Recommended reading, II

One positive impact blogging has had on me is that it's made me read more books. Some were specifically mentioned on blogs I read, and others seem like they'll help flesh out some of the ideas and topics I've been reading about and debating online.

One particularly enjoyable recent read was the widely-discussed Freakonomics, in which economist Steven D. Levitt applies quantitative analysis to unlikely questions, like whether Sumo wrestling matches are sometimes rigged. (Answer: Yes.)

However, the old saying notwithstanding, if you're going to write a book about unconventional, outside-the-box thinking, it really shouldn't copy someone else's cover:

Logic and its Limits Shaw 1997 (2nd edition)	Freakonomics Levitt, Dubner 2005

FOLLOW-UP:
A more substantive criticism of Freakonomics is Levitt's tendency to gloss over or ignore evidence that undermines his arguments. To his credit, however, he has started a blog where he acknowledges (although does not always fully address) criticisms.

In this post, titled "Does Freakonomics Suck?", he links to a detailed and persuasive rebuttal of his contention that legalization of abortion post-Roe v. Wade led to a decline in crime rates in the 1990's, as fewer unwanted children were around to turn into criminals.

I have similar problems with Levitt's offhanded dismissal of the idea that gun ownership deters crime, because criminals are less likely to attack people who might be armed. He mentions one researcher who found such a link, but then says there were "allegations" that he forged his data. Well, people can "allege" whatever they please, and often do, especially on contentious issues like gun control. That doesn't mean their allegations, without any substantiating evidence, are valid.

Levitt then mentions that other researchers failed to find that gun ownership deterred crime, without naming them, identifying their studies, or assessing the quality of their findings. He also fails to consider the possibility that gun ownership might deter certain types of crime, like burglary, more than others, like battery. Or that below a certain level of gun ownership, the risk that potential victims might be armed could be too low to deter criminals.

In short, Levitt is quick (and correct) to attack society's overreliance on "conventional wisdom", yet it seems he's not above creating a little conventional wisdom of his own. Perhaps the best thing I can say about Freakonomics is that I took its message of skepticism to heart and instinctively applied it to the book itself.

ANOTHER FOLLOW-UP:
I emailed Levitt about the similarity in covers, and he was gracious enough to respond, pointing out that he addressed the matter on his own blog in this post back in April. It reads:

Chika Azuma, a cover artist at William Morrow, created the present cover. According to the book jacket, it was derived from a photo collage by James Meyer/Getty Images; the inside orange slice is credited to Jan Cobb. Our guess, therefore, is that the apple/orange existed as some form of clip art, likely the same source used by another cover artist for another book cover.

Another mystery solved.

YET ANOTHER FOLLOW-UP:
In the notes at the end of the book, Levitt does cite two researchers who claim that increased gun ownership does not lead to lower crime levels. But he does not describe or evaluate their findings, and I doubt most readers will have the resources or inclination to dig up old issues of the Stanford Law Review or the Journal of Political Economy on their own.