By Mike Dorf
My Verdict column for this week is part 2 of my 2-parter on the 40th anniversary of Roe v. Wade. (Part 1 and the accompanying blog post appeared on Thursday of last week.) In Part 2, I consider the claim that Roe went too far, too fast--that a less sweeping decision merely invalidating the Texas law at issue in Roe would have invited a dialogue between the Court and the People, which in turn would have resulted in less contentiousness and roughly the same degree of liberalization of abortion law that Roe decreed. I argue that these claims are highly speculative at best and probably false.
Most of my argument against the too-far-too-fast hypothesis rests on two factual observations: (1) Its proponents overstate the degree to which state legislatures were in the process of liberalizing abortion law when Roe constitutionalized the subject; and (2) in the two decades since Planned Parenthood v. Casey, we have seen that legislatures given greater regulatory freedom regarding abortion do not typically respond with "dialogue." Rather, pro-life legislators (quite understandably and predictably) enact laws that make abortion as difficult as possible, going right up to the line of what the courts will uphold or even well over it, on the theory that simply making people go to court to secure abortions is costly.
But in addition to these concrete grounds for objecting to the too-far-too-fast hypothesis, I also raise a general epistemological point: Insofar as the too-far-too-fast hypothesis rests on a claim about what would have happened if we change one piece of history, it is inherently speculative. Anybody who has ever read or seen a time travel book or movie knows that even small changes can set off large and unpredictable chains of causation. I vividly remember as an undergraduate when a fellow student asked a question of one of my favorite teachers, the late political theorist Judith Shklar. The question began "what if . . . ." I don't now recall what followed the "what if" but it was something like "what if the first communist revolution had occurred in a highly industrialized country like Germany or England rather than Russia?" Shklar's answer has stuck with me pretty close to verbatim: "What is this 'what if'? There are no 'what ifs' in history. The past is determined."
The title of this post refers to a very old (like 1978 old) Saturday Night Live sketch that posed various preoposterous what-if questions, including "what if Eleanor Roosevelt could fly?". Shklar's point was that all such questions--including seemingly serious ones--are preposterous, at least from the perspective of history.
But I never understood Shklar to be making the point that "what if" questions lack all value. I understood her to be making a point about the study of the past. Shklar herself was a political scientist, not a historian, and presumably a good deal of what political science does is to make generalizations that necessarily entail some set of views about "what if" questions. Suppose you agree with Immanuel Kant, Thomas Paine and their modern supporters in thinking that democracies never (or at least rarely) go to war against one another. If so, then you will think you can say something substantive in response to a question like "would England and Argentina have fought a war over the Falklands if Argentina had not been ruled by a military junta at the time?"
Of course, anything you say will be probablistic and will lose force over time, because the path-dependence of the "butterfly effect" is real. But the whole point of a science of politics is that, at least within broad bounds, history is not simply one damn thing after another. There are patterns and causal mechanisms, albeit complicated ones. In retrospect, I realize now that Shklar was a favorite teacher of mine despite her dismissal of "what if" questions, not because of it.
To come back to the too-far-too-fast hypothesis, I want to emphasize that when I say in the column that it is highly speculative, I do not just mean that in the sense that all claims about what might happen in a counterfactual world are necessarily speculative. I also mean that we have specific reasons to doubt the particular counterfactual narrative that the too-far-too-fast hypothesis assumes.