The Wall Street Journal editorial board is laundering the Roberts Court’s documented pattern of one-sided doctrinal destruction as popular legitimacy. The Marquette Law School Poll has published another wave of public-opinion numbers on the Supreme Court, and the editorial board has treated the institutional confidence rating as if it were a constitutional defense of the rulings.
The steel-man of the editorial’s argument runs as follows. The Roberts Court has issued a series of historic rulings, and Americans, when polled, continue to express confidence in the institution. Therefore the institution’s work is legitimate, and the critics are a vocal minority animated by partisan grievance.
The editorial identifies several of those rulings by name. Dobbs v. Jackson Women’s Health Organization, 597 U.S. 215 (2022), overruled Roe v. Wade and Planned Parenthood v. Casey. Loper Bright Enterprises v. Raimondo, 144 S. Ct. 2244, 2273 (2024), overruled Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc. and dismantled the administrative-law deference regime the federal courts had operated under for forty years. Shelby County v. Holder, 570 U.S. 529 (2013), struck down the coverage formula of the Voting Rights Act and left the statute’s text on the books with no operative enforcement mechanism. Rucho v. Common Cause, 588 U.S. 684 (2019), declared partisan gerrymandering claims nonjusticiable and closed the federal courthouse door to a constitutional violation the Court itself acknowledged.
Each was decided by a five-justice majority along predictable coalition lines. Each departed from a methodological baseline the Court had previously claimed to honor. The argument has the virtue of being short. It has the defect of being wrong.
Start with what the poll actually measures. The Marquette Law School Poll asks respondents whether the Supreme Court is doing a good, fair, or poor job — an institutional rating. The Journal reads the residual confidence number — the “fair or good” share — as a defense of the Court’s specific rulings. The poll does not work that way. The same Marquette series the editorial cites has, in its recent waves, found that majorities of respondents believe the Court is driven by ideology rather than law on the cases the editorial does not name. The Journal cites the institutional number and treats it as if it answered a question the polling instrument was not designed to ask.
Now consider the audit. The Roberts Court has produced, across the last three terms, a body of work that systematically favors one political coalition over another on questions that touch the daily life of every American. The institutional confidence numbers the editorial cites measure the residue of prestige from an earlier era, when the Court produced something closer to bipartisan output. They do not measure — and cannot measure — consent to the doctrinal program the Roberts majority has actually executed. Loper Bright’s majority opinion, at 2273, claimed to restore judicial humility to agency review — the same humility the same majority had invoked in Kisor v. Wilkie five years earlier, and then discarded for reasons that never met the methodological standard the author had previously set.
The poll will move. By the next major Roberts Court ruling that lands as concrete injury on a politically salient question — agency deregulation, federal civil-rights enforcement, the next executive-immunity invocation — the institutional confidence numbers will catch up to the rulings. The Journal will then write the same editorial it is writing now, citing the same Marquette series, and treating the polling of the moment as if it answered the question the editorial has refused to answer for a decade: whether the Court’s actual work deserves the legitimacy the editorial board has been spending on its behalf.