Nearly 50 years after the US Supreme Court ruled that "separate but equal" schools for blacks and whites were unconstitutional, a new legal and cultural climate is now causing many school districts to shift their focus to a different kind of segregation problem: that which separates wealthy students from poor ones.
This week, Cambridge, Mass., joined a small but growing number of cities, from San Francisco to Charlotte, N.C., that have decided to make income, rather than race, the primary factor in deciding which students go to which schools.
The shift is in part the result of a series of recent court decisions that have struck down some racial integration efforts as unfair to nonminority students or putting too great a burden on school systems. For many districts, income-based school placement represents a legal path toward racial diversity, since income and race are correlated.
But it also reflects a growing awareness that income, even more than race, is a strong predictor of academic achievement. Notably, students in schools with high concentrations of poverty are much less likely to succeed.
"Two factors are combining to really promote this as part of an important trend," says Richard Kahlenberg, a senior fellow at the Century Foundation, and author of a book on economic desegregation. "Quite frankly, a number of communities are trying to find a legal way to maintain racial diversity as a byproduct. But there are also a lot of very sound educational reasons to want to do this on its own."
Still, the plan is controversial. Many middle-class parents here worry their children might suffer academically if forced to attend a school that has traditionally had a high percentage of low-income students - and a low rate of academic achievement.
Under the new system, Cambridge will continue to let parents list schools they would like their child to attend. But the percentage forced into schools they did not select, now about 8 percent, is sure to rise.
Despite these concerns, the economic-desegregation effort is moving forward after several heated school-committee meetings with big public turnout.
The concept is not new. In the 19th century, Horace Mann advocated "common schools" that would bring students of all classes together in a kind of training ground for democracy.
In the 1960s, racial-integration efforts showed clear links between students' success and the income level of their classmates. When black children were integrated into schools with mostly low-income whites, their scores showed little improvement. When they were integrated into schools with middle-class whites, they performed much better.
Experts say these differences have to do with a host of factors involving parents, teachers, and students themselves. Schools with more middle-class students tend to have greater parental involvement (middle-class parents are four times as likely to join parent-teacher associations), and less teacher turnover. And middle-class students more often expect to attend college - an attitude that can rub off on peers.
"All that has a big impact on achievement in a school," says Mr. Kahlenberg.
The issue is also gaining increasing attention as more and more racial desegregation plans are being successfully challenged in court. Since 1991, three US Supreme Court decisions have released school districts from court-ordered desegregation measures, such as busing. And lower courts have ruled that districts cannot use racial quotas for admission to certain schools.
Such legal challenges "have pushed the country toward abandonment of a lot of [racial] desegregation efforts, even where local jurisdictions want to do them," says Gary Orfield, director of Harvard University's Project on School Desegregation. "So they're trying to figure out what to do."
The first community to try economic desegregation was La Crosse, Wis., in the late 1980s. The plan was initially controversial - several school board members were ousted over it - but over time, it has proved successful. Since then, a number of other districts, including Wake County, N.C, Montgomery County, Md., and Manchester, Conn., have launched similar efforts.
The Cambridge plan, which will be phased in gradually, will ultimately ensure that each school's percentage of low-income students (defined as those who qualify for a free or reduced-cost lunch) roughly equals the percentage in the district overall.
Currently, that's far from the case. Although the district is very diverse racially, the percentage of low-income students ranges from 20 percent to 80 percent in individual schools. The district average is just under 50 percent.
Although most parents in this famously liberal community (many of whom are faculty members at Harvard or MIT) say they prize diversity in the schools, some admit the new plan is providing a difficult test of principles.
"Cambridge tries very hard to have as much equity ... as possible, and I think most individuals here, including myself, view that as a good thing," says Steve Atlas, a Cambridge doctor whose daughter starts kindergarten next year. "The question is how much I'm willing to experiment with my daughter's education."
In general, experts say schools have a bigger impact on low-income children than on middle-class kids, for whom family is a stronger educational influence. So putting poor students into a largely middle-class school usually benefits them without hurting the rest.
But "if you sprinkle a few middle-class kids in schools where there are overwhelming concentrations of poverty, they're not going to change the culture of the school, and they're likely to do worse," admits Mr. Kahlenberg. "Most studies seem to find a negative effect [in schools with] above 50 percent low-income."
The Cambridge school committee is trying to reassure parents by requiring that schools improve their instruction as well as their equality. If such safeguards don't work, the risk is that well-off parents may flee the system.