specially concurring:
I concur but write separately to explain my different perspective regarding white flight in this case and to make the point that this case is, to my mind, different from Lawrence County, Mississippi.1 The Hattiesburg School District has never integrated its elementary schools. Over thirty years after Brown v. Board, a newly proposed magnet plan with questionable prospects, and not testable for several years, is a weak answer to a suit on behalf of the district’s black children. Certainly not when so much can be had by some draw upon the proposed pairing and clustering plan, a draw that could offer few unmanageable problems, particularly if accompanied by a devotion to improving the quality *393of education as fervent as the devotion to the proposed magnet program.
I
At the outset, there is a caveat about which there should be no mistake. The panel is unanimous in its commitment to the imperative of Brown v. Board. Our disagreements may loom large, and indeed those disagreements at times are substantial, but it would be a mistake to read division over implementation and means as division over principle and ends; here principle and ends can be collapsed into quality education without regard to race. And the fear that both cannot be achieved fuels differences over means. The testimony of the experts and the opinion of the district court reflect a brooding sense that an already suffering quality in education will be further diminished by fully mixed classes. Long perpetuated differences in the quality of their education bring black children and white children to the elementary school desks with marked differences in levels of preparation, and critically, with resulting marked differences in the level and pace of class room instruction that teachers may pursue. It seems likely that without considerable effort by educators, class instruction will be pegged at the lowered median that inevitably seems to attend a full mixing.
The attractiveness of magnet schools is, in part, their offer of a possible escape from this loss of quality in education. The offer of quality is seen as a carrot to tempt a voluntary mix, a mix with narrowed black and white levels of preparation. The allure is strong. Magnet programs as a remedy for segregated schools speak to voluntary acts and quality education, a freedom of choice with incentives to make desegregated schools a reality, achievable without the disasters of Boston. Perhaps more fundamentally, magnet programs appeal to our ethic of self determination, to a marked sense of merit, both for individual students and for the governance of school districts. This is not a criticism but a caution, because important questions lurk behind the allure of magnet plans, such as whether a particular magnet reduces disparity in the education levels of mixed blacks and whites by taking the best of the black students and turning its back on, or at least showing less concern for, the fate of the greater number of “average” black students, unable to qualify or uninterested in trying, a deficit in motivation not easily separated from the deficit in education.
Such questions are best answered by persons with training not possessed by judges, true enough. Yet, the command to eliminate segregated schools, root and branch, is of constitutional order, and ultimately must find expression in a judicial decree. That is, courts must answer actual questions, not abstract inquiries, and do so by deciding whether a particular magnet plan implements the constitutional order or whether a particular magnet plan dilutes the constitutional command.
In short, mindful that remedies must be tailored to the wrong, magnet plans, as all plans, must be individually crafted to respond to unique needs of each district in order to implement the command to eliminate segregated schools, root and branch. Just as a mindless pairing and clustering that in a quest for an elusive percentage of mix transports black children across a district to attend a school, only to find that it is now also black, fails its implementing role, so also can a mindless subscription to magnet plans fail by leaving great numbers of black children to attend black schools; plans that can be little more than freedom of choice with the twist that the choice is exercisable only by a few.
Communities and school districts with bridges over racial separation are fragile and the task of change is delicate. As I will explain, I read Judge Wisdom’s opinion as responding with sensitivity; it does not reject the work of the district court, the parties and the community by a blind insistence upon the pairing and clustering of the Stolee plan, without regard to its per*394verse potential for greater segregation. Rather, the opinion urges further effort to reduce the numbers of children attending racially identifiable schools, an effort that draws upon both the Stolee and the magnet plans.
II
There are three proposed clusters 2 under the Stolee plan. Even were the pairing plan adopted in full, and the panel does not even suggest that it need be, pairing the Grace Christian, Jones, and Davis schools would require no travel in excess of 2.4 miles or fifteen minutes; Woodley to Walt-hall is a 2.5 mile trip of fifteen minutes. Travel between Grace Love, and Thames is a trip of 6.4 miles or eighteen minutes, the longest in the district; Thames and Bethune are 3.9 miles or fifteen minutes apart. Even the magnet plan contemplates the transport of students who reside more than one mile from their school and make a majority to minority transfer.
The district court was properly concerned by a possible loss of neighborhood schools for these children of tender years and its companion, white flight. In the abstract, these concerns are understandable and cannot be ignored. But the opportunity to do more is so plain and the travel distances are so short that the plea for neighborhood schools rings hollow.
Much of the debate has here centered on the role of “white flight” and its relevance to judicial efforts to desegregate public schools. Judge Wisdom confines its relevance to a choice between constitutional plans, ignoring its potential in deciding what is required by the Constitution. At this level of generality, I agree. But that does not take us far because Hattiesburg is unquestionably operating segregated elementary schools and that operation is unconstitutional. The specter of white parents removing their children from public schools plays no part in that judgment.
The difficulty comes from our task of reviewing the district court’s decision that a magnet plan is the preferred remedy for that unconstitutional condition because parents may otherwise remove their children from the district. The anticipated response of white parents cannot be ignored; at the same time fear of white flight cannot alter the constitutional command. The effectiveness of a remedy is the question and it makes no sense to construct decrees that do not grasp the real world.
The exquisite difficulty is that a decree contemplating defeat by white flight is a self-fulfilling prophecy. The experts who testified were impressive, including their opinions about anticipated reactions to the proposed plans, extrapolated from Los Angeles, California, Louisiana and Mississippi. Nonetheless, though they may be the best from the data, the studies, remain guesses if educated ones of how people may respond to various plans. Over a period of time these scholars see patterns in the responses to various plans. Apart from the fact that as yet, the data is not so rich that the perceived patterns ought to be regarded as anything more, these social scientists do not purport to gauge why people reacted as they did. So whether the responses are sheer racism, concern for educational quality, local schools, or all, remain anyone’s “judgment.” I welcome the confirming opinion of all the experts that local schools and quality education are essential ingredients — I venture the primary concern of parents, black and white. This concern can be met. Some draw upon the pairing and cluster plan proposed here in conjunction with a magnet program might achieve desegregation, and now. It or some combination of the magnet and pairing plans could do so with little loss of local or neighborhood schools, and minimal transportation. Even taken full measure, nothing in *395the pairing plan rejects the possibility for improving the quality of education in addition to desegregation.
In these cases, we tend to be less tied to the record, perhaps because our task is such an unusual judicial chore, a phenomenon I have criticized. United States v. Lawrence County School Dist., 799 F.2d 1031, 1052 (5th Cir.1986) (Higginbotham, J., concurring in part and dissenting in part). Whatever our freedom to call upon “judicial experience,” that experience must lead us to disagree with the assumption, implicit in the arguments, that white citizens of Mississippi will necessarily pull away from a plan for the sole reason that it causes their children to go to school with blacks. We know from other cases, such as Lawrence County, that today this is not necessarily so. We also know that while there was a loss of white students, the use of pairing and zoning in Hattiesburg’s secondary schools resulted in a system with no racially identifiable schools; surely we also know that parents in Mississippi and elsewhere, will protest with their feet dilution in the quality of education and transports of small children long distances from their homes. Particularly when, as the government’s impressive expert put it — there is nothing at the end of the ride. These concerns are understandable, not illegal, not ignoble, and they must be considered. Because they need not be realized here, I join the panel in its remand, a remand that contains more praise than criticism for the district court and all parties’ efforts below.
. It bears emphasis that we do not reject Dr. Rossell’s use of "interracial exposure” as a diagnostic aid. To the contrary, this inquiry can cast light on the potential success of a proposed plan, even its objectives.