Both Howard Bashman and Orin Kerr ask what happens to a statute that is declared unconstitutional if and when the Supreme Court later reverses itself on the same issue. This is a very relevant question regarding the future of abortion bans given the likelihood of more anti-Roe v. Wade Justices in the near future.
First principles could be invoked for either side.
Ideas of (small-r) republicanism and popular sovereignty suggest that the statute must be re-passed by the legislature before it can take effect. The court's first decision striking down the law short-circuited any political activity that might have been directed towards repeal — why bother repealing a dead letter?
On the other hand, a formalist view — and we don't in our law recognize desuetude, so we have a somewhat formalist tendency — would suggest that every statute on the books is in force except those blocked by court action. Remove the block, the statute springs back into action. In other words, the second decision erases the first as if it never was (and then only due process prevents charges filed for any violations of the rule during the period it was thought invalid?).
Orin asks if there are any cases on this, and indeed there is one, although somewhat aged, leading case, that of Jawish v. Morlet, 86 A.2d 96 (D.C. App. 1952), heard by what was then the Municipal Court of Appeals for the District of Columbia.
I've reproduced the full text below, but the thrust of it is captured in this sentence: “And since the [first supreme court case] never was the law, its only effect, to use the language of Justice Vinson in the Warring case, was 'that just about everybody was fooled.'” As a result, the old law springs back into bindingness.
As a decision by the D.C. local courts is not binding precedent anywhere else, the issue appears open at the federal level. It seems to me that this sort of thing is actually arguable either way but that the length of time between the two decisions is highly relevant. The longer the gap, the stronger the argument for not reinstating the decisions by fiat, for the stronger the argument that but for the first court decision coalitions would have formed to demand repeal.
UPDATE: Stuart Buck weighs in with more formalist arguments as to why Jawish provides the correct answer:
One thing to clear up right off the bat, however, is the common misconception that federal courts have the power to “strike” down a law. We usually imagine that a statute, once declared unconstitutional, “is not a law; it confers no rights; it imposes no duties; it affords no protection; it creates no office; it is, in legal contemplation, as inoperative as thought it had never been passed.”191 But this is not technically accurate. Despite such language used by courts and commentators, there is no such thing as “striking down.” A federal court has no power to erase a statute from a state's lawbooks.192 As one prominent scholar said, “No matter what language is used in a judicial opinion, a federal court cannot repeal a duly enacted statute of any legislative authority.”
To which I can can only say, “well, maybe.” The argument that a court doesn't void a statute, just puts it into a species of legal hibernation is strengthened by the observation that legislatures sometimes repeal statutes held to be unconstitutional, and if the statute were a total nullity, that would be a useless act. But then again, legislatures do useless acts sometimes, so what exactly does that prove, especially since there's no one with standing to challenge the repeal.
It seems to me that on this one, like many hard constitutional questions, how you come out on this one depends on what you consider relevant inputs to the issue. Is it the nature of democracy? A structural view of separation of powers? The underlying goal(s) of the Constitution (whatever you think that is, e.g. liberty)?
Update 2: See also Marty Lederman's excellent comments (broadly agreeing with Stuart Buck), and the interesting remarks of The Greedy Clerk (arguing that for state statutes the answer depends on state rules of decision).