I'm not going to enter the dispute about whether it was Albert Einstein, Benjamin Franklin or someone else who first defined insanity as doing the same thing over and over again, expecting different results. I'll just suggest that the U.S. government's program to build miles - and then more miles - of fences along the U.S.-Mexico border in hopes of stemming illegal immigration fits the definition well enough that therapy seems in order.

In recent months, the Department of Homeland Security has been fast-tracking the construction of walls and fencing along several stretches of Arizona's southern border, in accordance with the Secure Fence Act of 2006. There are many reasons to question the current border-fencing mania, but the most important is the most obvious: It has been tried before, and it has failed. Immigrants still climb under, through and over the long-standing border fences below San Diego, and from the Arizona border to El Paso, walls and fences haven't so much reduced illegal immigration as moved it to other, unfenced crossing points.

In theory, of course, the government could try to fence all 2,000 miles of the country's southern border. But the estimated construction costs for such a project run into the tens of billions of dollars. In places - mountain ranges, canyons and riverbeds - construction of a wall or fence would approach impossibility. If a full-border fence ever were built, it would still have to be patrolled, at a cost of billions of dollars, with billions more added every year. And migrants would still go over, under and around it.

Beyond its huge cost and tiny utility, the border-fencing project will have a variety of negative unintended consequences. Cultures will be split, from the cross-border megalopolis of El Paso/Juarez to the isolated lands of the Tohono O'Odham. Economies that have been entwined for decades will be divided, as border crossing - legal and illegal - becomes more time-consuming.

Perhaps the most pernicious side-effects of border barriers - which, according to recent news reports, will grow to a total of some 670 miles in the Southwest by the end of next year - will be environmental. As Jeremy Voas writes in our cover story, "Cat Fight on the Border," walls and fencing on the Arizona-Mexico border will almost certainly sever migration routes for the jaguar. The border barriers will all but doom hopes that these once-indigenous cats might repopulate their former Southwestern range without a contentious reintroduction program.

And the jaguar is just one of many species whose habitats will be bisected by the fence - a multibillion-dollar fit of temporary political insanity that Congress should quickly snap itself out of.

Inevitably, when I write about the futility of border walls, I receive bushel baskets of letters from the seal-the-borders, deport-them-all crowd. Inevitably, they accuse me of being soft on illegal immigration, but they are wrong. I just happen to believe that border walls and fences are expensive and useless.

My belief is not ideological, but evidence-based. For a sample of that evidence - and before you send that scathing letter accusing me of being a leftist alien-hugger - you might take a look at the Arizona Daily Star's four-part investigative series, "Sealing Our Border: Why it Won't Work." The Star sent six reporters on a wide-ranging expedition along the length of the U.S.-Mexico border. Their findings (which can be found online at http://www.azstarnet.com/secureborder/) seem, to me, well-presented and definitive.