concurring:
This is an unusual case in which the role of the district court as trier of fact seems to me so key that it would be extraordinarily difficult for an appellate court to reach a result contrary to the one reached below. Cf. Munoz-Mendoza v. Pierce, 711 F.2d 421 (1st Cir.1983). And the case presents a problem where the task of the trier of fact is in effect to evaluate the significance of policies. The question is: what is the impact, if any, of various administrative and business practices on a complex social phenomenon like resegregation; are the practices “symptoms” or “causes”? Although it is certainly possible on various points to disagree with the district court or with the majority, it would be difficult, given a problem of this sort, to conclude that the district court is so wrong as to merit reversal. Cf. Hope, Inc. v. County of DuPage, 738 F.2d 797, 817 (7th Cir.1984) (Cudahy, J., dissenting).
Thus, what conclusion can one draw about John Doe appraisals, a practice that logically would “tend” to hasten house sales, but which, in the view of the majority, was not shown to have caused or accelerated resegregation in Area A? Supra at 1426-27. It seems to me irrelevant that these appraisals are no longer used in Area B, id. at 1425 n. 7, since, of course, the Veterans Administration can reinstitute them at will. Because John Doe appraisals “tend” to speed up sales, the reimplementation of this practice in Area B might contribute to the future resegregation of that area. However, tracing the threat of re-segregation to the potential use of John Doe appraisals in Area B may be too speculative a connection on which to base standing, especially given the apparent dearth of data on the effect such appraisals had in Area A.
With respect to uncounseled VA loans, the district court discounted the testimony of a white resident that the existence of the FHA and the VA loan assistance programs fuels whites’ fears of resegregation. To the extent such fears do exist, the district court concluded they were “mistaken” because the evidence before the court did not establish a causal connection between VA-guaranteed financing and white flight. Jorman v. Veterans Admin., 654 F.Supp. 748, 766 (N.D.Ill.1986). It is not clear to me why the rationality of the fears is relevant. A connection may be created simply through perceptions: if residents believe rapid racial change and VA-guaranteed loans are linked, these perceptions by themselves produce effects in the real world. But, of course, the district court did not find that even this connection existed. At the end of trial, the court remained “unconvinced that the availability of VA-guaranteed financing was a substantial factor in the creation of white residents’ fears in Area A.” Id.
Although I am not persuaded that the programs the plaintiffs seek are without *1429value as antidotes to resegregation, I believe that, given the findings of the district court and the record developed below, we would not be justified in reaching a result different from that of the district court.1
. I think that one of the plaintiffs’ more formidable hurdles — but only as to Area A — is the apparent inadequacy of the record on remedies as they relate to redressability. The majority, of course, notes but does not rely on this point. Supra at 1423.