Without the law of unintended consequences, columnists would have a lot less to write about. Many of the problems with health care result from the fact that medical insurance is commonly linked with employment. Nobody would have set out to design a system in which someone who loses his job because he's too sick to work loses his health insurance too. But that's the system we have, as an unintended consequence of World War II wage and price controls followed by decades of high marginal income-tax rates. Companies had an incentive to offer, and employees to demand, compensation in the form of tax-free fringe benefits. Welfare-as-we-know-it, the sad cases of families that have been on welfare for generations with never a husband or a job in sight, is an unintended consequence of a Depression-era program designed to help widows. When it started in 1935, the program then called Aid to Dependent Children was primarily intended to help widows, and mostly that is what it did; 81 percent of recipients were widowed. In just one generation, by 1960, widows were only 8 percent of ADC recipients, and not because widowhood had become rarer. Divorce, desertion and illegitimacy all became more common. Illegitimacy continues to climb, from 5 percent in 1960 to 30 percent in 1992, and welfare is following it on up, from 3.7 million recipients in 1960 to 14.1 million in 1993. Certainly nobody intended that. Probably nobody could have foreseen it, or would have been believed if he had. Those figures are cited by Ben Wattenberg in "Values Matter Most." He calls the phenomenon "LUSE," the Law of Unintended Side Effects, but that tends to obscure the fact that the unintended effects may be vastly more consequential than the intended ones. It's enough to turn policy-makers humble, if humility were in the nature of policy-makers. But deliberate policy is not the only place where the results people aren't aiming for turn out to be more significant than the ones they are. Technology can exact an unanticipated price, and that's the topic of historian Edward Tenner's new book, "Why Things Bite Back." He calls it a revenge effect, "the tendency of the world around us to get even, to twist our cleverness against us." He's not anti-technology, just skeptical. As technology develops into more complex systems, the separately conceived parts of the system interact in ways the designers don't foresee, and certainly don't intend. Social factors play a part, too. "When a safety system encourages enough additional risk-taking that it helps cause accidents," Tenner says, "that is a revenge effect." Sometimes the revenge is catastrophic. The Iroquois Theater in Chicago opened in 1903, even though its sprinkler system wasn't working yet, because it had been designed to be fireproof. Then it burned, and more than 600 people died. But catastrophe tends to concentrate people's minds wonderfully, and that leads to improvements. Air travel, for instance, has become steadily safer even though it is already far safer than driving. So if airlines begin to require families to buy tickets for their very young children so the children can fly safely strapped in, some families will elect to drive instead, and statistically at least, more children will die in travel accidents. That's revenge, all right. The methods used of preventing catastrophes, though, may lead to greater catastrophes in the future. That's happened with fire suppression. Instead of frequent, small fires, which tend not to do much damage because the amount of available fuel is limited, an area may go unburned for decades and then be devastated by a fire that is uncontrollable because of the buildup of fuel. In other cases, the efforts to solve problems simply alter them, he points out, trading the possibility of a sudden, shocking disaster for a gradual, long-term dispersed problem that is far less tractable. In medicine, Tenner says, one result may be to replace acute illnesses with chronic ones. Tenner devotes several chapters to medicine, because, he says, people have never been healthier — or more anxious about illness. Despite the obvious improvements in medicine, whether you think of drugs or surgery, people don't necessarily feel better about being cured. More advanced medical techniques demand more of medical practitioners, and allow for a whole new class of things that can go wrong. The promise of better treatment for an ailment may also lead patients to seek relief where previously they'd have decided to grin and bear it. Patients perceive that laparoscopy will be easier on them than traditional surgery. So much more of it is done, with higher rates of complications. Drug resistant pathogens, health problems resulting from excessive dieting, low-weight babies who survive but with multiple handicaps — there are many medical problems that are the unintended consequences of generally beneficial medical improvements. The other topics Tenner mines for examples are the environment, office procedures and athletics. He finds the same patterns in all of them, and illustrates successfully that this is a useful way to think about innovations. Makers of social policy might take note.