RATHER THAN COMPARING the risks of producing nuclear energy with the risks of producing other kinds of energy, many opponents of nuclear power have come up with a special argument in the aftermath of Three Mile Island. It is that such plants must be unsafe because those who build and operate them insist on setting a limit to their monetary liability if there is an accident. The argument rests on a false logic that has nothing to do with the serious kind of reappraisal that should be made now that Three Mile Island has happened.

The implication here is that no company should ever undertake a project unless it is prepared to face the ultimate insurance consequences of the worst disaster that project could conceivably cause. And, in the affairs of most companies most of the time, this is the true. But it is not always in force - especially not when the government wants something done. The government, for instance, routinely guarantees loans and underwrites contracts because it wants certain projects to go forward.

The limitation on nuclear liability, known as the Price-Anderson Act, came into being 20 years ago because the government wanted nuclear power developed and wanted private enterprise to do the developing. The utilities said then - and it still may be true - that they could not raise the huge amounts of money needed to build those plants - at least not without a limitation on liability for which they could buy insurance.

That judgement had more to do with the risks investors perceived in atomic energy than those the experts foresaw. Atomic energy was then a promising but undeveloped technology and the financial risks the utilities took, even with the liability limitation in place, were pretty big. Without that limitation, the utilities thought the investors would simply shy away. It is possible, as some people claim, that without Price-Anderson there would be now be no nuclear plants. But it seems more likely that if private enterprise had not built those plants, the government would have done so. The enthusiasm for developing atomic energy was so great in the 1950s that it was not about to be abandoned.

The real issue about these plants does not concern how much exposure to financial risks a utility has in case of an accident. It concerns how dangerous these plants are - by themselves and also when compared with the alternatives.

Unfortunately the country is only now beginning to stare at the real - not just the dollar - costs of energy: air and water pollution, radiation, land destruction, injuries to workers and the rest. It is in this context, not that of insurance liability, that energy must be judged.