Thomas, J., dissenting
conception of “public use” adopted by this Court in Berman v. Parker, 348 U. S. 26 (1954), and Hawaii Housing Authority v. Midkiff, 467 U. S. 229 (1984), cases that take center stage in the Court’s opinion. See ante, at 480–482. The weakness of those two lines of cases, and consequently Berman and Midkiff, fatally undermines the doctrinal foundations of the Court’s decision. Today’s questionable application of these cases is further proof that the “public purpose” standard is not susceptible of principled application. This Court’s reliance by rote on this standard is ill advised and should be reconsidered.
A
As the Court notes, the “public purpose” interpretation of the Public Use Clause stems from Fallbrook Irrigation Dist. v. Bradley, 164 U. S. 112, 161–162 (1896). Ante, at 479–480. The issue in Bradley was whether a condemnation for purposes of constructing an irrigation ditch was for a public use. 164 U. S., at 161. This was a public use, Justice Peckham declared for the Court, because “[t]o irrigate and thus to bring into possible cultivation these large masses of otherwise worthless lands would seem to be a public purpose and a matter of public interest, not confined to landowners, or even to any one section of the State.” Ibid. That broad statement was dictum, for the law under review also provided that “[a]ll landowners in the district have the right to a proportionate share of the water. ” Id., 164 U. S., at 162. Thus, the “public” did have the right to use the irrigation ditch because all similarly situated members of the public—those who owned lands irrigated by the ditch—had a right to use it. The Court cited no authority for its dictum, and did not discuss either the Public Use Clause’s original meaning or the numerous authorities that had adopted the “actual use” test (though it at least acknowledged the conflict of authority in state courts, see id., 164 U. S., at 158; supra, at 513–514, and n. 2). Instead, the Court reasoned that “[t]he use must be regarded as a public use, or else it would seem to follow that no gen-
