge industry premised on ignoring the fact that limited data implies limited precision. The lesser failure is that our teaching is so bad that an intelligent and well-schooled man like Rebonato has never grasped the real point of a confidence interval.

But, like I said, this needs a full review.

Warren Ellis and Ivan Rodriguez, Doktor Sleepless: Engines of Desire

"Where is the future we were promised?" More mad prophets speaking to cities full of humanity transforming itself into something very strange. But Doktor Sleepless is not Spider Jerusalem; he really is mad, bad, and dangerous to know. I want to see where Ellis is taking this, but I know it's nowhere good.

There is, inevitably, a wiki.

Warren Ellis and Jacen Burrows, Scars

Fall of Cthulhu: The Gathering

Simon Oliver, Tony Moore, Ande Parks and Chris Samnee, Exterminators, vols. 2 (Insurgency), 3 (Lies of Our Fathers) and 4 (Crossfire and Collateral)

More of the struggle to keep chaos at bay. The interlude with the mothers in vol. 2 was particularly well-done.

Warren Ellis, Crooked Little Vein

This is a gross, at times nightmarish, book, but in the end it is really (as one of the characters says) about the nature of love; as is Dante's Inferno.

S. M. Stirling, The Sky People

Mind-candy: the first probes to Venus and Mars in 1962 revealed that we live in Edgar Rice Burroughs's solar system, which rather re-directs the Cold War. This is because Something set the solar system up that way, through millions of years of deliberate fiddling. (I refuse to regard this as a spoiler; it's obvious after a few chapters, though not to the characters.) Stirling combines playing the "planetary romance" genre straight — our competent and manly all-American hero woos a blonde Venusian barbarian princess in a fur bikini — while also thinking through the practicalities of things like setting up a bronze age civilization on a planet full of dinosaurs, which someone like Burroughs would never have dreamed about. (Though de Camp would've.) There are, as you might expect, a fair number of scientifictional in-jokes, some of which I found irritating, perhaps because I would've thought they were oh-so-clever when I was 15. Overal, fun; I will be tracking down the sequel.

Books to Read While the Algae Grow in Your Fur; Scientifiction and Fantastica; Enigmas of Chance; The Dismal Science

Supernatural Horror in Electoral Politics

What, you actually thought it was a coincidence that Election Day and Halloween are so close?

On the one hand, Barack Hussein Obama: is he the candidate of a nefarious African conspiracy of cannibalistic pseudo-Christian Muslim witches, or the candidate of Lucifer himself?

On the other hand, Cindy McCain is just like any other female human (via Pandagon).

— Pretty much every agrarian society has a fairly lively belief in witchcraft, of the putting-a-hex-on-your-neighbors-goat-or-genitals variety. Interestingly, however, this does not always translate into its elites actually taking it seriously as a threat. In early medieval Europe, for example, the quite sensible official theory was that, God being infinitely more powerful than the Devil, Christians didn't have to actually worry about magic, and magicians were to be condemned for what they wanted to do, rather than what they could do. The transformation of elite thinking to fearing that magic actually works was a fairly involved and curious process, hinging on the merger of the idea of witches and magicians with the also-very-old stereotype of the organized subversive conspiracy of murderous infidel perverts. (The story is well-told in Norman Cohn's Europe's Inner Demons.) Whether the modern American versions are lineal descendants of intellectualy respectable European demonology, or rather an independent evolution from the same kind of folk beliefs, I don't know. It's also a mystery to me how this complex of fears and myths can have any hold on people who sincerely believe and understand the most elementary tenets of Christianity. (It would make perfect sense if you thought that Satan was another god, and maybe stronger than yours.) But I suspect that is due to my own lack of understanding of moral psychology.

What I can see is that very few ideas like this — ones which, however stupid, speak strongly to our fears — ever really go away. In comparatively happy times, these fantasies may be able to do no more than terrorize the inhabitants of the cultural slums, rather than occupying the center and enacting a reign of blood. But the cultural slums are still where too many of us live, pending deliverance.

Manual trackback: Cognition and Culture

The Beloved Republic; The Running Dogs of Reaction; Psychocermamics

Socialism and the Financial Crisis

Comrade B. Ehrenreich explains all. This astonishing revelation must be read in full; no excerpt can do it justice. (Via comrade A. Swartz, who also explains all.)

The Dismal Science The Progressive Forces

Ghost Peaks, Buried in Ice

I haven't been posting very much, have I? It's felt hard to make the time. Yes, yes, the global financial system is imploding, and the conventional wisdom among the chattering classes is to embrace the policies which brought Herbert Hoover such fame, while the Central Planning Board of the future socialist state may need to make room in its lobby for statues of Paulson and Bernanke beside those to von Mises and von Hayek, and Engels is vindicated as a prophet. (Of course Fox news reassures us that everything is fundamentally sound.) The political question of the moment seems to be whether America is ready to assassinate its first black president*. (With ancillary queries: Can a finite-state automaton — not even a push-down stack machine! — be elected to national office? Can there be anything more un-American than enabling ordinary people to discover their united powers and individual potentials, and is it Maoist or Stalinist un-Americanism, or, even more insidious, rootless cosmopolitanism?) Momentous though they are, it's important to recognize that there are all crises of the day, raised up for a time by Fortune's wheel to vex us, but soon to fall as the wheel turns and others take their place.

A while back some people got their shorts in a twist over turning on the Large Hadron Collider, which would, indeed, have mattered more. (Not that the election doesn't matter.) But that was idiotic; far more energetic reactions routinely happen with cosmic rays in the upper atmosphere, just inconveniently located away from detectors. If they were going to turn the Earth into a black hole or a lump of strange matter, it would've happened already.

No, none of these are causes for lasting concern. (Unlike bugs in Soviet-era pattern-recognition code.) But this has me terrified:

It is perhaps the last great Antarctic expedition - to find an explanation for why there is a great mountain range buried under the White Continent.

The Gamburtsevs match the Alps in scale but no-one has ever seen them because they are covered by up to 4km of ice.

Geologists struggle to understand how such a massif could have formed and persisted in the middle of Antarctica.

Now, an international team is setting out on a deep-field survey to try to get some answers.

The group comprises scientists, engineers, pilots and support staff from the UK, the US, Germany, Australia, China and Japan.

The ambitious nature of the project - working in Antarctica's far interior - has required an exceptional level of co-ordination and co-operation.

..."There are two easy ways to make mountains," explained Dr Robin Bell, from the Lamont-Doherty Earth Observatory, who is a lead US researcher on the expedition. "One is colliding continents, but after they collide they tend to erode; and the last collision was 500-million-plus years ago. They shouldn't be there.

"The other way is a hotspot, [with volcanoes punching through the crust] like in Hawaii; but there's no good evidence for underneath the ice sheet being that hot.

"I like to say it's rather like being an archaeologist and opening up a tomb in a pyramid and finding an astronaut sitting inside. It shouldn't be there."

... The expedition gets under way in the next few weeks and will take some two-and-a-half months to complete.

Space-travelers. In tombs. In an inaccessible, highly anomalous mountain-range in Antartica. Do the fools know nothing? Or do they know only too much?

I am forced into speech because men of science have refused to follow my advice without knowing why. It is altogether against my will that I tell my reasons for opposing this contemplated invasion of the Antarctic — with its vast fossil hunt and its wholesale boring and melting of the ancient ice caps. And I am the more reluctant because my warning may be in vain...

*: I was sure this was an Onion man-in-the-street question, but evidently not; I wish I knew who I stole it from.

Update, 11 November 2008: It just gets worse and worse; the Gamburtsevs appear to be both as tall as the Alps, and 500 million years old. These are things that should not be (geophysically).

At least the continent of our doom is beautiful and sublime...

Manual trackback: Inverse Square; Brad DeLong (I believe his title translates as "In his office on Frew Street, the author procrastinates on grant proposals")

Cthulhiana; The Continuing Crises; The Beloved Republic; The Dismal Science

Books to Read While the Algae Grow in Your Fur, September 2008

Donna Kossy, Strange Creations: Aberrant Ideas of Human Origins from Ancient Astronauts to Aquatic Apes

Donna Kossy is one of the best writers about weird beliefs around. (Go read Kooks right now if you haven't.) This is a look some ideas about where humanity came from, and where it might go, which are currently on the fringes of respectability at best, although in some cases (creationism, eugenics) they didn't use to be. She's particularly good, here, on the influence of Theosophy. As always, Kossy tries very hard to be sympathetic to those she reports on, even in cases where that's obviously very hard (like eugenicists). Her readers, however, are under no such obligations.

(She misses the Lovecraft/ancient astronaut connection, however; in a sense the Cthulhu Mythos is what you get when you take Blavatsky's elaborate scheme of pre-human, material-life-but-not-as-we-know-it races, and add the assumption that aliens will be alien.)

Kage Baker, The House of the Stag

Another intricately-plotted fantasy novel, including a lengthy section where Epic Fantasy is depicted as a kind of harlequinade. Also, a portrait of the saint as a young woman.

— Brad DeLong has posted a long excerpt of one of the more amusing scenes.

Warren Ellis, Chris Weston and Laura Martin, Ministry of Space

Alternate history of the British conquest of space, aided by Wernher "we aim for the stars, and sometimes we hit London" von Braun. The technology thus closesly follows von Braun's actual proposals.

Mat Johnson and Warren Pleece, Incognegro: A Graphic Mystery

Historical mystery about lynching and passing.

Denis Bosq and Delphine Blanke, Inference and Prediction in Large Dimensions

Some aspects of prediction problems in the presence of either infinite-dimensional data (such as continuous processes observed over finite intervals of time), or infinite-dimensional parameters (such as arbitrary smooth probability distributions for noise), or both. The general setting is that of mixing processes; the major tools are (1) limit theorems for mixing processes, especially exponential inequalities; (2) kernel estimators (as in Bosq's earlier book on Nonparametric Statistics for Stochastic Processes); and (3) projection estimators, i.e., projecting the data, suitably massaged, on to a countable orthonormal basis in an appropriate Hilbert space, and adaptively truncating small components. The authors refer to all such basis decompositions as "Fourier analysis", whether the basis functions are trigonometric functions or not. Much of this comes together in the last few chapters, on "functional linear processes", i.e., linear processes taking values in infinite-dimensional function spaces, such as those of continuous curves. As is often the case, by moving to Hilbert space one can linearize an otherwise non-linear time-evolution, and so do a lot more than you'd think with autoregressive models.

This is very much a book for advanced students; familiarity with the measure-theoretic treatment of stochastic processes is essential, as is a good grasp of ordinary, finite-dimensional estimation theory, and of course familiarity with Fourier analysis and Hilbert space methods. (Knowledge of ordinary time-series methods is not so much a requirement.) The exposition is very dense and sometimes confusing, though only one section is actually impenetrable. (This is the tangential one on Blackwell's method for forecasting the probabilities of arbitrary events; it's a hard topic, but the discussion in Prediction, Learning and Games really is infinitely clearer.) Many of the results are either new to this book, or might as well be new.

Scott Lynch, The Lies of Locke Lamora and Red Seas Under Red Skies

Mind-candy: a weird yet very entertaining hybrid of fantasy and crime-caper novel.

Books to Read While the Algae Grow in Your Fur; Scientifiction and Fantastica; Enigmas of Chance; Psychoceramics; Creationism

Books to Read While the Algae Grow in Your Fur, August 2008

Patricia Briggs, Cry Wolf

Werewolf-flavored mind-candy.

Peter Goin, Humanature

Photos of "the earth as transformed by human action" (mines, kudzu, over-grown ruins, straightened rivers, artificial wetlands, flood control systems, zoo habitats carefully constructed to look untouched by human hands), largely in the south. Tedious introduction (by the photographer) tries to get beyond the pristine wilderness vs. totally spoilt dichotomy, but ultimately I think fails. (Cf.) Worth it for the pictures.

Kat Richardson, Poltergeist

Sequel to Greywalker. Continuing adventures of a Seattle PI who finds herself, much against her will, a shaman (though she doesn't call it that); this time investigating Parapsychology Gone Horribly Awry.

Karin Slaughter, Fractured

Typically engrossing, and squickening, crime fiction. Sequel to Triptych, but works as a stand-alone as well.

Walter Jon Williams, Implied Spaces

Post-singularity struggles over the future of humanity and the nature of the universe, with sword-fights, poetry, and talking cats. Not one of Williams's most emotionally intense books, but definitely one of his most enjoyable, which is really saying something. (I suspect the lack of emotional depth is, itself, a reflection of the deliberate superficiality of the view-point character, part of his way of coping with being a very, very, very old man in a strange and fluid world.)

Samuel A. Goudsmit, Alsos

Tracking down the German atomic bomb effort, just behind the advancing Allied armies. His conclusions — that the Germans never got very far, but that this was entirely due to their being on the wrong track technically, and complacement about their superiority, rather than a deliberate humanitarian effort, as e.g. Heisenberg liked to imply — appear to be entirely correct. Also includes broader thoughts about the organization of the German war-research effort.

Eric Rauchway, Blessed Among Nations: How the World Made America

How 19th-century globalization let the US develop into an economic juggernaut with an usually weak and incapable central government, and the difficulties this caused when that globalization collapsed during the First World War, leaving us in charge. No other country worked like us because no other country had, or has, our position in the global flows of goods, money and people.

(As a methodological point, Rauchway seems to find it unproblematic that a certain set of institutions should form Back In The Day, when they fit conditions (or at least fit-well-enough-for-the-purposes-of-the-powerful) and then tend to survive later, when they did not fit so well. But I would like some explanation of why adaptive processes had an easier time working in the earlier period, as opposed to the later one. Or perhaps this is just an effect of historical foreshortening, that there were lags and mis-fits as the focal institutions were established and supplanted earlier ones, but this is shaded off into the past, and more time is spent on the end of the period, when new mis-fits developed.)

Update, 25 September 2008: Eric was gracious enough to post a reply.

Jonathan Israel, The Dutch Republic: Its Rise, Greatness, and Fall, 1477--1806

An attempt at a total history of the country which, perhaps more than any other, was the furnace in which modernity was forged, embracing political, military, social, economic, cultural, artistic, theological and scientific developments and their inter-relations. (British readers may find the description of the Glorious Revolution of 1688 as the last successful invasion of the British Isles a bit hard to swallow, but facts are facts.) Surprisingly readable despite running to 1100 pages.

Stephen Murdoch, IQ: A Smart History of a Failed Idea

Popular-science debunking. Not deep, but plenty deep. enough. (Of course, I would think that.)

Neil Gaiman and Michael Zulli, Creatures of the Night

Karin Slaughter, Beyond Reach

Meth-heads! Skinheads! Lies! Outstandingly gruesome murders! (I can't think of anything more to say that isn't full of spoilers: the ending involves sudden killing off a highly-sympathetic major character. This comes across like a surprise kick to the gut. While in retrospect I appreciate the reasons for this — see Slaughter's own explanation — part of me is still going "Noooo!!!", which is, of course, exactly effect she was aiming at.)

Dale Furutani, Death at the Crossroads

Historical mystery/Kurosawa homage.

Books to Read While the Algae Grow in Your Fur; IQ; Physics; Scientifiction and Fantastica; The Beloved Republic; The Dismal Science; Writing for Antiquity

Fire, Metal, Form

For almost as long as I can remember — at least since seeing the ancient Chinese bronzes at the Sackler as a teenager — I've wanted to learn metal-casting. To my great good fortune, Carly Jean Parrish and Ed Parrish (of hot metal happening) offered a class on iron-casting for beginners this summer at the Pittsburgh Center for the Arts near my house, which I've been taking. This Saturday the class cast our molds. (Here's mine, before and after.) It was one of the most awesome things I've ever seen or been part of; I took a few pictures, but really you needed to be there.

Fortunately, our teachers will be performing on Saturday the 23rd at the PCA, as part of the closing of the biennial exhibit; you will not find a better entertainment value in the city that evening.

(Of course, lots of the things which went towards making it an incredible experience as an occasional spectacle — the heat of working under layers of protective clothing, the noise of the furnace, the hammers, the yelling, the flames, the muscle-tension of carrying and controlling a big bucket of molten metal, the adrenaline shock of noticing that your glove doesn't quite meet your sleeve around your wrist and there are sparks going everywhere — are also the things which go towards making iron-working a very unpleasant job. As for doing it twelve hours a day, six days a week, with periodic 24-hour shifts and no insurance, the way Mr. Carnegie and co. used to run the plants — well, there was a reason a visitor called old Pittsburgh "hell with the lid off", and they had to keep the unions down by shooting people.)

Update, 17 October: I somehow forgot to post a link to the actual product. I am a bit disappointed that the metal which should have been the bottom of the vessel fell to about its lip — I need to form the negative mold better next time — but not so disappointed that I'll stop; and I'll keep this as, you should pardon the expression, an object lesson.

Self-Centered; Heard About Pittsburgh PA

Books to Read While the Algae Grow in Your Fur, July 2008

Nicolò Cesa-Bianchi and Gábor Lugosi, Prediction, Learning, and Games

A wonderful synthesis of the literature on competitive, individual-sequence forecasting with expert advice. That is, the problems considered are all variants on a situation where you need to make a prediction about the future (or more generally take an action whose consequences will only be revealed in the future), have access to a range of "experts" or forecasting algorithms, and want to ensure that, no matter what actually happens, your performance will be close to that of the best expert. This is thus a study of sequential decision-making under uncertainty without probability. Often, but not always, the solution lies in taking weighted averages of the experts, giving more weight to those which have done well in the past. This works not because past performance provides any kind of inductive evidence of future success, but merely because it keeps your predictions from drifting too far from what is, in fact, working. (Perversely, many of the proofs rely on probabilistic arguments, but they don't make probabilistic assumptions.) Of course, it may be that even the best expert is very bad, but the possibility of improving on the experts is not really considered, though it's certainly possible (at least with convex loss functions).

Anyone at all interested in machine learning, forecasting, information, game theory, or decision-making under uncertainty needs to read this. It may also be useful to epistemologists (cf.).

Max Brooks, World War Z: An Oral History of the Zombie War

An oral history (a la Studs Terkel) of the early 21st-century global struggle against the zombie apocalypse (a la George Romero). This is a happy choice of form, because it lets him tell the story of a global disaster from many viewpoints, without taking the space which would be required in a conventional cast-of-thousands novel. Also, he gets to tell lots of variously creepy, horrifying, thrilling, and/or moving stories this way.

Query: does this qualify as a "modern epic", sensu Moretti?

Marc Van De Mieroop, The Ancient Mesopotamian City

As every school-child knows, "history begins at Sumer", with the first cities and the first writing. This book is the only accessible synoptic view of the cities of ancient Mesopotamia as such.

After opening by quoting some jaw-droppingly ignorant (and recent) remarks by classicists on how there were no real cities before the Greeks, Van De Mieroop describes the geographic scene, and lays out some of the limitations on our evidence — peculiarities in what scribes thought worth recording, and other peculiarities in what archaeologists have thought worth excavating. Next he considers theories of the origins of cities in Mesopotamia, a peculiarly difficult problem since there were no other cities to learn from or be influenced by. He favors the idea that they originated around the temples, which acted as institutions for redistributing the products of multiple ecological regions, but he is fair to other ideas. (He is even fair to Jane Jacobs's wacky idea that cities preceded, and caused, agriculture, which is to say he does some simple calculations to show it makes no sense whatsoever.) He then goes on to consider social organization, leading institutions like the palace and the temple, the hints of self-government among city-dwellers and their growth over time, the relations between cities and their agricultural hinterlands, how food moved into the cities, long-distance trade, credit and finance, and cities as centers of religion and learning, including divination and astronomy. (He says scribes were taught "calculus", presumably meaning "calculation".) He quotes frequently from Mesopotamian documents, without any philological apparatus, and despite a ritual rejection of strict "positivism", he is very cautious in advancing hypotheses, and very good about marking conjectures as such, and emphasizing that we simply have little or no evidence about many matters.

Mesopotamian history is usually considered to last from the first writing around -3100 to the Macedonian conquests around -300. As Van De Mieroop says, this period of 2800 years is longer than the interval separating us from Homer. It is an astonishing act of hubris, or at least of abstraction, to try to summarize the features of all cities over such a period, even in a restricted region — one can only presume that there must have been extensive variation. Nonetheless, Van De Mieroop does a really remarkable job.

Lucy Snyder, Installing Linux on a Dead Badger

You remember "Installing Linux on a Dead Badger", don't you? Well, imagine a full hundred page book of such stories. C'mon, you know you want it.

Jack Campbell, Valiant

Continuing science-fictional anabasis; see here for previous installments.

Charles E. Lindblom and David K. Cohen Usable Knowledge: Social Science and Social Problem Solving

This is a short (100 pp.) book from 1979, largely given over to sketches of arguments and directions for further inquiry (mostly not undertaken in the ensuing years) about why the social sciences, and "professional social inquiry" more generally, have not been very directly useful for social problem solving. They suggest that this rests on a number of basic widespread mistakes about how social problems are solved. In particular, they allege, social scientists vastly under-rate the importance and competence of ordinary-life social knowledge, and, yet more consequentially, fail to see that social problems can be solved either by analyzing them in some discursive/analytic form, or by setting up patterns of social interaction where the participants' acts collectively solve the problem, though none of them need to grasp the solution or even realize that is what they are doing. Markets are of course one example of such "interactive problem solving", but they also, and quite correctly, emphasize others: democratic politics, bargaining processes, and the "republic of science". They emphasize that interactive problem-solving should not be seen as a poor substitute for formal problem solving, to be displaced in due time by scientifically-informed social engineering, but rather as inevitable, and indeed often superior.

The alternative, of analytically finding solutions to social problems, is basically impossible, because the problems are too complex, and even systematic investigation into them is not just prohibitively expensive, but so slow that the world has moved on before research findings can become very accurate or precise. (Obviously these obstacles can all be bigger or smaller in various cases, and I don't think they'd quibble if someone wanted to assert that very small, stable social problems could be successfully analyzed if enough resources were thrown at them.) Worse, the very definitions of "social problems" are themselves contested, and properly so. The authors' view is that while the natural sciences can (often) legitimately claim independent authority, for social scientists to aim at such authority is to set themselves a target they cannot possibly hit. Since they do aim at that target, however, social scientists and other "practitioners of professional social inquiry" systematically waste their efforts.

Given all this, fruitful roles for social analysts become things like advising individual participants in the interactions, or looking at the over-all performance of an interactive mechanism and searching for ways in which it might be improved. (They suggest that economists are better about this than other social scientists. Given the recent vogue among economists for replacing all kinds of institutions with arbitrary intellectual constructions, planned by analogy with the idealized markets of their Micro 1 textbooks, I suspect the authors might wish to revise and extend these remarks.) A further, if more diffusive, constructive role would be in hoping to shape the general framework within which participants in interactive problem-solving think about things; and of course the kind of detailed reportage which statistical bureaus engage in.

The name "Hayek" does not appear anywhere in this book.

Douglas W. Tallamy, Bringing Nature Home: How Native Plants Sustain Wildlife in Our Gardens

This is a rather straight-forward argument for North American suburbanites (and urbanites, too) to plant more native plants, roughly as follows. (1) We like having wildlife such as birds and (some) mammals around. (2) These animals sit high in the food-web; below them are many insects, especially insect larvae. (3) Most insects are specialized to eat only a small range of plants with which they have co-evolved, in no small part because of the chemical defenses evolved by plants. (4) Thus, native plants support a much larger and much more diverse population of insects than do introduced ones. (5) The process of evolutionary adaptation is very slow, and even plants introduced almost 500 years ago are still substantially less good as insect hosts than natives. (6) Suburbia occupies such a huge part of the American landscape that if native plants are to thrive anywhere, it has to be there. Therefore, (7) suburbanites should shift what they plant towards natives.

Obviously, the key empirical parts of this argument are (4) and (5); here the evidence that Tallamy presents is good, but — he is admirably up-front about this — not conclusive, and he is happy to admit that there are some species of introduced plants which are so closely related to natives that bugs like them just fine. Unlike many books on native plants, this is empirical, consequentialist, modest, and un-mystical.

David Pollard, Empirical Processes: Theory and Applications (full text free online)

The simplest sort of empirical process arises when trying to estimate a probability distribution from sample data. The difference between the empirical distribution function F_n(x) and the true distribution function F(x) converges to zero everywhere (by the law of large numbers), and — this is non-trivial — the maximum difference between the empirical and true distribution functions converges to zero, too (by the Glivenko-Cantelli theorem, a uniform law of large numbers). The "empirical process" E_n(x) is the re-scaled difference, n^1/2[F_n(x) - F(x)], and it converges to a Gaussian stochastic process that only depends on the true distribution (by the functional central limit theorem). Empirical process theory is concerned with generalizing this sort of material to other stochastic processes determined by random samples, and indexed by infinite classes (like the real line, or the class of all Borel sets on the line, or some space parameterizing a regression model). The typical objects of concern are proving uniform limit theorems, and with establishing distributional limits. (For instance, one might one want to prove that the errors of all possible regression models in some class will come close to their expected errors, so that maximum-likelihood or least-squares estimation is consistent. [For more on that line of thought, see Sara van de Geer's book on Empirical Processes in M-Estimation.]) This endeavor is closely linked to Vapnik-Chervonenkis-style learning theory, and in fact one can see VC theory as an application of empirical process theory. (I'd guess Vapnik himself would disagree with that, however.)

This short book by Pollard is an introduction to empirical process theory by a statistician for statisticians. As such it succeeds admirably; as always, Pollard does a really good job of explaining what the technical apparatus is doing and why it takes the form it does. People coming to it from other backgrounds (I am particularly thinking of computer scientists) will probably find it harder going, not least because the implied reader has an extremely sure grasp on measure theory. (Such as one might acquire from, oh, Pollard's User's Guide to Measure-Theoretic Probability.) If you can handle Pollard's 1989 survey paper, then you will probably enjoy this book; and if not, not. The applications he describes are all interesting and challenging, though I was a bit disappointd that none of them involve dependent data.

Clubbing

Fall of Cthulhu: The Fugue

Scalped: Indian Country

Proof: Goatsucker

Various flavors of comic-book mind-candy.

Jenny Davidson, The Explosionist

It is impossible to describe this better than the author, so I'll steal her words:

the story of a 15-year-old girl growing up in an alternate version of 1930s Edinburgh, one where the legacy of Napoleon's victory a century earlier at Waterloo is a standoff between a totalitarian Federation of European States and a group of independent northern countries called the New Hanseatic League. This world is preoccupied with technology (everything from electric cookers to high explosives) but also with spiritualism, a movement our world largely abandoned in the early twentieth century; Sigmund Freud is a radio talk-show crank, cars run on hydrogen and the most prominent scientists experiment with new ways of contacting the dead.

My biggest complaint with this book is that it ends in the middle of the story, and nothing warns the reader about this. Grrrr.

Charles Tilly, Democracy

An attempt to explain the mechanisms by which states come to engage in "broad, equal, protected, mutually-binding consultation" with their citizens, singling out three especially important processes: (1) integrating "trust networks" into public politics, (2) screening off public politics from categorical forms of inequality, and (3) suppressing non-state centers of coercive power. As usual with Tilly, he draws on a huge range of historical sources, in an impressive display of erudition and clear thinking. Also as usual with Tilly, one does not get a comprehensive theory, but perhaps this is the sort of material where such a theory isn't really possible, and the best one can hope for is a catalog of recurring mechanisms.

Charles Tilly, Big Structures, Large Processes, Huge Comparisons

A brisk debunking of pernicious ideas about how societies work and change that we have inherited from the 19th century, together with a smart and enthusiastic brief for comparative, historical social science. (The description at Powell's is definitely for another book!)

Thanks to Doug White for lending me his copy.

Books to Read While the Algae Grow in Your Fur; Enigmas of Chance; Biology; Scientifiction and Fantastica; The Dismal Science; Cthulhiana; Writing for Antiquity

In Memoriam Abdussattar Shalizi

My grandfather passed away at his home in Kabul on July 4th. He was, most likely, 92.

Nothing I could say right now would be adequate.

Books to Read While the Algae Grow in Your Fur, June 2008

James R. Flynn, What Is Intelligence? Beyond the Flynn Effect

A full review will be forthcoming in a magazine, so I don't want to spoil that, but will just hit a few key notes.

(1) This is mostly about explaining the large, long-term, world-wide rise in IQ, a.k.a. the Flynn Effect (not so named by Flynn!), and especially the fact that parts of IQ tests show different rates of gain, or none at all, with no particular reference to their correlation with the (IMSAO mythical) "general factor of intelligence" g.

(2) Flynn's preferred explanation is much closer to mine than I would have guessed before reading this: it has to do with the cultural diffusion of new habits of thinking, what he calls "putting on scientific spectacles". I do not think he follows this argument as far as he could, or should. (I will elaborate on that in the full review.)

(3) A reasonable chunk of this already short book consists of Flynn's animadversions on ethical relativism, postmodernism, etc. I like seeing these bashed as much as anyone else who thought "Transgressing the Boundaries" was the funniest thing ever, but here it's a tangent on a tangent, and Flynn doesn't even bash them especially well; his editor shouldn't have let him indulge himself to this extent.

(4) Flynn is clearly trying to write for a general audience, but I am not sure that someone who reads, e.g., his account of the model he and Dickens devised of "social amplification" would understand it, unless they had read and grasped the relevant papers first.

— On re-reading this seems more negative than my actual opinion. It's definitely worth reading if you care at all about the IQ controversy; it's probably not so helful as a first exposure to that subject.

Roger Th. A. J. Leenders, Structure and Influence: Statistical Models for the Dynamics of Actor Attributes, Network Structure and Their Interdependence

(In lieu of a full review): Two extremely important phenomena in social networks are that (1) people don't make social ties randomly, but tend to link up with others who are either similar to them in some salient way, or to whom they are complemntary; and (2) people learn from and imitate each other. This creates a very serious inferential problem: when we observe that neighbors in a social network are more similar than random members of the population, is that because being linked made them similar, or did they link because of pre-existing similarities? (In the jargon, there is confounding between homophily and contagion.) Leenders's book, a revision of his dissertation, is a first attempt at prising this appart, by using models which represent both how people might influence each other and how they might decide on who to interact with. It is straightforward but worthy stuff, and I can think of a number of high-profile recent papers whose authors — or, better yet, referees — should be whacked over the head with this. (It's only a 250 pp. paperback so that's not that bad.)

The writing is exactly as bad as you'd expect from a doctoral dissertation in mathematical sociology. Nonetheless, I strongly recommend it for the collection of anyone seriously interested in social networks or dynamic network analysis.

C. J. Sansom, Dark Fire

Historical mystery, sequel to Dissolution. This time our hero has the misfortune to get mixed up in a complicated plot involving alchemists, royal marital dissatisfaction, and politico-theological disputes. Manages to mater-of-factly convey the awfulness and alienness of Tudor England, without slipping into the trap of making the narrator a modern man on the inside.

Jane Haddam, Cheating at Solitaire

Haddam takes on Martha's Vineyard Margaret's Harbor and the culture of celebrity, with special reference to pop tarts. It would have been easy, in the interest of entertainment, to make many of the characters completely unsympathetic; she doesn't. And the mystery was baffling, at any rate to me.

Books to Read While the Algae Grow in Your Fur; Enigmas of Chance; IQ; Networks

Short Story Sunday Reading

Three for your delectation:

• K. J. Bishop, "The Art of Dying" [Featuring the return of a character from The Etched City, which you should go read right now if you haven't. When did The Boston Review start publishing fantastica?]
• David Moles, "Long Past Midnight"
• Catherynne M. Valente, "A Buyer's Guide to Maps of Antarctica"

Scientifiction and Fantastica

An Undertaking of Great Advantage, But Nobody to Know What It Is

Shorter FBI to Congress:

Our "National Security Analysis Center" data mining project is too important and super-secret to explain to mere legitimate authority, but it's expanding exponentially, so the money you give us for it had better grow too.

As Noah Shachtman notes, the remarkable thing is that this did, in fact, lead to the House appropriations committee voting to at least not expand the program. (They did not, apparently, vote to kill it altogether, though I can't see why not.)

(Speaking of wasteful, fradulent, and/or abusive data-mining...)

Manual Trackback: Earning My Turns

The Continuing Crises

Chris Anderson: Aware of All Statistical Traditions (with bonus fall course announcement)

Attention conservation notice: Someone is wrong in Wired magazine.

I recently made the mistake of trying to kill some waiting-room time with Wired. (Yes, I should know better.) The cover story was a piece by editor Chris Anderson, about how having lots of data means we can just look for correlations by data mining, and drop the scientific method in favor of statistical learning algorithms. Now, I work on model discovery, but this struck me as so thoroughly, and characteristically, foolish — "saucy, ignorant contrarianism", indeed — that I thought I was going to have to write a post picking it apart. Fortunately, Fernando Pereira (who actually knows something about machine learning) has said, crisply, what needs to be said about this. I hope he won't mind (or charge me) if I quote him at length:

I like big data as much as the next guy, but this is deeply confused. Where does Anderson think those statistical algorithms come from? Without constraints in the underlying statistical models, those "patterns" would be mere coincidences. Those computational biology methods Anderson gushes over all depend on statistical models of the genome and of evolutionary relationships.

Those large-scale statistical models are different from more familiar deterministic causal models (or from parametric statistical models) because they do not specify the exact form of observable relationships as functions of a small number of parameters, but instead they set constraints on the set of hypotheses that might account for the observed data. But without well-chosen constraints — from scientific theories — all that number crunching will just memorize the experimental data.

I might add that anyone who thinks the power of data mining will let them write a spam filter without understanding linguistic structure deserves the in-box they'll get; and that anyone who thinks they can overcome these obstacles by chanting "Bayes, Bayes, Bayes", without also employing exactly the kind of constraints Pereira mentions, is simply ignorant of the relevant probability theory.

By coincidence, I am going to teach our data mining course (36-350) again in the fall. The theme for the semester, which I decided on back in the spring, will be "waste, fraud and abuse" — not so much detecting suspicious activity, though some examples of that might be fun, as warnings against wasteful, fraudulent and/or abusive data mining.

Update, 29 June: see next post.

Update, 2 July: A correspondent writes to let me know that Anderson's essay and the linked pieces from Wired are up at Edge.org, along with responses from some of the other clients of John Brockman's literary agency leading public intellectuals associated with that site. So far, the only one whose reaction is both substantial and not completely clueless is Danny Hillis, who politely says that Anderson's idea does not have "even a little bit of truth in it".

There's no reason we couldn't have an interesting public discussion about what big data, and data-mining, could contribute to science. We already have a very large and successful scientific discipline which routinely generates and deals with petabytes of data, namely experimental high-energy physics. Its example suggests that theory becomes more rather than less important with huge volumes of data. That may not hold for the biological and social sciences, but I'd like some argument as to why. Of course, if one looks at actually-existing quantitative models in those sciences, it seems clear that part of what they are doing is representing scientists' substantive knowledge and/or guesses, but another part is just put in for tractability, especially statistical tractability — linear or logistic dependence, Gaussian noise, etc., etc. One of the things modern statistics and big data could do is to drastically weaken those tractability constraints. (To repeat a slogan from my class, "More science, fewer t-tests.")

We could have a conversation about these matters. But its participants would have to know something about scientific practice, about statistics and about data-mining. Some of these participants might even argue quite strongly that discovery can be automated, if one goes about it the right way. If someone — say, a literary agent and impresario whose client list includes just about every well-known popular science writer in America — wanted to organize such a discussion, it would certainly be possible and a contribution to public enlightenment. That would, however, require such impresarios to have somewhat more critical acumen than a puppy, which evidently is not the case. So the actually-existing conversation is a source not of light but of noise.

Why oh why can't we have a better consciousness industry?

Manual trackback: Entertaining Research; Tongue but no door; O Hermenauta; Whimsley; Quantum of Wantum; The Statistical Mechanic; Lies and Stats; sciber

Enigmas of Chance; Corrupting the Young

Books to Read While the Algae Grow in Your Fur, May 2008

Stan Washburn, A Moral Alphabet of Vice and Folly: Embellished with Nudes and Other Exemplary Materials

A series of etchings by "California's foremost sixteenth century artist", accompanying short, mordant little fables. For example, under "P": "A Philosopher concluded that man's pretentions are absurd, and that worldly endeavor is without purpose. So assiduous were his ruminations on this insight that he neglected to publish, and in due course he perished. Moral: publish."

(Thanks to Carl Worth for introducing me to this book.)

John Sutton, Marshall's Tendencies: What Can Economists Know?

A wonderful little book about how economists do, and (what is not quite the same) should confront models with empirical data. Along the way he discusses the history of econometrics, the theory of the tides, option pricing, the origins of thermodynamics, the price of taxi-cabs in San Diego, how to bid for oil rights, etc., etc. It's a wonderful performance in only about a hundred pages, and requires no technical knowledge of econometrics or game theory, though some would probably help. Much of what he says would apply, mutatis mutandis, to any social or natural science, though many of them will not have such strong convictions about the form models should take.

Timur Kuran, Islam and Mammon: The Economic Predicaments of Islamism

This is a collection of essays on self-proclaimed "Islamic economics" and "Islamic banking", along with a final one on the causes of the comparative economic backwardness of the Islamic world (by which Kuran means the Middle East and not e.g. India or southeast Asia), as compared to western Europe.

The key point of the essays on Islamic economics and Islamic banking is sound: these are recent ideological creations, entirely a product of the last half century, and their goal is expressive, rather than actually being concerned with understanding or improving economies. Islamic banking, in particular, while it purports to avoid interest-based loans in favor of risk-sharing, in fact does so only through transparent dodges, and that for very good reasons. (It is, in fact, not at all clear that the Qu'ran prohibits interest as such, as opposed to certain usurious practices.) Kuran also speculates, plausibly but entirely without evidnece, that the creation of an "Islamic sub-economy" of banks, manufacturing and service firms, grocers, etc., allows up-and-coming emigrants to the great cities of the Islamic world to find networks of reasonably trust-worthy peers in similar situations, since they are shut out of the existing elite networks.

These points are made in most of the essays here, many times over, and highly repetitiously. They are also marred by what I can only call a very strange ethical scheme. Kuran claims, fairly enough, to be a Hayekian, and so what worries him about Islamism is whether it might be redistributive, or mess with the rights of property owners. To quote from p. 68,

Modern Islamist movements possess, then, the ideological capacity and flexibility to sustain a liberal economic agenda. Even if they promote illiberal policies while in opposition, they may be able to assume a liberal orientation once in power. In any case, to pursue effectively liberal policies they need not make deliberate or explicit ideological adaptations. By giving low priority to economic issues, they may end up promoting private investment, self-management, private ownership, and free trade by default. Such unintended liberalism is all the more likely where illiberal economic goals are overshadowed by objectives concerning family, sexuality, manners, and education. Though a prominent theme in Khomeini's pre-revolutionary rhetoric was the elimination of poverty and exploitation, once he rose to Iran's helm he subordinated his stated economic objectives to the general goal of restoring the centrality of Islam in public life &emdash; even to such particular objectives as eliminating the consumption of alcohol, veiling women, banning Western music, and severing Iran from its pre-Islamic heritage. After the revolution, he dismissed demands for concrete economic reforms on the ground that economic well-being is worthy of the donkey.

To put this in "shorter" form: "Sure, the new regime is using the coercive power of the state to impose a single scheme of values on all citizens, censoring all forms of expression, and forcibly subjecting half the people to lives of grossly restricted choices, but they're against rent control, the minimum wage, industrial policy and tarriffs — liberal values are safe!"

The last chapter tries to construct an explanation for the economic backwardness of the Islamic Middle East, drawing on Kuran's truly impressive book on preference falsification. This is extremely unpersuasive. The fact that Kuran sets out to explain is that, while this part of the world used to be on an economic level with western Europe, or even more advanced, it is no longer and has not been for some centuries. However, this is not just true of the Middle East but also of India, China, Japan, etc. By constructing an explanation which could, at most, apply to one of those cases, he is asking us to believe in a remarkable coincidence... The alternative, which he mentions but doesn't really address, is that something very peculiar started happening in one place, namely western Europe, and that this explains the relative decline of the rest of the world. (Don't get me started on his cross-country regression of growth rates.)

To sum up, Kuran has some important points to make, but also some truly remarkable lapses and blind-spots. His book is short (147 pages + notes) yet highly repetitive; it should have been even shorter. It's worth reading only if you are extremely interested in the subject; worth buying only for specialists.

Fritz Stern, The Politics of Cultural Despair: A Study in the Rise of the Germanic Ideology

Studies in the lives and works of three intellectuals who helped, each in their own way, to lay the road to ruin: Paul de Lagarde, Julius Langbehn and Arthur Moeller van den Bruck. Good at not just seeing them in that light, and at seeing that many of their trends they participated in — for example, a driving hatred of liberalism — were not just German. (A nice study could be done comparing these ideologues' idea of "the West", home of "civilization", as opposed to German "culture", to the idea of decadent, socialist Europe among contemporary American conservatives.) The contrast cases Stern seems to have in mind, but who mostly show up in his footnotes and asides rather than his main text, are Nietzsche and Thomas Mann, who shared (at least for a time, in Mann's case) these writers' pretense of being "unpolitical", but with vastly more sense and vastly less moral depravity.

Robert Pinsky, Gulf Music: Poems

The poet wrestling with things, disasters, and his own "insomniac monkey-mind". I'll quote the last part of "First Things to Hand", titled "Door", because it is nice, but it's not representative of the range. (No one poem here is.)

The cat cries for me from the other side.
It is beyond her to work this device
That I open and cross and close
With such ease when I mean to work.

Its four panels form a cross—the rood,
Impaling gatepost of redempton.
The rod, a dividing pike or pale
Mounted and hinged to swing between

One way or place and another, meow.
Between the January vulva of birth
And the January of death's door
There are so many to negotiate,

Closed or flung open or ajar, valves
Of attention. O kitty If the doors
Of perception were cleansed
All things would appear as they are,

Infinite. Come in, darling, drowse
Comfortably near my feet, I will click
The barrier closed again behind you, O
Sister will, fellow mortal, here we are.

Andrea Camilleri, The Patience of the Spider and The Paper Moon

Wonderful as always; Montalbano continues to be a superb detective, and the tone of outrage at injustice and astonishment at human depravity and folly is nicely balanced with self-mockery (the scene with the alarm clock at the beginning of The Paper Moon, for example) and good food. (This I think distinguishes itself from American hard-boiled crime stories, which seem to take themselves and their disillusionment so seriously.) Previous installments discussed here, here, here, here and, most recently, here. — Many thanks to "Uncle Jan" for copies!

George R. Milner, The Moundbuilders: Ancient Peoples of Eastern North America

Over-view of the archaeology of the pre-historic inhabitants of what is now the eastern US, with a little bit of Ontario thrown in, emphasizing the mound-building cultures of the mid-west and south-east. Milner seems somewhat more confident in some of his statements (e.g., about artifacts circulating by gift exchange rather than trade, or about social organization) than the evidence he presents would seem to warrant, but then I often have this problem when reading archaeologists.

Jessica Hagy, Indexed

Fun with Venn diagrams and little graphs on two axes. The effect is a little hard to describe, but fortunately you can just see examples.

David Rees, Get Your War On II

Reading this in 2008 brings back, in a truly vivid way, just how much of a feverish nightmare 2002--2004 really was.

Books to Read While the Algae Grow in Your Fur; Islam; The Dismal Science; Enigmas of Chance; The Running Dogs of Reaction; Writing for Antiquity; The Commonwealth of Letters; The Continuing Crisis

A Note to My Alma Mater: John Yoo Is Not a Campus Treasure

Dear UC Berkeley: I understand you have reasons, more or less good, for not firing John Yoo immediately. But there is no call to put puff-pieces about him in the magazine you send to donors. This is not making me any more likely to contribute. I'm not asking for displays of public repentance in publications like this, though that would be nice; just not rubbing my nose in the fact that I got my degree from a school which has as one of its faculty a man who helped make us a nation of torturers. Sincerely yours, Cosma Shalizi ('93).

Update: see also Marty Lederman at Balkinization.

The Running Dogs of Reaction; Learned Folly; The Continuing Crises

End-of-Semester Inventory

New classes taught alone: 1
New classes co-taught: 1

Thesis committees I was on at beginning of semester: 5
Number of those students who successfully defended their dissertations this semester: 1 (congratulations, Dr. Damouras!)
Number of thesis committees I joined: 2
Co-supervised Ph.D. students who successfully proposed and are now ABD: 2 (yay, Linqiao and Justin!)
Undergrad RAs going on to graduate school: 1 (yay, Shawn!)
Undergrad RAs not heard from in months: 1

Grant applications rejected: 3
Grant applications still pending: 3

New papers mostly finished at beginning of semester: 3
Papers mostly finished at end of semester: 5
Papers under review or revision at beginning of semester: 5
Papers under review or revision at end of semester: 5
New papers half finished at beginning of semester: 4
Papers half finished at beginning of semester: 5
"We should really do a paper together on X" ideas at start of semester: 5
"We should really do a paper together on X" ideas at end of semester: 8

Blog posts written: 31
Unfinished posts in my drafts folder: 50

Papers submitted: 0
Papers accepted: 0
Papers published: 0

Self-Centered

Memos to Self, re: Pedagogy

Attention conservation notice: An exercise in public self-embarrassment as an aid to behavior modification.

1. [REDACTED]
2. The next time a research student gets you a book from your wish-list as a gift, do not let the first words out of your mouth be "Wow, I just bought that the other week!".
3. Keep the wish-list up to date.

Self-Centered; Corrupting the Young

Today in Et in Arcadia Ego Blogging

Both brought to you via Bill Tozier.

First, from LOL Manuscripts:

The post is worth at least a sardonic glance.

Second, a discussion of the whole Rennes-le-Château/Priory of Sion mythology as an alternate reality game devised by Pierre Plantard. In other words, Holy Blood, Holy Grail and The Da Vinci Code are the "Tlön, Uqbar, Orbis Tertius" we deserve.

— The fact that death and delusion are on my mind has nothing whatsoever to do with the fact that my students in 462 are turning in their final papers today.

Psychoceramica; Writing for Antiquity; Linkage

Assorted Link Roundup, May 2008

Without style or grace.

Wolfgang Beirl explains why financial engineers (like the ones I've been teaching this semester) are also known as "rocket scientists". There are connections here to Wolfgang's thoughts on telephones and the foundations of statistics.

Man's role in changing the face of the Earth dep't.: Ben Fry's map of the 48 contiguous states, showing only streets and roads. Everything else, astonishingly, emerges from that. (Via Unfogged.)

The radical right revives the theory of magical kingship propounded by Sir J. G. Frazer in The Golden Bough, in which the health of the land is sympathetically tied to the character of the ruler, as an account of the American presidency, and correspondingly prophecies doom, doom, DOOM! should Hillary be elected. Illustrated with kittens. Note: WorldNetDaily, unlike the Landover Baptist Church, is not a parody. (I've mentioned them before.)

Speaking of signs of the apocalypse, Thomas "The Baffler" Frank is now writing op-eds for the Wall Street Journal. (Via Aaron Swartz.)

Mind Hacks offers two neurologically-themed tattoos, observing of the second that it produces "a markedly different effect, despite the fact it resides in the same location".

Steve Laniel and Tom Slee review Clay Shirky's Here Comes Everybody. You have probably already seen or read Shirky's talk "Gin, Television and Social Surplus". His social history is over-simplified, and I get a bit leery of my own response to things which push my buttons so thoroughly, but nonetheless — preach it, brother Clay, preach it!

Brooks Simpson, in an interview with the Southern Poverty Law Center, briskly shreds various lies about the US Civil War and the Confederacy propagated by modern apologists for "treason in defense of slavery". Via Abiola Lapite, who has a good post on the genetics of height.

Sierpinski cookies (via Dave Feldman).

Because I am a mean and vicious person, I take great pleasure at reading Kathy G. toy with someone who pretends to know something about economics (1, 2, 3, 4). G. is a public-spirited person, so when she says "I write about economic theory because I believe it is Really. Fucking. Important. Bad economic models make for bad economic policies.", I believe her. But I enjoy reading her for the sheer pleasure in the evisceration. Similarly, I think that in a juster world, Camille Paglia would now be remembered only as the occasion for this 1991 Molly Ivins essay.

Kit Whitfield explains the concept of a "Macho Sue":

A disagreeable variant of Mary Sue, often found in action films, cop shows and the more battly kind of science fiction. While Mary Sue is a fictional character who bends the universe around herself with her amazing specialness, Macho Sue bends the universe around his manhood. He has a particular ability to get away with behaviour that would be considered bad in a woman — to the point of behaviour that would be considered typically female by a misogynist if displayed by a woman.

These traits usually involve poor self-control, such as outbursts, tantrums, sulks, and a refusal to take responsibility for his own behaviour towards others when he's upset. It's not uncommon for Macho Sue to be prejudiced, or at least suspicious of the unfamiliar, and he's almost always unusually disrespectful to others; he has a particular propensity for taking an unreasonable dislike to somebody on sight (only to have it validated later). When thwarted, he tends to be affronted as well as frustrated, in a way that suggests neither he nor the narrative think it right that anyone but him should ever get their way. The story tends to throw straw men at him by way of obstacles, but they're never shown as equally masculine, and thus are without any heroism of their own. Macho Sue is emotional, but with such an assumption of gendered authority that nobody questions the manliness — in the rightful sense of 'adulthood' — of his behaviour.

She instances (the characters played by) John Wayne, but, oddly enough, neglects to mention Achilles.

Further on the literary-critical vein, a remarkably funny, yet thoroughly horrifying, review of a set of novels I will not be reading. It ends thus

The PALADIN OF SHADOWS series is arguably the most horrifying series of books I have ever read. It has a hero I can't stand, politics so strong they're comical, and sex scenes that are downright horrifying. And I cannot stop reading it. I am going to buy every single one, and if Ringo ever comes out with a spin-off featuring Katya as Cottontail the Bionic Whore, I will buy that too. Because dammit, there's bad, and then there's so bad you have to memorialize it for future generations.

but you really need to read what comes before it to get the full effect. The reaction by the author of the books in question is — startling. (Via Kate Nepveu.)

Thematically not-unrelated, an experiment with a famous comic book author. (For the record, I liked Ronin well enough when I read it as a teenager, but generally haven't seen what there was to get excited about in Miller's work; at most a "lower and distorted form" of a general theme.)

Second in our series of great moments in Afghan Buddhism: the earliest known oil paintings may be from Bamiyan (via Matthew Berryman).

You should read Existence Is Wonderful. She changes my mind about things.

I become more and more convinced that one of the keys to understanding our intellectual life is the Skolnick Effect. It is hard to understand the success of neuromarketing otherwise, for example. It's not that functional brain imaging can't be scientifically useful (I'm involved in some projects myself), but the level of the usual study which gets popular attention is to tell us, on the basis of tiny samples, that some part of the brain is differentially activated by thoughts of attaining money, chocolate, justice and sex and/or dirty pictures. (That last link in particular offers a glimpse into a remarkable clusterfuck of bad science journalism amplifying sloppy thinking.) At this point what you are really learning is that there isn't a straightforward mapping from our psychological concepts to paticular brain regions, which is something the neuropsychologists have been trying to tell you for quite a while now. You can even say it with math, but that doesn't seem to make people any more inclined to listen.

Worse, the journalists — and even many of the scientists — seem incapable of separating "implemented in the brain" from "innate". (A recent offender, via Abiola. [It wouldn't surprise me in the least if some sense of social hierarchy is innate in human beings — with all the disclaimers about what such statements mean hereby incorporated by reference — but the point is that the results reported are completely irrelevant to the question of innateness.]) I realize we have thousands of years of ingrained ideas about mind-body dualism and human nature to work through here, but honestly, people, could we at least get into the eighteenth century? All our thoughts and actions involve our brains somehow; detecting them in the brain with current technology says nothing about their being innate, unless you want to seriously say that the rules of chess are hard-wired into our genomes. But if I pursue this further I will get into the bog of free will, and the idiotic conclusions about it people draw from weird experiments...

(Meanwhile, the fact that people can get papers in Science out of the astonishing prediction that territorial ethnic conflict requires the geographic proximity of (self-perceived) ethnic groups, and is rare in locales where one group is an overwhelming majority, suggests that there is a version of the Skolnick Effect involving toy-model simulations.)

The newly-risen Fafblog shows that prophetic parody is the only way to keep up with the real news.

Carlos Yu has, sadly, stopped blogging. I feel a bit bad because I always liked his stuff and rarely told him so. I will miss very much the only blogger capable of writing about ancient Sanskrit plays featuring "creepy horny drunk carnivorous beggars covered in human ash, accompanied by hott chick acolytes, carrying around someone's skull, asking you for money" (parenthetically adding "I think they used to squat in Tompkins Square Park"); the grand unified theory of wingnuts; Garry Wills; paleobiochemistry; football and other pure products of America; and God knows what else. I hope that, like Fafblog, he will one day return.

A while back, Brad DeLong linked to a parody of Thomas Aquinas's proofs of the existence of God, Five Ways of Proving the Existence of Santa Claus. This is ridiculous, of course, but really not much more so than such serious topics treated by the Angelic Doctor as the relation of the saints to the damned (the saints in Heaven will see the suffering of the damned perfectly; will have no pity towards them; and will in fact rejoice in their suffering); whether the weeping of the damned will be corporeal (yes, but there will be no tears); and whether the damned will be in material darkness:

The disposition of hell will be such as to be adapted to the utmost unhappiness of the damned. Wherefore accordingly both light and darkness are there, in so far as they are most conducive to the unhappiness of the damned. Now seeing is in itself pleasant for, as stated in Metaph. i, "the sense of sight is most esteemed, because thereby many things are known."

Yet it happens accidentally that seeing is painful, when we see things that are hurtful to us, or displeasing to our will. Consequently in hell the place must be so disposed for seeing as regards light and darkness, that nothing be seen clearly, and that only such things be dimly seen as are able to bring anguish to the heart. Wherefore, simply speaking, the place is dark. Yet by Divine disposition, there is a certain amount of light, as much as suffices for seeing those things which are capable of tormenting the soul. The natural situation of the place is enough for this, since in the centre of the earth, where hell is said to be, fire cannot be otherwise than thick and cloudy, and reeky as it were.

Some hold that this darkness is caused by the massing together of the bodies of the damned, which will so fill the place of hell with their numbers, that no air will remain, so that there will be no translucid body that can be the subject of light and darkness, except the eyes of the damned, which will be darkened utterly.

On which note, I have a final exam to give.

Linkage; The Commonwealth of Letters; Minds, Brains, and Neurons; The Natural Science of the Human Species; The Beloved Republic; Afghanistan and Central Asia; The Dismal Science; Math; Learned Folly; The Running-Dogs of Reaction; The Continuing Crises; Philosophy; Psychoceramics

Books to Read While the Algae Grow in Your Fur, April 2008

Richard Bookstaber, A Demon of Our Own Design: Markets, Hedge Funds, and the Perils of Financial Innovation

One part "financial crises I have known" to one part general thoughts about market dynamics, and in particular the difficulties that arise due to complexity, "tight coupling" of markets, and leverage. The stories are going to be familiar to most people interested in the subject. The latter are interesting but under-argued. This is true even when I agree with him, about, e.g., the limitations of statistical modeling in financial markets. (The pages on Gödel's Theorem, the Heisenberg Uncertainty Principle and chaos were painful, but also completely logically independent of the stuff about finance.)

This may get a full review later. For now I'd just say that his main recommendations — avoid complex and novel financial instruments, avoid leverage, and avoid trying to optimize to current conditions, in favor of responding adequately to a wide range of situations, including ones you can't currently anticipate — are not bad as words of wisdom, but he has no hint as to how they could be implemented under current conditions, i.e., in the actually-existing capitalist financial system he describes.

This interview with Andrew Leonard in Salon serves as a decent summary.

John McGowan, American Liberalism

Unapologetic advocacy of modern liberalism as an attempt to provide equal and, crucially, effective freedom to all. Liberalism tries to achieve this by creating institutions which make arbitrary, unaccountable, unchecked power ineffective, because powers are checked and balanced by other sources of power and made to answer for theirs actions to those over whom power is exercised. (This distinguishes it from anarchism, whose ideal is simply to eliminate power.) The means by which these things are achieved are secondary, and evaluated pragmatically, by their effectiveness and side-effects in given conditions as compared to available alternatives. (Liberalism, though he doesn't put it this way, becomes in his hands a general ideology of the second best.) Seen thus, there is a clear line of descent between the 18th century liberalism of (most of) the American founders and the modern ideology, with the main development being taking seriously the bit about all men being created equal.

McGowan tries very hard here to reach the general educated public, rather than fellow academics, and almost succeeds. (There are turns of phrase which make it obvious that he's read his post-structuralists, but they're not unreadable ones.) The ideal book along these lines would be something at the level of, say, Milton Friedman's Free to Choose, and McGowan isn't there, is still a little too committed to academic forms, but this is clearly a labor of love, and I hope it will succeed in being influential.

(I confess, though, that I don't get why he thinks cell phones are worse for involvement in the public sphere than land-lines. The reverse, if anything.)

John McCleary, A First Course in Topology: Continuity and Dimension

Well-written textbook of topology, with a historical flavor (but modern methods), and an emphasis on (as the subtitle suggests) the problem of showing that dimension is invariant under continuous and invertible mappings (homeomorphisms). The reader needs a solid grasp of basic real analysis, linear algebra and abstract algebra.

William R. Cross, The Burned-Over District: The Social and Intellectual History of Enthusiastic Religion in Western New York, 1800--1850

A very solid historical work, though it presumes a fair degree of familiarity with the Protestant sects of early 19th century America, and even with the political history of New York. (I lose any right to review this by the fact that I had to look up the Holland Company, and was boggld by what I found.) Though he does not put it this way, a big part of his thesis as to why much but not all of western New York was so susceptible to religious and semi-religious fads then was that the pure products of Yankeedom go crazy. He makes this very plausible, in a way which nonetheless manages to be sympathetic to the enthusiasts.

Some remarks about feminine weaknesses, and the places where he seems to blame the Civil war on, of all people, the Abolitionists, are distasteful, but also a sign of the moral progress separating us from 1950...

David Ruelle, The Mathematician's Brain: A Personal Tour Through the Essentials of Mathematics and Some of the Great Minds Behind Them

An eminent mathematical physicist's take on mathematics and mathematicians. It manages to be sane, pragmatic, thoroughly unromantic, and yet highly enthusiastic for the subject. I actually think anyone who remembers high school math could follow everything; his trick, here, is to start with that sort of stuff and explain how mathematicians generalize it, why they generalize it, and especially why they generalize it in certain ways and not others. — Despite the title, this is strictly psychological, with negligible neuroscience. Given the utter lack of useful neuroscientific data about mathematical thinking, this is sound.

Draws on his "Conversations on mathematics with a visitor from outer space" (PDF), but with all traces of Gallic whimsy removed. (They would probably have become unbearable at book length.)

Matthew Yglesias, Heads in the Sand: How the Republicans Screw Up Foreign Policy and Foreign Policy Screws Up the Democrats

Young master Yglesias finally delivers on that early promise with a book, which, mercifully, is not about blogging and not just a collection of his blogging. Rather it is a sustained, sober, well-written argument in favor of robustly and forthrightly re-embracing the tradition of liberal internationalism, which tries to create institutions that will channel international affairs in peaceful directions and restrain raw power, in order to create a better world for all, including the powerful. As against this we have various strains of nationalist and/or imperialist viciousness and idiocy. Yglesias argues for liberal internationalism and against other ideologies on grounds of morals, practical benefits (the life of a hegemonic power being nasty, brutish and short), and sheer political expediency for the Democratic party, since the alternative hasn't been working out all that well. (He also offers up some brisk but sincere mea culpas.) I would have preferred more argument about morals, e.g. reminding people that the point of our country is not supposed to be a thousand years of crushing global military dominance, but suspect my own impulses in that direction.

Can be read in a day, if you're stuck on planes. Highly recommended if you're in to this sort of thing.

Warren Ellis and Salvador Larroca, Newuniversal: Everything Went White

Comic-book candy. — OK, it deserves a little more than that. From time to time Timothy Burke complains about how astonishing things happen in comic books, which ought to transform the world, but somehow life goes on exactly as before. This series starts from a world slightly askew from our own, where the appearance of superhumans does, in fact, change things.

Charles Stross, The Jennifer Morgue

Sequel to The Atrocity Archive. More lightheartedly chilling Lovecraftian spy fiction, from the perspective of the geeks in IT. Only, this time, haunted by the ghost of James Bond.

John Dewey, Liberalism and Social Action

"And now abideth liberty, individuality, and the critical use of intelligence, these three; but the greatest of these is intelligence." (Not an actual quotation.)

Brian K. Vaughan et al., Ex Machina: Tag; Fact vs. Fiction; March to War; Smoke Smoke; Power Down

Comic books. Actually, I read these back in February, not too long after the first in the series, but forgot to mention them here. I suspect I can guess where this is going, but even if I'm right I want to see how they get there.

Margaret Maron, Up Jumps the Devil; Killer Market; and Home Fires

More unreasonably charming mystery novels about murder in increasingly-exurban North Carolina. Series fatigue will doubtless set in eventually.

Books to Read While the Algae Grow in Your Fur; The Progressive Forces; Scientifiction and Fantastica; Cthulhiana; The Continuing Crises; Mathematics; Psychoceramica; The Beloved Republic; Minds, Brains, and Neurons; The Dismal Science

May Day 2008: Strike Against the War

What follows is a letter my friend John Burke has been circulating to friends. John used to blog as "reprieved" a.k.a. "rootlesscosmo", but gave that up. I wish he'd start again; but in the meanwhile I have his permission to reprint this.

I well remember how indignant a lot of antiwar people were at US organized labor's late, feeble, and sometimes dead wrong positions during the Vietnam War. Much of the then AFL-CIO leadership supported the war (though this support grew less vocal as the war dragged on under a Republican administration); so did a lot of union members, notably the building trades "hard hats" who waded into an antiwar rally in Manhattan in 1969. There were exceptions, including the International Longshore and Warehouse Union (ILWU) on the West Coast and, eventually, the United Auto Workers and a number of public employee unions; there was a labor coalition against the war, which formed a contingent at rallies, bought ads in the print media, and lent support to antiwar candidates.

What there wasn't, though, was any use of labor's economic strength--the strike weapon--to express opposition to the war, and that baffled and irritated some antiwar activists, especially those who didn't know much about labor law or labor history. (I know this doesn't apply to a lot of the recipients of this message; feel free to skip ahead if this is familiar material.) In particular, students from middle-class families weren't aware that under the Taft-Hartley amendments to the National Labor Relations Act, the use of the strike weapon for any purpose except in disputes about collective bargaining agreements is explicitly prohibited. They also may not have grasped the context of Taft-Hartley, which--though labor opposed it and Truman vetoed it, only to be overridden by a Republican-majority Congress--set in stone the main outlines of the postwar, Cold War-era "social compact:" labor would save job action for "pork chop" issues, confine its political action to endorsing candidates, impose a "loyalty" test on union leaders (which led to the expulsion of the Left-led unions from the CIO in 1949) and become a partner in the worldwide struggle against Communism. In return, major corporate employers would recognize unions and accept contracts that included regular productivity and cost-of-living increases; there were occasional disruptions in this cozy arrangement, but strike activity fell sharply from the big upsurge in 1946-47 and stayed low until the "stagflation" and mass layoffs that began in the mid-70's.

So job action against the Vietnam War would have been not only a challenge to the law but a sharp break with the postwar social compact, at a time when that compact's real meaning was thrown into sharp focus: labor was called on to support a Third World military intervention against a Communist-led liberation movement, at a moment when that intervention was producing a flush of prosperity and job growth. (Harry Bridges of the ILWU, when he launched a campaign to recruit new members from high-unemployment communities in response to the growth of war-related Pacific shipping, admitted ruefully that it was blood money.)

But the social compact started falling apart in the 1970's--the war turned out to be a large part of the reason, though I've promised myself not to use the word "dialectical" in this brief survey--and Reagan shredded it after 1980. The Cold War is over, the steady-growth postwar economy is over, union density as a percentage of the workforce is down from 35% to 13% (and less in the once-powerful industrial sector), anti-labor policies have been entrenched at the NLRB for many years, and neither the Carter nor Clinton administrations achieved labor's goal of legislative reform. (How hard did they try? Good question.)

In short, the deal that undergirded labor's qualified support for the Vietnam War has fallen apart.

The postwar social compact was a tradeoff; the other side went back on the bargain. It's time for labor to begin reclaiming its full range of tactical options in support of a robust participation in political life, on an agenda of labor's choosing without the artificial constraints imposed by Taft-Hartley. This will be, inevitably, a gradual process, and it may get ugly; I don't think there are any US Attorneys dumb enough to try to indict the ILWU leadership, but I may be being too generous. (It's a grave failing of mine.)

In any case, the first big crack in the ice is the ILWU's planned coastwide work stoppage tomorrow,

http://maydayilwu.googlepages.com/ which will also coincide with and support an immigrants' rights rally (and it certainly is refreshing that the immigrants' movement has reclaimed May Day as a day of workers' action; sure, the sectarian Lefties will try to hop aboard the bandwagon, but who cares?) I'll be marching tomorrow, with my United Transportation Union button on, prouder of the labor movement, my movement, than I've ever had a chance to feel in my life. Hope to see you there.

The only thing I have to add is that when John says "my United Transportation Union button", he means "the button of the union I belonged to during the more than a quarter century I worked on the railroads".

Manual trackback: Chaotic Soliloquy; Stripes with Plaid

The Continuing Crisis; The Progressive Forces

Putting the CART before the Horse-Race

Finally, from Amanda Cox at the Times, a decision tree students can believe in (click for full size):

Now, if I wanted to be a hobby-horse-riding pedant, I would compare this to a regression of vote-share on these covariates, and ask you rhetorically which one was easier to understand, and which gave more of a misleading impression of being more than a summary description; but I'll save that for the poor souls who take data mining in the fall. (This is from the 16th; it would be interesting to see how it changed after last Tuesday. Not that I'm bitter.)

Local interest note: Ms. Cox will be judging the final-project posters produced by the students in Prof. Nugent's graphics and visualization class (36-315), next Friday, 2 May, 12:30 to 1:20 pm in Porter Hall 125C. Friendly, non-psychotic visitors are welcome.

Via Flowing Data, via K. (Owing to the silly limits of the Times archives, I can't find the direct link to the story!)

Manual trackback: A Well, With Two Buckets

The Beloved Republic; Enigmas of Chance

From Your Lips...

I swear I had nothing to do with this; blame them.

Via Bill Tozier.

Complexity; Learned Folly

Voting will continue until morale improves (Public Service Announcement)

If you (1) live in Pennsylviana, (2) are registered to vote and (3) are not sure about where to vote, what to bring, etc., votesPA.com has the information you are looking for.

Manual trackback: Cranial Darwinism.

The Beloved Republic; Heard About Pittsburgh, PA

Chaos, Complexity, and Inference (36-462): Lecture Notes

This page will be updated as the semester goes on, if you want to use this RSS feed to track them. Alternately, lecture notes will be linked to on the course syllabus, which includes the readings.

Lecture 25 (April 22): Inference on networks

slides

Lecture 24 (April 15): Contagion on networks

slides

Lecture 23 (April 10): Collective phenomena and self-organization in agent-based models

slides

Lecture 22 (April 8): Agents and Agent-Based Models

slides

Lecture 21 (April 3): Complex networks 2

slides

Lecture 20 (April 1): Complex networks 1

slides

Lecture 19 (March 25): Inference from simulations

slides; R

Lecture 18 (March 20): Special problem-solving session for homework 2

partial solutions; R

Lecture 17 (March 18): Error Statistics and Severe Testing

slides

Lecture 16 (March 6): Heavy tails 4, testing and evaluation

slides and R

Lecture 15 (March 4): Heavy tails 3, estimation

slides

Lecture 14 (February 28): Heavy tails 2, origins

slides

Lecture 13 (February 26): Heavy tails 1, basics

General R files for the next several lectures

slides; R

Lecture 12 (February 21): Self-organization 2

slides

Lecture 11 (February 19): Cellular automata 2/Excitable media

slides

Lecture 10 (February 14): Cellular automata 1

slides

Lecture 9 (February 12): Self-organization 1

Philip and Phylis Morrison and the Office of Charles and Ray Eames, Powers of Ten

slides

see also: Pattern Formation in Cocktails

Lecture 8 (February 7): Determinism and randomness

Henri Poincaré, "Chance" (from Science and Method, 1908)

slides

Lecture 7 (February 5): Information theory

slides

M.C. Hawking, "Entropy", from Fear of a Black Hole [lyrics; mp3 (radio-safe Brief History of Rhyme version)]

Ray and Charles Eames, A Communications Primer

Lecture 6 (January 31): Inference for Markov chains and related processes

slides

Note: Maximum Likelihood Estimation for Markov Chains

Lecture 5 (January 29): Symbolic dynamics; stochastics from dynamics

slides

Note: More on the Topological Entropy Rate

Lecture 4 (January 24): Attractor reconstruction and nonlinear prediction

slides (see slides for R examples).

Note: Nonlinear prediction, nearest-neighbors, kernel methods

Lorenz time-series generator, written in Perl.

the Lorenz time series used in the lecture

Lecture 3 (January 22): Attractors

slides; R

Lecture 2 (January 17): More chaos

slides; R

The Arnold Cat Map Movie (starring Marlowe the Cat, directed by Evelyn Sander)

Lecture 1 (January 15): What is a dynamical system? What is chaos? What is a simulation?

slides; R

Corrupting the Young; Complexity; Enigmas of Chance; Networks

"Thou Shalt Not Follow a Multitude to Do Evil"

From William R. Cross, The Burned-Over District: The Social and Intellectual History of Enthusiastic Religion in Western New York, 1800--1850 (reprint; New York: Harper, 1965, pp. 81--82):

The whole tribe of Yorkers exhibited a trait which bears on the nature of Burned-Over District credulity. It ranks in importance with the canniness and moral intensity customarily attributed to Yankees and relates to both, but has been less noticed because it is difficult to define and isolate. Against the "holy enterprise of minding other people's business," which produced a marked community-mindedness, these folk balanced a stubborn intrspection in the fashioning of personal beliefs, which recognized no authority this side of Heaven. Frank curiosity, pride in independent thinking, a feeling that action should be motivated by sound logic and never by whimsy, a profound skepitcism of any rationalization looking to less than the supposed ultimate good of society, and, once arrived at, an overweening confidence in one's own judgment — all these attitudes differently demonstrate the same trait. The mores of the community must definitely be observed when established and agreed upon, but in practice they remained forever open to challenge and subject to revision. No apology was required for unorthodoxy dictated by conscience in conference with Scripture; rather, any difference from custom created a compelling obligation for the individual to press toward conformity with his own new light.

Cross goes on in a footnote to add that "Certain angles of [this trait] survive the generations of Yankee descendants, and my discussion of it is based in part upon observation of acquaintances, my family, and myself".

The Beloved Republic; Psychoceramica

The End of the Age

Vanquished, the hero sails into the west, but legend says that he will return when his people's peril is most dire. (The legend does not say whether that was a promise or a threat.)

(Photo via Warren Ellis, who got it from English Russia)

Linkage

Solvitur ambulando

A: Hey, you over there, the one walking! You're doing it wrong.
B: Excuse me?
A: You're only using two feet! You should keep at least three of your six in contact with the ground at all times.
B: ...
A: Look, it's easily proved that's the optimal way to walk. Otherwise you'd be unstable, and if you were walking past a Dutchman he could kick one of your legs with his clogs and knock you over and then lecture you on how to make pancakes.
B: What? Why a Dutchman?
A: You can't trust the Dutch, they're everywhere! Besides, every time you walk it's really just like running the gauntlet at Schiphol.
B: It is?
A: Don't change the subject! Walking like that you're actually sessile!
B: I don't seem to be rooted in place...
A: It's a technical term. Look, it's very simple, these are all implications of the axioms of the theory of optimal walking and you're breaking them all. I can't get over how immobile you are, walking like that.
B: "Immobile"?
A: Well, you're not walking properly, are you?
B: Your theory seems to assume I have six legs.
A: Yes, exactly!
B: I only have two legs. It doesn't describe what I do at all.
A: It's a normative theory.
B: For something with six legs.
A: Yes.
B: I have two legs. Does your theory have any advice about how to walk on two legs?
A: Could you try crawling on your hands and knees?

Disclaimer: A is the one with the weird Batavophobia, not me.

Manual trackback: Vukutu

Learned Folly; Enigmas of Chance

Behold the Masses (Next Week at the CMU Statistics Seminar)

Attention conservation notice: Publicity for a talk at CMU next week. Of limited interest if you're not free and in Pittsburgh at 4 pm on Monday the 14th.

We are very happy to have Nathan Eagle, of the Media Lab and SFI, as our seminar speaker next week, talking about the extremely cool work he's been doing on some extremely large social networks.

"Inference in Complex Social Systems: Insights and Applications from the Behavior of the Aggregate"

Monday, 14 April 2008, 4 pm, Porter Hall 125C at Carnegie Mellon

Abstract: I have used mobile phones to continuously gather information including proximity, location, and communication from 100 human subjects at MIT. Systematic measurements from these people over the course of nine months has generated one of the largest dataset of continuous human behavior ever collected, representing over 300,000 hours of daily activity. Additionally, in collaboration with several European and African telecommunication companies, I am currently analyzing the call logs of entire countries - dynamic social networks consisting of up to 250 million nodes and 12 billion temporal edges.

In this talk I describe how this type of data can be used to uncover the structure in behavior of both individuals and organizations, infer relationships, and study social network dynamics. By combining theoretical models with rich and systematic measurements, we show it is possible to gain insight into the underlying behavior of complex social systems.

While results such as uncovering scaling laws from the communication patterns of hundreds of millions of people will certainly be one emphasis in this talk, of equal importance is how this data can enable applications that improve our society. I will demonstrate a variety ways these insights into our own behaviors can be used to develop applications that better support both the individual, organization and society.

The talk is of course free and open to the public; come if it sounds interesting (unless you're like some people who attend talks in Santa Fe [not that Nathan's work is remotely comparable to Sheldrake's]).

Networks; Enigmas of Chance; Complexity

He Is Risen

Fafblog has returned to us. Long live the new era of Fafno-Gibletsian rule over the cosmos!

Linkage

Books to Read While the Algae Grow in Your Fur, March 2008

I slacked off on posting this until mid-April, if anyone cares about why it's out of sequence.

Joss Whedon, Fray

Comic book. Mmm, candy.

Simon Oliver & c., Exterminators, vol. 1: Bug Brothers

Comic book. LA low-lives hold the line against the forces of chaos, embodied by vermin and large corporations. A bit of a come-down for the combat myth from Python, perhaps, but I'll be reading the others.

Sandra McDonald, The Outback Stars

Aussies in space, after a never-quite-spelled-out ecological collapse on Earth, with mysterious ancient alien artifacts and naval-procedural elements. Mind-candy.

Warren Ellis et al., Apparat: The Singles Collection

Four "issues" from four different otherwise-nonexistent comic books, each imagining a different line of descent from early twentieth-century pulp fiction to sequential graphic story-telling, without the invention of the superhero genre. Angel Stomp Future revisits the future-shock and technology-driven social liquification of Ellis's glorious Transmetropolitan; it's more deliberately shocking that Transmet (which is saying something), but not as good (which says little). (That sentence also applies, mutatis mutandis, to Ellis's stand-alone City of Silence.) Frank Ironwine is a detective/cop story, almost a dry run for the (excellent and continuing) Fell. Quit City descends from aviator stories (by way of confrontation with personal, and perhaps I should add metaphorical, demons); Simon Spector from old detective serials like The Shadow. They're all not just clever exercises in genre bending and para-literary archaeology, but also well-told and well-drawn tales.

Update: see also Aetheric Mechanics.

Margaret Maron, Southern Discomfort and Shooting at Loons

Sequels to Bootlegger's Daughter (discussed here). Continues in a light-hearted, quirky-semi-rural-stories vein, which ought to clash with the fact that they're really stories about poverty, attempted rape and multiple homicide (Discomfort), and the collapse of traditional livelihoods and the values they supported (Loons), but, somehow, doesn't.

Phil Rickman, The Fabric of Sin

Latest in his Merrily Watkins series of "procedural ghost story" mysteries. A haunted house story, involving family feuds, the fiction of M. R. James, and people with obsessions about the Templars. Less of a supernatural element here than usual, everything is satisfactorily explicable as people being either creepy or creeped out. (Previous installments: here, here, and here.)

Jorge Cham, Piled Higher and Deeper, Chapter 3: Scooped

If the idea of a comic strip about the travails of geeks in graduate school appeals to you, then you are probably already reading Ph.D. Comics, but should buy this anyway, as a contribution to the fund for the support of cartooning roboticists. If, on the other hand, that sounds dreadful, reading this would probably only confirm your darkest suspicions about the lumpentechnocracy.

A. E. Stallings, Hapax: Poems

Highly formal (sonnets!) but also very good poems, many with classical themes, ranging in tone from the funny ("XII Klassikal Lymnaeryx") and the drily amusing ("Dead Language Lesson") to the darker "old standards" of transience and loss (e.g. "Arrowhead Hunting") — or amusement and sentiment, as in "Last Will". The "Antiblurb" on the back cover may give some idea of the contents:

This is not necessary. This is neither
Crucial nor salvation. It is no hymn
To harmonize the choirs of seraphim,
Nor any generation's bold bellwether
Leading the flock, no iridescent feather
Dropped from the Muse's wing. It does not limn,
Or speak in tongues, or voice the mute, or dim
Outmoded theories with its fireworks. Rather

This is flawed and mortal, and its stains
Bear the evidence of taking pains.
It did not have to happen, won't illumine
The smirch of history, the future's omen.
Necessity is merely what sustains —
It's what we do not need that makes us human.

Some other poems are available online here and here, and there's a version of "Recitative" (not in this collection) charmingly illustrated by R. Kikuo Johnson.

Books to Read While the Algae Grow in Your Fur; Scientifiction and Fantastica; Learned Folly; The Commonwealth of Letters

Career Advising Day

Attention Conservation Notice: Another thousand-odd-word rant about reactionary idiots pretending to be scientists.

Contemplating the writings of the now-deservedly-forgotten Pierre Teilhard de Chardin, the late, great Peter Medawar was driven to observe that "Just as compulsory primary education created a market catered for by cheap dailies and weeklies, so the spread of secondary and latterly tertiary education has created a large population of people, often with well-developed literary and scholarly tastes, who have been educated far beyond their capacity to undertake analytical thought." (Whether that is a trained incapacity is itself a nice question.) Some of those people, owing to those tastes, pursue careers in academic research; the problem for them is that they are not actually very good at what they are supposed to do, which is come up with novel, insightful, important, precise, and accurate findings. Suppose that you are such a person, and that you do not want to switch to some other line of work to which you might be better suited. What to do?

Perhaps the best thing which could happen to you would be to run across a new and controversial theory which speaks to you at a deep level, both intellectually and temperamentally. If you are what William James called "tender-minded", like Teilhard de Chardin, then Medawar has already mapped out your trajectory, though nowadays the Templeton Foundation would likely be involved. If instead you are what James called "tough-minded" — "materialistic, pessimistic, irreligious, fatalistic, sceptical" — then edification-through-obfuscation is not an option, but it wouldn't even occur to you. Instead, you take your theory and you write papers about it, where you make claims about lots of hot-button topics, especially sex and current political controversies. The papers seem to carry the signs of rigor, but are actually deeply fallacious — maybe you see this, but are so convinced the conclusions are right you don't care, or maybe you're so convinced of the conclusions you can't see the errors. (There is some peer-reviewed venue where you can publish almost arbitrarily sloppy papers, so getting into print won't be a problem.) Then — and this is the key — you start promoting your papers, and find that more salacious and provocative your spin on them, the bigger the response. Your possibly-unconscious shamelessness about publishing rubbish will not only give you an advantage in sheer publications over other mediocre scholars who happen to have an intellectual conscience, but will also get you media attention. The reason it will get you media attention, and credibility with the media, is that they will see your institutional affiliation and your peer-reviewed papers, and so you become not just another crank but a Serious Scholar Contributing to the Debate. The whole package — carelessness, provocation and publicity — is wonderfully self-reinforcing, so you write even more careless papers, with yet more provocative conclusions, which you push even harder. (As a wise woman once said, "No one ever forgets how to do something that's worked for them in the past.") With a bit of luck, book contracts, magazine columns, etc., will follow in their train. Your career becomes like two drunks supporting each other as they stagger down the street: neither crappy academic research nor media presence could stand up on their own, but together they can lurch and shamble in glorious, glorious inebriated freedom, bellowing about the fierce joys to be found in facing what's revealed by the harsh light of your pseudo-scientific prejudices.

Ladies, gentlemen, and distinguished others, I give you Dr. Satoshi Kanazawa of the London School of Economics, the Fenimore Cooper of sociobiology, a man who has leveraged an inability to do data analysis or understand psychometrics into an official blog at Psychology Today, where he gets to advocate genocidal nuclear war as revenge for 9/11. He seems to mean it, rather than be fukayaming.

His argument — to the extent that it is an argument and not just a wish-fulfillment fantasy — has to do with his earlier attempt to explain "why most suicide bombers are Muslims". Leave to one side whether his attempted explanation is coherent, two things strike one on reading that. The first is his near-total disconnection from the literature on, precisely, the causes and motivations of suicide bombing — no Sageman, no Pape, the only mention of Scott Atran (an actual evolutionary psychologist, and very aware of the problems with the kind of crude ad hominid argument Kanazawa pushes) basically misses Atran's point, etc. The second is that the fact he is trying to explain something which isn't true: the tactic was pioneered [Update: a poor choice of words; see below] is and long has been heavily used by the decidedly non-Muslim Liberation Tigers of Tamil Eelam. This is well-known to anyone even slightly interested in understanding this horrible practice; Kanazawa doesn't even try to explain it away. But facts like these don't matter when the real goal is some combination of, on the one hand, constructing and projecting a fantasy ideology, and, on the other, sheer self-promotion.

I don't know of any systematic data on whether James's distinction between tender-minded and tough-minded thinkers really holds up, but at the level of casual empiricism it's pretty persuasive, and I fall very much on the tough-minded end of the spectrum. I find that sort of position persuasive, but at the same time it takes only a minimal amount of self-knowledge (certainly that's all I've got) to realize that it exposes one to certain characteristic errors or temptations. One of them is self-congratulation at being, precisely, so tough-minded. And one prominent expression of that is a delight in one's superior ability to perceive things as, supposedly, they really are, stripped of sentimental ornament; more than that, a delight in imagining how the tender-minded will be shocked by having to confront these realities. It is especially a delight in reductionism, not as a productive if not inevitable explanatory strategy, but as a series of "nothing-but" claims. This is one of our characteristic forms of wishful thinking, just as much devising imaginary consolations for real sufferings is a characteristic of the tender-minded. With these thoughts in mind, I invite you to read the conclusion of Kanazawa's article on suicide bombers:

Maybe the Muslim suicide bombings are not "terrorist" acts, as the term is usually used. Maybe it has nothing to do with Israel or the American and British troops. Maybe it's all about sex, as everything else in life is. Men do everything they do in order to get laid (Kanazawa, 2003). Maybe young Muslim men are no exceptions.

Satoshi, mon semblable, mon frère: whoever she is, I really hope the sex is worth it.

Obligatory disclaimers:

1. I am agnostic, and uninterested, as to whether Kanazawa, like one of Thorndike's cats, developed these habits by trial, error, and environmental reinforcement, or whether, like one of Köhler's chimpanzees, it came to him in a flash of insight that he could reach the seemingly-unobtainable prize, if only he stacked publicity atop bullshit.
2. Of course, Kanazawa has a perfect right to express whatever opinions and beliefs he might have (within the usual limits of the laws of libel, etc.). Moreover he ought to have the right to pursue whatever lines of inquiry seem to him most promising. Reciprocally, the rest of us have every right to criticize his research as idiotic, ideological rubbish.
3. I actually think there is a lot to be said in favor of evolutionary psychology, but if anything that only makes it more important to stomp on things like this.
4. I cheerfully admit that "I hope we catch the bastards who did this and nuke them till they glow" was one of my first reactions to seeing the pictures from 9/11. (The very first was that I was being shown some kind of disgustingly tasteless viral marketing for a disaster movie.) This sort of reaction is not a sound basis for making policy, and I leave it as an exercise to explain why on evolutionary grounds. I would still like to see us catch the bastards who did this, however.

Credits: Kanazawa's blog found via the appropriately dumbfounded reaction of Michael Meadon. I owe the insights, and much of the phrasing, of my second paragraph to a correspondent who prefers to keep their name out of this.

Manual trackback: Entertaining Research; Ionian Enchantment; Pharyngula; 3 Quarks Daily; Soob; Flagrancy to Reason; O Hermenauta

Learned Folly; The Continuing Crisis; The Natural Science of the Human Species; The Running-Dogs of Reaction

Follow the Oil Money!

Via Skye Bender-deMoll, an old acquaintance from Santa Fe days, a lovely little example of network mapping and the visual display of quantitative information in the service of the public good: Follow the Oil Money. This website lets you track the network of campaign donations from the oil industry, in its various tentacles, to U.S. politicians, with nifty pictures and charts.

As Skye explains, while all this information is a matter of public record, available from the FEC, working with that data is surprisingly hard. (I didn't appreciate just how hard when I wrote this.) Skye and his collaborator Greg Michalec have done a really impressive job of making this accessible. The result is good for hours and hours of entertainment and enlightenment, even if you think that it's all about constitutionally-protected and democratically legitimate freedom of speech, in the form of dollars, on the part of the companies. (Incidentally, if you do think that, please contact me about exciting business opportunities in Lagos.)

It would be fascinating, if perhaps scary, to see a parallel website for military contractors.

Manual trackback: Three Quarks Daily; Media Theory for the 21st Century; The Monkey Cage; Statistical Modeling, Causal Inference, and Social Science

Networks; The Beloved Republic

Profiled

I am this week's profilee at Norman Geras's blog. This is very pleasing to me, as my chance discovery of Norm's Solidarity in the Conversation of Human Kind (followed by his Marx and Human Nature) was an important part of my intellectual development in graduate school. There are some very important areas of politics where I believe that he is very wrong, but he's always sincerely benevolent and worth reading; and sometimes, simply right.

Self-Centered; The Progressive Forces; Linkage

Books to Read While the Algae Grow in Your Fur, February 2008

Norman Geras, Discourses of Extremity: Radical Ethics and Post-Marxist Extravagances

Four essays in two parts. The first half is about the need for Marxists to be explicit about their moral commitments, and a call to work out what constitutes justice in the conduct of a violent revolution, by explicit analogy with just war theory. The second is an extended controversy with Laclau and Mouffe, denouncing them for their manner of leaving Marxism, misrepresentations of the Marxist tradition, idealism, etc., etc. (The analogy of the chain is very good, but not good enough to save historical materialism.)

Jack Campbell, Dauntless, Fearless, Courageous

Mind-candy. Covers unusually horrible and not actually indicative of contents. Fairly grim military science fiction, mixing a take on the Anabasis, the disasters-of-total-war (parts of which seem intended as comments on current events, but not especially heavy-handedly so), and many well-thought-out relativistic battle scenes. Not my usual thing but oddly compelling. Third volume ends with a huge cliff-hanger; more are forthcoming.

Wiktor Stoczkowski, Explaining Human Origins: Myth, Imagination, and Conjecture

If you read the fifth book of Lucretius's De rerum natura, you will find (starting around line 925) an account of the development of the human race from a primitive condition like that of the other animals to the civilization of the last century BC. It sounds startlingly modern --- obviously wrong in some details, but not that different from what one would get from a synoptic over-view of human evolution today. Stoczkowski's thesis is that this is not because Lucretius was very smart and/or very lucky, but that the "hominisation scenarios" one finds in such works of paleoanthropology are really exercises in speculative or conjectural history, part of a continuous tradition which descends from antiquity through the Enlightenment, of which Lucretius was very much a part, and that this tradition has very rarely had all that much contact with archaeological findings or proper scientific procedures more generally. His book is an analysis of the tradition, especially as found in a few dozen prominent scholarly texts, accompanied with arguments that the recurrence of its themes cannot be explained on the grounds that they are empirically well-supported, or even the only conceivable alternatives. They just sound plausible. The argument that cooperative hunting on the savannah requires spoken language would seem to entail that lions can, in fact, speak, only we've failed to understand them; this is absurd, but remarkably popular. (Stoczkowski is especially good on the subject of teeth, but would take too long to summarize.)

As Stoczkowski is at pains to state, none of this means these recurring ideas are wrong, but they are weak, and it's distressing to see them recycled from generation to generation, at most reshuffled and occasionally inverted. (He has obviously been influenced by Levi-Strauss, but no familiarity with structuralism is needed, or even helpful.) He concludes with a plea to abandon conjectural history in favor of seeking truth from facts, accompanied by a reminder that it simply may not be possible to learn the answer to many questions about the evolutionary history of our species.

Books to Read While the Algae Grow in Your Fur; The Progressive Forces; Scientifiction and Fantastica; Writing for Antiquity; The Natural Science of the Human Species; Philosophy

The Caveman's Valentine

Wiktor Stoczkowski, Explaining Human Origins: Myth, Imagination and Conjecture, p. 110:

Since society is [thought of as] the fruit of necessary cooperation, it is believed that this necessity did not exist previously, and that man's ancestors led individual, presocial lives. It is easy to recognise the ancient view of a period of paradisal abundance, when "each one went his own way in search of fruit and herbs", all then being capable of obtaining food without the help of others. Lucretius and Diodorus Siculus were already painting a similar picture of the primordial existence, and in the eighteenth century the idea of the solitary life of the earliest humans became more firmly embedded in popular imagery. In the twentieth century, colorful speculations concerning that grave event, the first encounter between two humans, still persist. Here is how E. Haraucourt imagined it in a "prehistoric novel" which portrays the first tête-à-tête between a male and a female:

A punch on the forehead stunned but did not defeat her and she returned to attack. She buried her teeth in the shoulder of the male who had grabbed her round the waist; it was his turn to scream; picking up a stone, he dealt her such a vicious blow on the top of her head that she collapsed: circles of light were whirling in front of her and she was vaguely aware of a violent mass hurling its weight on her back... When she reopened her eyes, the conqueror was still clasping her but was not devouring her.

This is a good illustration of the firm belief that existence was originally solitary and that the first meeting was not without some difficulties.

The novel, incidentally, was entitled Dâah, le premier homme, and was first published in 1914 and apparently reprinted as recently as 1996. Whether this work is the actual origin of the drag-her-back-to-the-cave meme (and so of its variants), or itself merely another iteration, I couldn't say.

(More on Stoczkowski's book under recommendations for February.)

Writing for Antiquity

First International Causal Prediction Contest: Your Ticket to Perdition

In his History of Western Philosophy, Bertrand Russell claimed that a special section of Hell was reserved for those who claimed to have refuted David Hume on the impossibility of establishing causality. Here (via Isabelle Guyon in e-mail) is your chance to risk damnation in exchange for valuable cash prizes, and a paper in JMLR.

Causality Challenge #1: Causation and Prediction

Deadline April 30, 2008

This challenge bridges the gap between data mining/machine learning and causal discovery. Several datasets drawn from real data, or emulating real data, are provided, with the goal of making predictions under "manipulations".

The setting is similar to a usual predictive modeling setting: We have a training set and a test set; a target variable, whose values are concealed in test data, must be predicted. But, the test data are not distributed like the training data: some variables in test data are "manipulated" by an external agent, i.e. set to given values instead of being drawn from the "natural" distribution. Such problems are encountered in many application domains: In medicine to predict the effect of a new treatment, in economy or ecology to predict the consequences of new issued policies, in marketing to predict customer response to marketing campaigns. We anticipate that the tasks of the challenge should require the knowledge of causal relationships between variables since acting on causes of the target may result in a response change while acting on consequences should not. However, we encourage participants to enter the challenge with any approach to the problem.

Despite being nearly synonymous with causality among machine-learners, the use of graphical models is not required — they're serious about the "any approach" bit. (Parochial boosterism, however, leads me to guess that the winner will use graphical models extensively.) If you're interested, do check out the contest homepage, especially the FAQ.

Mr. Hume and Lord Russell could not be reached for comment.

Enigmas of Chance

Clothes Make Working for the Man Easier

Attention conservation notice: 700 words on a stupid op-ed about how academics dress. Contains ludicrous over-generalizations about the rhetoric of cultural criticism. Don't you have paint to watch dry?

I have just had one Prof. Erik M. Jensen's op-ed "A Call for Professional Attire" referred to me by multiple sources (none especially pointedly, thanks), and I find myself greatly irritated. Jensen says that contemporary American academics generally fail to dress up, in the modes that are supposed to reflect seriousness and status, and spends about 2000 words bemoaning this; longing for a lost "golden age" (his phrase); and trying to ridicule, brow-beat, and shame his audience into complying with his wishes. The closest he comes, in all of this, to present an actual reason for doing so is saying this: "People generally act better when they're dressed right. If a professor is sending a signal of seriousness, of civility, students will pick it up." This is backed up by a casual, second-hand reflection on how "in DiMaggio's day ... [t]he men wore white shirts and ties under coats and hats, the proper attire in public, even at a ball game."

This is a style of cultural commentary which drives me up the wall, so I try to avoid it. It is not that hard to think of an actual rationale for what Jensen wants; it would go something like this. (These are, of course, my words, not his.)

Academics are supposed to impart knowledge and skills to their students, to critique their work, to direct their intellectual and to some extent their moral development; in all these tasks they are supposed to exercise authority over students. They may also be called upon to supervise student or other employees, which is another exercise of authority. They will do so more effectively if they display the recognized external markers of high status and of seriousness, which includes dressing in certain ways and adopting certain demeanors. In fact, if they do this, their authority is more likely to be accepted as legitimate, leading to fewer occasions on which it must be explicitly insisted upon and made into naked acts of domination. Furthermore, academics are often called upon to represent their schools and/or their scholarly communities to the outside world, and this, too, will be done more effectively if they dress in ways which their audiences take to convey seriousness.

This is a reasonable argument for what Jensen says he wants. It refers to consequences, rather than insinuating some mythical intrinsic desirability; it is also an argument with empirical premises, and one susceptible to balancing — how much extra effectiveness is the extra expense, hassle, restriction of personal choice, etc., of this mode of dress worth? Supposing that, at the margin, I would be a slightly more effective teacher if I wore a tie, is that worth enough (to me? to my students? to my university?) to make up for wearing something so utterly ridiculous, an arbitrary self-sustaining convention made silk? One could imagine a reasonable essay which went into these points, backed them up, thought through the trade-offs.

Jensen, on the other hand, just wants to take his internalized norms, however transparently parochial ("faculty members shall dress in a way that would not embarrass my mother"), and pretend that they are the maxims of universal laws, as well as purporting to tell us what various cultural changes mean or signify. This is by far the more common rhetorical mode when people try to criticize manners and customs, and it strikes me as deeply stupid. Or at least deeply stupid to be moved by, since it gives you no reason to believe that acting as the author wants will make things better. However, I must confess that it relies on the strengths of East African Plains Apes (emotionally manipulating conspecifics, devising intentional explanations) and not their weaknesses (establishing quantitative cause-effect relationships, balancing diverse objectives). I have no idea whether this mode of argumentation (if it can be called that) achieves its object, supposing that to be persuasion, and not, e.g., making the like-minded feel better about their shared views.

John Dewey once wrote that, so far from their being no point arguing over tastes, there are actually few things so worth arguing about; but I don't think Jensen's essay was the kind of thing he had in mind.

Update, next day: I'd be remiss if I didn't point out that Jensen could also have tried to persuade us that the way academics dress is just plain ugly, and the world would be at least a bit more beautiful if they adopted his dress code. But I think it's fair to say he doesn't attempt that, either.

Manual trackback: Quantum of Wantum; Matt McIrvin; The Quantum Pontiff; Brad DeLong; Cosmic Variance; 3 Quarks Daily; Need Coffee; The Inverse Square Blog; Physicality of Words; the Monkey Cage; Entertaining Research; Weirleader's Lair

Learned Folly

Pedagogical Value

Would it be wrong to make the topic of my April 1st lecture the implementation of butterfly mode in R?

Modest Proposals Complexity Corrupting the Young

Friday Cat Blogging (Seasonal Affective Disorder from the Souls of Artists and Writers Issues of Non-Science-Geek Edition)

Because it is rainy and gloomy and February. With Futurist art!

States of Mind: Those Who Go

Best of Craigslist (Seattle), To my Ex-GF's Cat:

I don't miss her, but I miss you. You are the only cat I ever liked...and I think you liked me as I'm the only person you let pick up and walk around with. Sure, you were crabby, sounded like a rusty can when you were meowing, would ignore the laser pointer and got pissed at me when I needed to work and not pet you. Oh sure, you'd complain and make me feel bad for feeding you the same thing and at the same time as her other 2 cats, but did you notice I'd always slip you a piece of meat from my dinner plate? I know you were old and stairs were not as easy as they used to be, so I was always secretly glad and flattered to hear your voice by the bedroom door when I'd stay over. I know her kids liked the other animals in the house more then you, and I'm sorry, but I liked you better then her kids anyway. And yes, I know you watched me walk away that last time I left; I knew I wouldn't be coming back so I hope you found that catnip mouse I left in your secret hiding spot...you deserved 1 last rush in your old age.

I'm not sure if you are even still alive as I haven't been by the house since March of 05, but I hope that you are happy, warm and still catching the beam of sunlight in your favorite spot.

Anyway, just wanted you to know that you were the only cool cat I've ever known and that I miss you.

States of Mind: Those Who Stay

Wislawa Szymborska, "A Cat in an Empty Apartment"

Dying--you wouldn't do that to a cat.
For what is a cat to do
in an empty apartment?
Climb up the walls?
Brush up against the furniture?
Nothing here seems changed,
and yet something has changed.
Nothing has been moved,
and yet there's more room.
And in the evenings the lamp is not on.

One hears footsteps on the stairs,
but they're not the same.
Neither is the hand
that puts a fish on the plate.

Something here isn't starting
at its usual time.
Something here isn't happening
as it should.
Somebody has been here and has been,
and then has suddenly disappeared
and now is stubbornly absent.

All the closets have been scanned
and all the shelves run through.
Slipping under the carpet and checking came to nothing.
The rule has even been broken and all the papers scattered.
What else is there to do?
Sleep and wait.

Just let him come back,
let him show up.
Then he'll find out
that you don't do that to a cat.
Going toward him
faking reluctance,
slowly,
on very offended paws.
And no jumping, purring at first.

(Note: My cat is fine, I'm fine, etc.)

Friday Cat Blogging

Books to Read While the Algae Grow in Your Fur, January 2008

An Inconvenient Truth

I may loose some friends by saying this, but having finally watched it I really liked it, and approve. It's true that Gore's occupying a familiar social role, that of the prophet, but so what? (How consciously he's doing this, I couldn't guess.) As Scott Aaronson says in a different context, this is a familiar role because it is is something that works, given human psychology and social organization. "Even when the prophet exhorts us to reason, skepticism, and empiricism, he does so by hijacking a delivery system that is thousands of years old. And that is why he succeeds." The value of a prophet depends on what they prophesy. What Gore is saying about climate change is true, and his recommendations for what to do about it are reasonable; maybe not optimal, but reasonable, and certainly not presented dogmatically. He's calling on us to come up with an industrial infrastructure which doesn't qualify as a self-inflicted wound, not to repent of our whoring after the false gods of material progress and Mammon.

This is a technological issue, but it is also a political one, because it involves large, persistent, consequential externalities, which must be dealt with somehow if we are not to all be in a lot of pain; or, more realistically, a lot more pain than we are already stuck with. It is further a political issue because of the collective action problems, which arise from the way our choices are made in contexts of larger networks. (In large parts of the country, for the most part making a living entails driving from one suburb to another, which entails burning fossil fuels. Preferences don't enter into it.) It is finally a political issue because it involves competing interests, and politics is how we trade those off against each other.

The Wire, season 4

"No corner left behind." In some ways, the saddest part of the story yet.

Michelle Sagara, Cast in Secret

More on the magical power of words, names, and writing-on-the-body. (Previous installments.)

Patricia Briggs, Moon-Called; Blood Bound; Iron Kissed

Mind-candy contemporary fantasy, with auto repair and struggling with bills in addition to the shapeshifters, vampires, and the very unpleasant denizens of fairy-tales. WARNING/SPOILER: the ending of the third book is very and unexpectedly brutal. It works, narratively, and is not gratuitous, but it's not pleasant.

Catherynne M. Valente, The Orphan's Tales: In the Night Garden

This may well be the best fiction I read in 2008.

When not actually under Valente's words' spell, the thing I admire most about this is the sheer control of rhetoric it displays on her part; the multiple stories-within-stories-within-stories all have their distinctive voices and styles, which all seem appropriate to the teller and the tale. And then on top of that she manages to make them weave through each other and twist back on each other, in ways which change their meaning and hint at other, unrevealed connections.

I think I might have been directed to this by a post of David Moles's, but if so I can't find it again.

Denny Borsboom, Measuring the Mind: Conceptual Issues in Contemporary Psychometrics

A thorough, searching, sober, drily funny, restrained, and in the end mostly damning survey of psychology's attempts to measure mental attributes. (It did, however, make me look more favorably on Piaget and his balance-beam task.) Borsboom gives particular attention to the quite hopeless way in which the validity of measurement has been tackled --- which seems to have involved every consideration conceivable, except the one of whether what you are trying to measure exists, and can influence your measuring device. This is a scientific question which no amount of calculating Cronbach's alpha will answer. (This paper by Borsboom, Mellenberg and Van Heerden is a self-contained version of the argument about validity, and strongly recommended to those who are interested.)

Some familiarity with basic statistics, especially linear regression, will be helpful. So would some knowledge of philosophy of science, though it's probably less important, since Borsboom spends more time explaining it.

Brian K. Vaughan et al., Ex Machina: The First Hundred Days

Comic book. Man acquires super-powers; parlays them into a successful run for mayor of New York City; then gets to deal with the fun of being a very human mayor. First in a series; I've bought the others.

G. Willow Wilson and M. K. Perker, Cairo

Your basic magical-realist Muslim love/adventure comic book about the Arab-Israeli conflict, life in the City of Victory, the ancient Egyptian afterworld, journalism in authoritarian countries, Orientalism and its after-images, and Sufism.

Read on Aziz Poonawalla's recommendation.

Clifford A. Wright, A Mediterranean Feast

Or: Cooking with Fernand Braudel. I mean that fairly literally; this is a combination of cookbook with a thematic history of Mediterranean food, cooking and cuisine, strongly and visibly influenced by Fernand Braudel and the rest of the Annales school, which makes for a possibly unique reading expereince. Certainly there can be few cookbooks which so emphasize grinding poverty, famine and mass death. — Many of the recipes are very tasty, but Wright has consciously made no concessions to American tastes, or to the typical equipment of American kitchens.

Read, over the course of a year and a half, on Carlos Yu's recommendation.

The Wire, season 3

The rise and fall of the city of Hamsterdam. (Am I a bad person because I found myself rooting for Stringer Bell?)

Fred Van Lente and Ryan Dunlavey, Action Philosophers Giant-Size Thing Volume 2 and Action Philosophers Giant-Size Thing Volume 3

Great Thinkers in comic-book format. Highly variable, ranging from excellent (i.e. clear, accurate and funny: Aristotle, Hume, Spinoza, Descartes, Macchiavelli, Sartre, Aquinas, Wittgenstein) to weak (i.e. not very deep and not very funny: Confucius, Lao Tzu, Hobbes; and the chapter on Marx is a class of its own, propagating Maoist deviationism as it does). Probably still funny even if you have not read the philosophers in question. (Volume 1 not reviewed becaue they didn't have it at my local comics store.)

Clark Glymour, Thinking Things Through: An Introduction to Philosophical Issues and Achievements

As I recently mentioned, in Larry Laudan's great philosophical dialogue, Science and Relativism, the positivist position is represented by one Rudy Reichfiegl, author of Everyman's History of Philosophy: Great Thinkers from Frege to Carnap.

This is not quite that book, for two reasons. First, Glymour was the student of one of the students (Wesley Salmon) of Hans Reichenbach, so he can't also be Reichfiegl. Second, it's even less historically-oriented than a book of great thinkers would be. But as an introduction to the kind of philosophy which can claim achievements, it's first-rate.

(Disclaimer: Glymour teaches at CMU, and in fact his office is one floor below and a little way up the hall from mine. On the other hand, I read this seven years ago, long before I came here.)

Warren Ellis and Max Fiumara, Blackgas

Zombies! Yankee zombies! Gross Yankee zombies! (Just eating the brains would waste the rest, you know.) Contagious gross Yankee zombies!

Discuss: Ellis should adapt The Road as a graphic novel.

Books to Read While the Algae Grow in Your Fur; Scientifiction and Fantastica; Food; Writing for Antiquity; Philosophy; The Beloved Republic; Minds, Brains, and Neurons; Enigmas of Chance; The Progressive Forces; Islam

"Here, bookie-bookie-bookie!"

Would the persons (person?) who borrowed my copies of Bosq's Nonparametric Statistics for Stochastic Processes, and Sageman's Understanding Terror Networks, please give them back? I need them for my class.

Self-centered; Corrupting the Young

Glory and $500 (VIGRE-funded Undergraduate Research in Statistics at Carnegie Mellon, Spring 2008)

Attention conservation notice: a repeating announcement, irrelevant to anyone who isn't a student at Carnegie Mellon.

Hey, kid! Interested in winning eternal intellectual glory and entering the glamorous world of scientific research? Interested in $500 for the semester? Are you an undergrad at Carnegie Mellon University? If so, the statistics department has no less than five possible projects for you. Apply now!

Engimas of Chance; Corrupting the Young

MSS. in Preparation, and One-Way Hiatus

Speaking of "comments welcome", there are now pages for two works-in-progress, The Statistical Analysis of Complex Systems Models, and Almost None of the Theory of Stochastic Processes, with Aryeh (Leo) Kontorovich. (Almost None is the latest incarnation of my lecture notes for advanced probability.) Comments that prod me into working on the manuscripts more often are especially welcome.

And speaking of that, it's once again the point in the academic year when to keep from spending all of my time procrasting by reading blogs I delete my RSS reader until the semester's teaching is done. So the next few months, posts you especially think I should see should be brought to my attention either by e-mail or del.icio.us.

Self-Centered; Engimas of Chance; Complexity

Chaos, Complexity, and Inference (36-462): Syllabus

There is, finally, a detailed syllabus for my course. Comments welcome. (I know it's a lot of reading, but since we're a real university and not a glorified finishing school or agricultural junior college, I think the students will be able to handle it.)

Update, 15 January: lecture notes will be appearing in the syllabus as the course goes on.

Corrupting the Young; Complexity; Enigmas of Chance; Networks

"Networks in Political Science" Conference

Via James Fowler in e-mail:

Call for Papers: Conference at Harvard on Networks in Political Science

The study of networks has exploded over the last decade, both in the social and hard sciences. From sociology to biology, there has been a paradigm shift from a focus on the units of the system to the relationships among those units. Despite a tradition incorporating network ideas dating back at least 70 years, political science has been largely left out of this recent creative surge. This has begun to change, as witnessed, for example, by an exponential increase in network-related research presented at the major disciplinary conferences.

We therefore announce an open call for paper proposals for presentation at a conference on "Networks in Political Science" (NIPS), aimed at all of the subdisciplines of political science. NIPS is supported by the National Science Foundation, and sponsored by the Program on Networked Governance at Harvard University.

The conference will take place June 13--14. Preceding the conference will be a series of workshops introducing existing substantive areas of research, statistical methods (and software packages) for dealing with the distinctive dependencies of network data, and network visualization. There will be a $50 conference fee. Limited funding will be available to defray the costs of attendance for doctoral students and recent (post 2005) PhDs. Funding may be available for graduate students not presenting papers, but preference will be given to students using network analysis in their dissertations. Women and minorities are especially encouraged to apply.

The deadline for submitting a paper proposal is March 1, 2008. Proposals should include a title and a one-paragraph abstract. Graduate students and recent Ph.D.'s applying for funding should also include their CV, a letter of support from their advisor, and a brief statement about their intended use of network analysis. Send them to networked_governance [at] ksg [dot] harvard [dot] edu. The final program will be available at www.ksg.harvard.edu/netgov.

Program Committee: Christopher Ansell (UCBerkeley), James Fowler (UCSD), Michael Heaney (Florida), David Lazer (Harvard), Scott McClurg (Southern Illinois), John Padgett (Chicago), John Scholz (Florida State), Sarah Reckhow (UCBerkeley), Paul Thurner (Mannheim), and Michael Ward (University of Washington).

Had I been consulted, I might have suggested not chosing the same acronym as one of the premier conferences in machine learning, especially not after this, but the mere fact that this didn't occur to anyone suggests that the communities are distinct. In any event, this looks very promising.

Networks