IN PRACTICALLY every state with a significant black population, congressional and legislative redistricting plans have been challenged for violating the standard of "racial fairness." That test is imposed by both the Voting Rights Act, which covers most of the South and some parts of the North, and the constitutional amendments passed during Reconstruction to guarantee blacks the right to vote. Over the years the test of racial fairness has produced significant gains for blacks. Recently, however, there seems to be some confusion about what the law requires.

The standard rule, followed recently by courts in Illinois and Missouri, is that racial fairness requires maximizing the number of black-majority districts. But a Texas court, in a decision reversed on other grounds, decided that racial fairness required splitting a black-majority district created by the legislature and spreading blacks out over two districts, in order to assure black influence in a maximum number of districts.

Plausible arguments can be made for both results. Which argument you accept depends on how you think the political process works. If you think that an area has what we would call racial politics --with all blacks voting for one set of candidates and all whites for another, and with white-supported candidates caring not a fig for what black voters regard as their interests--then racial fairness probably requires maximizing the number of black- majority districts.

If you suppose, however, that the politics of an area is not strictly racial, then you might decide that blacks might be better served by maximizing the number of districts in which they are a significant presence. If voters do not automatically regard black and white interests as mutually exclusive, then it is not only possible but likely that legislators with a significant number--35 percent, 25 percent, 15 percent, even 5 percent--of black constitutents are going to pay attention to what they want.

We think the evidence is clear that large parts of the country have moved from racial to non-racial politics since the middle 1960s. You almost never see monolithic blocs of white voters in the South any more or in major cities, nor do blacks vote monolithically any longer in cities like Cleveland and Gary, which were racially polarized a decade ago. Racial issues are no longer the major factor in the politics of southern states or most major cities. Nor do voters of either race assume that they can be represented fairly only by someone of their own background.

We do not mean that racial politics is gone; it is still with us in many corners of the rural South and in some racially changing cities and suburbs. But in many parts of the country, the Voting Rights Act has worked so well that blacks are part of the mainstream of American politics, with access to the ballot and proven ability to exert influence in the political process.

So in redistricting cases, it makes sense for courts to scrutinize district lines in situations where race is still the basis of politics, and to continue in those situations to maximize the number of black-majority districts. Where politics is non-racial, however, it makes sense for the courts to give much more deference to a legislature's districting plans, except in an egregious case where a black community is sliced to ribbons and left with negligible influence in any district. Otherwise, we think the courts can trust the political process to work, and can assume that black voters and politicians interested in black support will sooner or later assert with vigor and success any interests blacks believe they have in legislative district lines.

Over the years the courts have in fact done pretty much that. They have looked at a number of factors --history of discrimination, nominating procedures, election results, responsiveness of elective officials --to judge whether a voting system is discriminatory. Language making it clear that such tests should continue to be used--and specifying that failure to elect blacks by itself does not prove lack of racial fairness--has been added to the Voting Rights Act in the version which passed the House by a 389-to- 24 vote and which has 65 co-sponsors in the Senate. Unfortunately the administration, though it supports renewal of the Voting Rights Act generally, has opposed this language. Passage of the House version would make it clear that courts should consider the surrounding circumstances--which means, among other things, making an informed judgment about whether a community has a racial or a non- racial politics--in deciding whether a redistricting plan violates the principle of racial fairness.

One more thing needs to be said about the application of this principle to the redistricting process. The Justice Department and courts in states as diverse as New York, Virginia, Illinois and Texas have accepted the argument that no district can be counted as black-majority unless 65 percent of its residents are black. One should always take a second look at any argument developed by lawyers that would strike the ordinary fair-minded person as absurd, and this is one of them. It makes much more sense to use language literally and say that any district more than half of whose adult residents are black is black-majority under the law. If blacks are prevented unfairly from registering and voting, then the stringent and successful provisions of the Voting Rights Act can and should be invoked.

If the problem is only that blacks do not register and vote as frequently as whites, then the sensible solution is political, not judicial: blacks and others who want to see more blacks voting can conduct the kind of registration and get-out-the-vote drives that have enfranchised so many Americans in the past. The Voting Rights Act was not intended to be a substitute for the political process, but rather to enable it to work.