Suppose that for a while, the president and Congress have been engaging in a constitutional practice that is not authorized by the text (and was not authorized at the founding) — maybe it’s allowing the president to make recess appointments to offices that became vacant while the Senate was in session. There is one sense in which an observer might say that the practice represents a constitutional “change.” It used to be that such appointments were thought to be invalid. Now they are thought to be valid. [Let’s put aside for now whether this hypothetical is accurate.]

But there is another, important, sense in which the Constitution has not changed. After all, the text of the Recess Appointments Clause is still right there, (and that text is dated 1787, if you care).

Now suppose that there is extremely widespread belief in America that the text of the Constitution (in its original sense) is the law, and that — allowing for the fact that it is often ambiguous or vague, and allowing for longstanding rules like precedent — it controls over contrary practice. Maybe that widespread acceptance includes the president and Congress, and so even they would give up the practice if they really thought the Constitution forbade it. Maybe not — maybe it merely includes the people, and by happy chance, some members of the federal courts. [Either way, put aside for now whether this hypothetical is accurate, too.]

In my view, under these hypotheses, the best way to describe what is happening is to say that there has been a “constitutional change,” but an invalid one. That is, one of the constitutional clauses is now being implemented much differently than it used to be. But that practice conflicts with the higher law of the United States.

If you want to be Austinian about it, you can say that the Constitution represents a command from the ultimate authority, the People of 1787, and that while the Peoples of 1791, 1795, 1804, 1865, 1868, 1870, 1913, 1919, 1920, 1933, 1951, 1961, 1964, 1967, 1971, and 1992* have all modified those commands, none of them have modified the relevant command of the Recess Appointments Clause. If you prefer to be a modern positivist about it, you can say that the current social fact is that we treat the text as authoritative. Either way, since the officials promised to support the Constitution, and hold office only because of the Constitution, that means they ought to quit violating it.

Not everybody agrees with my hypothesis about the role of the text in current American constitutional law. (Think of Solicitor General Verrilli, asked what to do “when there is a practice that flatly contradicts a clear text of the Constitution,” who said that “the practice has to prevail,” though he mostly fought the hypo). But that is what I was trying to get at in my previous post, and that is one of the many issues that divides me from Eric Posner (here is his post).

Eric concludes that I have made:

a large concession that is hard to square with originalism. Even if he thinks that no such evolution has occurred, he seems constrained to acknowledge that if it can occur, justices and other government agents should be on the lookout for changes in the people’s constitutional understandings, and should act accordingly if such changes occur.

I think originalism can fairly be described as the empirical belief that we in America have not decided to overthrow the original Constitution, even if we amend it sometimes, even if the bosses violate it sometimes. Yes, I acknowledge that government agents should be on the look out for such an overthrow — whether complete or partial. But remember the agent should also continue to ask, “Who gave me the authority to implement this new rule?” In any event, I am pretty sure the bigger risk is that officials will try to perpetrate such an overthrow, not that they will fail to notice one that has actually occurred.