Over the weekend, L.A. Times Supreme Court reporter David Savage criticized John Roberts for breaking his pledge to be a minimalist judge, pointing to the Chief’s votes with the “conservative” majority in Gonzales v. Oregon and Rapanos v. United States as evidence of his hypocrisy. Ilya Somin and Orin Kerr critique Savage here and here. Ilya, in particular, makes a good point:

Roberts[’] position in Rapanos was that the Corps has very broad discretion under the CWA’s grant of power to regulate discharges into “navigable waters,” but not the power to assert what he rightly called virtually “boundless” regulatory authority. The Corps had claimed the right to regulate virtually any body of water, no matter how small, remote, or nonnavigable, and had persisted in this stance in the face of an adverse Supreme Court decision. No broad assertions of judicial power here.

[H]is positions were hardly contrary to “judicial modesty.” Indeed, in my view, a stronger criticism of Roberts’ performance is that he was too deferential to assertions of federal authority in both cases, particularly Oregon.

See my recent short piece, and earlier blog post, on Roberts’ Rapanos concurrence here for more in that vein.

The confusion points out that “judicial minimalism” can mean different things. Here are just two (of several) possible meanings:

1. Minimalism could mean protecting expectations created by current precedent, or current understanding of precedent, by reading new, potentially revolutionary decisions, like Raich, narrowly rather than broadly. In fact, Cato’s amicus brief in Gonzales v. Oregon tried to portray a vote for Oregon as a minimalist decision in this sense, arguing that Raich should be read as narrowly as possible in order to protect the pre-Raich expectation of states.
2. Minimalism could be read to mean that courts should displace democratic politics, rather than floating “expectation interests,” as little as possible, by deciding “no more than is necessary,” leaving as much of a class of questions implicated by the case as possible to democratic judgment. This is much closer to what Roberts appears to believe, and his votes in both Oregon and Rapanos are consistent with this version—carving out as much room for a (federal) political response to the problem as possible.

Notably, though, even if you accepted version two, you could reach a pro-state outcome. Again, Cato’s amicus brief tried to make such a connection by portraying a narrow reading of Raich in that case as a reading favoring the decisionmakers with the superior democratic pedigree—that is, Oregon voters who, by direct referendum, had voted for legalizing assisted dying (twice). Even if agencies have a greater democratic pedigree than courts, state referenda have a far better democratic pedigree than the rulings of federal agencies.

The upshot is that minimalism is an indeterminate concept, less a theory of judging than a mood or intuition whose outcome depends on unexpressed background preferences. Oregon suggests that Roberts’ operative background preference is to preserve the flexibility of agencies under executive control by weighting their interests more than competing state interests when the law is open-textured. Perhaps, this might be called a kind of pro-executive minimalism.