Few will argue against the Defense Department saving money. And when the Pentagon announces it can save $8.56 billion over the next eight years by merely changing the way it issues contracts, the first response is to cheer it on.

But the idea of expanding the much-ballyhooed program of "multi-year procurement contracts" hardly seems the right medicine for the current defense budget indigestion.The multiyear contracts, which have the general support of the Bush administration, would use economical, long-term contracts to buy 32 major weapons systems in bulk quantities, providing stability to the program and preventing costly start-and-stop production.

Ideally, more of the contracts would be paid in the early years as contractors purchase parts in quantity. In theory, the later years of the contract would cost less.

At least that's the idea. Now it appears Pentagon claims of savings are not only wildly exaggerated but are based on the purchases of at least four untested or unreliable weapons systems, according to Defense Department documents.

A growing number of critics are concerned that the multiyear procurements could end up costing taxpayers more every year instead of saving them money, because it binds the government to a long-term purchase agreement regardless of whether the weapons system works.

Those opponents of the current proposal before Congress say they are not opposed to long-term contracts per se but only to long-term contracts involving weapons systems that have not been proven reliable. And that is a very legitimate concern.

Just as most people would never buy a new car without first test driving it, Congress should not be put in a position of buying any weapons system without making certain it's a good one.