The Undercover Philosopher’s Blog

Mon, 2 Mar 2009 13:56:33 -0800

The “Science” of Economics

According to economist James Galbraith, just ten to twelve of the fifteen thousand professional economists in this country foresaw the current mortgage crisis. This tells us a lot about the scientific status of economics.

The so-called science of economics begins with an untested and in fact absurd assumption, namely human beings will make rational economic decisions. Well, actually, the assumption is that they will make rational decisions if they have perfect knowledge of the market. By “rational” decisions, economists mean decisions that best promote one’s own economic self-interest. This is a ridiculously narrow idea of rationality, but never mind that. The problems with prediction lie elsewhere.

When it comes to model making, economists typically assume that all players do have perfect knowledge of the market. They recognize that this is an idealization, but apparently believe it comes close enough to reality to be useful. In fact, it’s laughable. No one even comes close. Imagine all the time and energy it would take to acquire perfect knowledge of the market, that, even in relation to soaps, sponges, carrots and other items on a typical shopping list. And think about the difficulty investors face when they try to determine the financial health and future of the companies they might invest in. This problem is complicated by the fact that one can’t trust a company’s balance sheets (see the section on accounting fraud in _The Undercover Philosopher_).

But even give accurate information about markets, it is ridiculous to suppose people will make rational economic decisions. As every trader knows, investors are emotional, and markets are heavily influenced by both fear and mania. It is truly astonishing that economists have only very recently started to acknowledge that human beings have irrational attitudes toward risk, or, in any case, that our attitudes toward risk do not conform to their own professional understanding of rationality. But even economists who take this seriously have not figured out how to incorporate this new understanding into most of their models (we owe this new understanding to the work of Amos Tversky and Daniel Kahneman).

Since mainstream economic theory is heavily mathematical and highly rigorous, most people assume that it must accurately represent some reality. It is, in fact, a kind of applied mathematics (beginning, as it does, with a priori assumptions and deducing consequences based on those assumptions). But applied mathematics is useful only to the extent that it accurately models some aspect of the world. And economic theory fails miserably to represent the aspect of the world it is intended to model. In fact, it is less a science than a branch of applied mathematics in search of a domain (or subject matter). Well, that’s a little too strong. It does work well for the kind of highly constrained choice situations expert players face in Monopoly and other board games.

Why Only Psychiatrists?

Fri, 19 Dec 2008 15:02:33 -0800

Did you ever wonder how psychotherapists and psychiatrists diagnose “mental problems”? Well, they rely on an authoritative book called the Diagnostic and Statistical Manual of Mental Disorders. That book is published by the American Psychiatric Association. The association selects psychiatrists who meet small groups to discuss particular problems areas (for example depression and other mood disorders and substance abuse). These groups report their conclusions--or divided opinions--to a 28 member panel of psychiatrists, also selected by the association, who decide what goes in the manual. Their decisions have an enormous social impact. Doctors, psychotherapists, social workers, judges and others rely the manual to identify mental disorders. Insurance companies rely on the manual to decide what to cover (what kinds of treatments for what kinds of conditions). The manual is now in it’s fourth edition and soon to be in it’s fifth. There is an article in the December 19 New York Times (front page) about some of the heated debates surrounding the creation of the next version.

The fact that the authors of The Diagnostic and Statistical Manual of Mental Disorders (or DSM, as it is called) need to reach agreement on controversial issues within the profession is a clue that psychiatry is not a science. When there are controversies within a science, the scientific community acknowledges that there are competing theories and looks for evidence or experiments to decide the issue. It does not try to settle the question by issuing an official document. One reasons psychiatry settles issues by fiat in this way is that many of the controversial issues can not be decided by experiment or evidence. That’s because they are based (at least in part) on value judgments. Mental disorders, after all, are at bottom just patterns of thinking, perceiving, imagining, feeling, and acting that are in some way harmful or dangerous to the person who does them or harmful or dangerous to others. Of course, we all have tendencies that are harmful or dangerous. So the question is: at what point do these tendencies become harmful or dangerous enough to ourselves or others to deserve to be called “mental disorders”, with all the social implications that category carries. And that is a value question.

For example, delusional thinking is an element of many mental disorders. But we are all delusional in various ways (in denial or otherwise self-deceptive, overly optimistic, overly pessimistic, etc.). Roughly, these delusions become elements of a mental disorder when they are harmful or dangerous enough (by themselves or combined with other elements of a syndrome). This is rough because we also make exceptions based on our values. If I give away all my family’s property on the demand of Harvey, the invisible six foot rabbit, I am likely to be diagnosed with a mental disorder (especially if I also exhibit a few other symptoms). But if I give it all to a church at the behest of Jesus or my guardian angel I am simply devout (See “Who’s Crazy?” on the Excerpt page). Because we value religious freedom, we tend not to classify religiously based thoughts, perceptions, or imaginings as manifestations of mental disorders even when they are harmful. Again, people who reject medical treatment for diseases, starve themselves, whip themselves, and give away all their belongings because God told them to do are not regarded as suffering from mental disorders. Whatever else we might say about Islamic terrorists, we do not consider their actions manifestations of mental disorders. Instead, we hold them responsible for their acts and put them on trial like ordinary criminals.

Among other things, the authors of the DSM are now trying to decide how to deal with sexual identity issues. Does “a man born in a woman’s body” have some kind of mental disorder? The transgender community is apparently divided on this issue. Most are offended by attempts to classify their condition as a mental disorder. That suggests that there is something wrong with them. On the other hand, many of them also seek medical treatment for their condition that they would like to have covered by medical insurance plans. That would be easier if their condition were classified as a mental disorder. Is there something wrong with transgender people? Should health insurance cover transgender surgery? What should our social attitude toward these people be? Again, these are value questions. They need to be informed by the facts. But they are not settled by the facts.

This is not to say that there is no science in psychiatry or that the DSM is nothing but a series of value judgments. It is just to say that classifying a condition as a mental disorder involves making value judgments. And since these classifications have important social consequences, the question arises: why are these decisions placed exclusively in the hands of psychiatrists? Of course, psychiatrists have important contributions to make. They have knowledge of the facts that should inform these value judgments. But they have no expertise when it comes to making the judgments themselves, and it is ridiculous to leave those judgments in their hands. People suffering from psychoses may be locked up against their will, deprived of civil liberties, and forced to take drugs against their will. Why should psychiatrists decide when they are dangerous enough to themselves and others to warrant that treatment? And why should psychiatrists decide who’s treatments should be paid for by medical insurance? Again, they may provide us with the facts relevant to deciding these value questions, but they are value questions, we all have a stake in deciding them, and we (the public) should all have a voice in making them.

The first edition of the DSM (1952) classified homosexuality as a mental disorder. That persisted until 1973, when that diagnosis was replaced by “sexual orientation disturbance”, which later became “ego-dystonic homosexuality”. Ego-dystonic homosexuality was finally dropped altogether in 1987. Psychiatry no longer views homosexuality as a disorder. Some say that this is because we have new information about homosexuality; most importantly, that there is a genetic predisposition toward it. But that is not the reason. After all, a condition may be genetic and still be a disability, disease or disorder even in psychiatry (there may also be genetic predispositions to some psychoses). No, the real reason is a change in values. Although homosexuals are very much in the minority--that is, homosexuality is not normal, in the statistical sense--most of us no longer regard it as unnatural. Instead, most of us now regard it as an acceptable expression of human diversity. Our values are more humane than they were before.

Wednesday, December 10, More on scientific fraud in health related areas

Wed, 10 Dec 2008 16:17:39 -0800

Scientific fraud threatens us all, especially in health related areas. I had space for only a short discussion of that problem in the book. For a powerful and detailed account of how industry paid scientists massage data to defend products and practices that threaten us all see David Michael’s book Doubt is Their Product: How Industry’s Assault on Science Threatens Your Health (Oxford University Press, 2008). Some of these techniques are nicely summarized by Sharon Begley in a Newsweek review of the book that appeared in the May 12, 2008 issue titled “Whitewashing Toxic Chemicals” (you can easily access this by going to the Newsweek archives or, I hope, by clicking this link http://www.newsweek.com/id/135408 ).

Wednesday, December 10, More “scientific” nonsense about will power

Tue, 9 Dec 2008 15:16:58 -0800

There is an entry in the index of my book that reads “study, astonishingly bad”. The study in question is featured in a New York Times op ed piece on April 2, 2008 titled “Tighten Your Belt, Strengthen Your Mind”. The authors of that piece are Sandra Aamodt and Sam Wang. She is editor in chief of Nature Neuroscience and he is an associate professor of molecular biology and neuroscience at Princeton. They cite the astonishingly bad study as the main support the conclusion that (in effect) will power is like gas in a tank--the more you use it, the less you have left over to fight temptation. I cite that study in my book to illustrate one feature of really bad science--the failure to consider alternative interpretations of the data.

Here’s how the study worked. The experimental subjects enter a room in which there is a table. On the table is a platter of radishes and a platter of cookies. The subjects are told they can eat from one of these but not the other. The scientists carefully note who ate from which and how much. After that, the subjects are asked to perform a task that is in fact impossible to perform (a paper and pencil task). The scientists carefully note how long it takes for each subject finally to quit. After analyzing the data, they noticed that the radish eaters tended to quit sooner. They took this to establish that will power is like gas in a tank. Because the radish eaters used up some of their reserve to resist the cookies, they had less left over to persist with the impossible task.

Are there other possible explanations? Might there be personality differences between the radish eaters and the cookie eaters that explains the difference? Might the radish eaters and the cookie eaters be in different moods? Might differences in blood sugar levels been significant? The authors of the op ed piece actually talk about blood sugar levels, suggesting that lower levels reduce will power (as any really hungry person knows). But they don’t seem to realize that this undermines the conclusion of the radish/cookie study. It’s not that the radish eaters used up will power by resisting the cookies, it’s just that they have lower blood sugar levels as a result of not eating them.

What’s really depressing about all this is the way that bad science makes it’s way into the public consciousness. In this case, the bad science also flies in the face of common sense. If the conclusion of the study were correct, we should expect people who give in to one temptation to have an easier time with the next (given they have conserved their will power). Tell that to the binge eaters. We should also expect people who are quite disciplined at the beginning of the day to start breaking down toward the end. Tell that to the Marines.

This oversimplifies the matter a bit because it’s not clear in the op ed piece over how long a period of time all this is supposed to operate. In the cookie/radish study the time lapse between one temptation and the next is short. The authors of the op-ed piece seem to expand it a bit (if you don’t want to drink too much at a party, don’t use up will power by window shopping without buying on the way over, they suggest; they might also have advised caving in to any temptation one encountered on the way over to preserve will power for the party).

The authors of the op-ed piece also claim that studies show that we can increase our will power by exercising it. That may well be true. In fact, it sounds like common sense. But the example they cite in it’s defense is also surprisingly weak. “People who stick to an exercise program for two months report reducing their impulsive spending, junk food intake, alcohol use, and smoking. They also study more, watch less television and do more housework.” This is another example of not considering alternative explanations. Could it also be that exercising regularly makes a person feel better (endorphins,anyone?) and reduces the need to shop, eat, drink or smoke his or her way to satisfaction? As far as studying and housework go, could it also be that exercising regularly is energizing? That’s certainly been my experience and the experience of everyone else I’ve talked to who exercises regularly.

All this reinforces my conviction that all Phd students in the sciences should be required to take at least one course in the philosophy of science taught by a philosopher.