Because abortion involves the interests not only of the mother but also of the unborn child (and the government’s interest in both), the court’s abortion jurisprudence has always been a mess of contrived categories and multifactor tests, resembling legislative logrolling more than judicial reasoning. Griswold, by contrast, simply created a categorical right to contraceptives for married couples, extended to the unmarried in Eisenstadt v. Baird (1972). Even if these cases were wrongly decided, they are far more workable than Roe.

Such stare decisis considerations are ultimately prudential judgments. But that is precisely the point: Taking precedent seriously in our constitutional system means taking each precedent’s specific details and circumstances seriously. Like all prudential judgments, a judge’s analysis of stare decisis factors will surely spur debate and criticism. And the judge may exercise his judgment poorly—but so may the critics, especially when they attempt to reduce the doctrine of precedent itself, one of the most nuanced parts of constitutional self-government, to slogans.

Abandoning Roe would improve this part of our constitutional discourse. Precedents in other areas of law will be evaluated and debated in much more nuanced and open-minded terms without lawyers, judges and citizens looking over their shoulders at what Sen. Arlen Specter once called the “super precedent” of Roe. Cases would be taken more seriously on their own terms, and not as a proxy war for a court-made right to abortion.