AMORY LOVINS CLAIMS nuclear power was born in a climate of "mutual intoxication," in which the Atomic Energy Commission and the utilities persuaded each other that "it was just another way to boil water." Belatedly, he says, they realized that it was far more than that. He says 40 years of experience have revealed that nuclear power's shortcomings lie in three broad categories:

* Nuclear power is not economical. Lovins says producing electricity with a new nuclear power plant costs approximately 10 to 25 cents per kilowatt hour, compared with about 5 cents per kilowatt hour for new wind turbines, small dams or cogeneration (in which a factory captures waste heat to make electricity).

"These technologies not only cost less, but they are less financially risky, because they are smaller, cheaper, and built faster. When you rely on a technology such as nuclear power that takes a decade and billions of dollars to build, you are basically playing 'You Bet Your Utility Company' that your forecast of demand 10 or more years down the road will be right. But our forecasting record is abysmal, which shows the advantage of these build-as-you-need, pay-as-you-go modular systems."

Lovins says that is one reason why from 1981 to 1984, dozens of orders for large nuclear and coal plants were canceled -- 65,000 megawatts' worth -- while at the same time there were 45,000 megawatts of new orders for small wind farms, small dams and cogeneration.

Lovins contends nuclear power persists mostly through government largess. "In 1984, it got about $15 billion in federal subsidies out of a total subsidy to energy of about $46 billion," he says, pulling figures from an article he wrote for The Wall Street Journal. "To put it another way, a dollar of subsidy to nuclear delivered about 1/80th the energy as a dollar of subsidy to efficiency and nonhydro renewables.

"Given the very unlevel playing field on which these technologies have competed, it is all the more impressive that efficiency and renewables have lately swept the market, while no new nuclear plants have been ordered since 1978 Potomac Electric Power Co., for example, in 1977 scuttled plans for two nuclear plants that would have cost $1.8 billion . It makes you wonder how much faster this would happen if they were allowed to complete fairly."

* It is accident prone. In 1979, a Freedom of Information Act request by the Union of Concerned Scientists unearthed "The Nugget File," a secret, Nuclear Regulatory Commission file composed of thousands of safety-deficiency reports at U.S. nuclear plants. Lovins' favorite is a 1969 report detailing the time a 3,000-gallon radioactive waste tank was attached to the drinking fountain at an unnamed plant. The file is a litany of sensitive equipment that was frozen, burned, flooded, dirtied, corroded, bumped, dropped, overpressurized, unhinged, miscalibrated, miswired or poorly designed in the first place. As one official noted in a report, "the absence of more serious effects is largely the result of good luck."

"What impresses you when you go through the literature is that these are very complicated machines run by people who are fallible, like oneself," says Lovins. "It is wrong to look at nuclear safety as an engineering problem. It is fundamentally a people problem." Though reactor mishaps such as the ones at Chernobyl, Three Mile Island, and the Davis-Besse plant in Toledo (which came within 90 minutes of a core-damaging accident last June) have gotten the most press attention, Lovins believes the longest term accident potential comes from the difficulty of continuously and indefinitely containing 14,000 tons of high-level waste; an amount that the Worldwatch Institute estimates will quadruple in the next 15 years. The wastes remain toxic for more than 10,000 years -- or, as Lovins puts it, "time periods that are more theological than geological."

He says the nuclear industry's declining public image is worsening the safety problem, since "for some years now, the best graduate students have not dreamed of going into nuclear engineering. Indeed, the best half of the people in the industry have already left . . . in a real sense, the business is slipping into ever less competent hands." Add to that what Lovins calls "a rubber-stamp-approval NRC," and the result is "a good chance for all-American Chernobyls in the near future."

* It jeopardizes national security. In their 1982 book Brittle Power, the Lovinses laid out their thesis that "the main driving force behind the spread of nuclear weapons to countries that don't already have them is nuclear power, because it provides the materials, skills, equipment and, above all, the innocent civilian cover for bomb programs." Lovins says the 1981 Israeli air strike on an Iraqi reactor indicated a growing recognition of the fact by world governments.

On the positive side, says Lovins, "The collapse of nuclear power from an incurable attack of market forces offers us a wonderful opportunity to start unspreading the bomb."

A second security danger, says Lovins, is that "it only takes low-technology sabotage to release most of the radioactivity in our nuclear plants. There have been well over 100 substantial breaches of security at nuclear plants around the world -- we are just lucky that most of them were meant as theater rather than to cause serious damage."