Steven Leinwand, who sits on the committee overseeing President Clinton’s proposed national mathematics exam, has written an essay explaining why it is “downright dangerous” to teach students things like “six times seven is 42, put down two and carry the four.” Such instruction sorts people out, Mr. Leinwand writes, “anointing the few” who master these procedures, and “casting out the many.” As Mr. Leinwand tells it, there might once have been an excuse for such undemocratic goings-on, but we can now, because of technology, throw off “the discriminatory shackles of computational algorithms.”

House and Senate conferees who will soon be deciding what to do about the Clinton plans for national testing ought to read Mr. Leinwand’s essay, in part because it helps explain why the committee on which he sits recommended a national math exam that would avoid directly assessing “certain knowledge and skills such as whole number computation.” And in case the exam might indirectly assess whether eighth-graders can add, subtract, multiply and divide, the committee recommended that every student be armed with a calculator throughout.

But the most important reason for conferees to read this essay is to gain an appreciation for the kind of thinking that has become all too common in the educational establishment. Although his ideas might seem extreme, Mr. Leinwand is not a marginal figure. He is not only on the committee overseeing the president’s proposed math exam; he serves as a consultant to the Connecticut Department of Education, sits on the board of a $10 million National Science Foundation math program and advises a standards-setting project funded by the Pew and MacArthur foundations.

Since 1989, when the National Council of Teachers of Mathematics set forth a radical vision for how mathematics should be taught, ideas like Mr. Leinwand’s have increasingly become the order of the day. Mathland, an instructional program widely used in California’s elementary schools, never does show kids the standard U.S. procedure for multidigit multiplication. But, in a fit of multiculturalism, it does offer instruction on the very complicated way in which the ancient Egyptians managed these matters.

In the current debate, many otherwise sensible senators have been convinced that safeguards can be put in place to keep harmful fads from influencing standards and assessments. Recent history does not support this optimism. A few years ago, as chairman of the National Endowment for the Humanities, I awarded a contract to develop national history standards. Although I required detailed plans from the contractor and had them thoroughly reviewed by knowledgeable people, the standards that were finally delivered were so suffused with political correctness that I felt obliged to condemn them--as did 99 members of the Senate.

The federal effort to set English/language arts standards produced such a muddle of trendy thinking that in 1994 the Department of Education cut off funding. Late last week, Secretary of Education Bill Riley backed off from the math test into which his department has recently poured thousands of hours and hundreds of thousands--perhaps millions--of dollars. Calculator use should be narrowly restricted, the secretary said.

Some in the Senate advocate turning national testing over to the National Assessment Governing Board, a bipartisan group appointed by the secretary of education, but that won’t prevent foolish ideas from making their way into national tests. In order to fulfill the vision of the National Council of Teachers of Mathematics, the board reduced the computational part of its 1996 National Assessment of Educational Progress math exam by 20% for 17-year-olds and increased the portion of the exam on which they can use calculators to well over a third.

Meanwhile, there is good work going on in the states: sound history standards in Virginia and Texas, admirable English/ language arts standards in Massachusetts. To be sure, there have been fiascos, most notably with reading and math standards in California, but pressure from concerned parents has turned California’s reading program around, and there has been significant progress in restoring to the state a mathematics program strong in the basics.

Setting standards and tests at the state level is no guarantee of success, but the accomplishments there certainly outshine the federal record, and the good work that has gone on outside of Washington could very well be rendered moot by a test created inside the Beltway. Even if a certain state decided not to participate in the Clinton testing plan, a federal test would strongly influence the textbooks used in that state’s schools and determine the way its teachers are trained. And if that test is the disaster that the record indicates it will be, the result could be a national calamity.

The president has threatened to veto the labor, health and human services appropriations bill if Congress blocks his plans for national testing, and many members of the Senate are hesitant to oppose him, particularly since a Wall Street Journal/NBC poll that the administration has been passing around shows overwhelming public support for national testing. But Sen. John Ashcroft (R., Mo.), trying to rally his colleagues to kill the Clinton plan, cites another part of the same poll. It shows that when pollsters explain that the federal government would establish the test, and spell out standard pro and con arguments, support for national testing drops to less than half.

“We should test our kids,” Sen. Ashcroft observes. “We need that accountability in education. But what we don’t need is the federal government coming up with the tests we use.”