I appreciate Robert's post, but feel I must respond on several counts. He makes the point that all social science rests on some normative judgments. For example, it is a normative judgment that declarations of unconstitutionality represent one possible indicator of activism. In that sense, in a study of activism, the decision to examine votes in cases challenging the constitutionality of legislation constitutes a form of "value-laden observation." No question there. But most social scientists would shy away from the determination of whether a statute was constitutional or unconstitutional a priori
, because that is an interpretive question that social science does not address. Instead, influenced by the behavioral revolution, social scientists generally study patterns in behavior (in this case, the propensity to strike statutes), and seek to explain and predict that behavior. That is to say, there are no social scientific methods that enable a researcher to create a baseline regarding whether a statute is or is not constitutional in some absolute or universal sense. Whether a statute is constitutional or unconstitutional varies with time and interpreters. While at some point in time most Americans would have agreed that segregated schools were constitutional, now few would adhere to that view. Determinations of constitutionality are dependent on a series of variables that are not themselves time or context-independent. To borrow a phrase from Richard Rorty and popularized by Stanley Fish, constitutionality depends on the nature of the "interpretative community" and whether consensus can be reached within that community by whatever decision rule prevails as to the constitutionality of a statute. At the U.S. Supreme Court, given its discretionary jurisdiction, often there is conflict within the interpretative community regarding just that question.
On that other hand, there may be valid indicators reflecting whether, at the time the decision was rendered, more or less consensus exists
regarding a statute's constitutionality. A unanimous decision striking down a statute--especially one that affirms the lower court--might be one way to determine whether such a consensus exists. In that situation, a vote might be perceived as "less activist" than one in which the Court was divided 5-4 (this dynamic is explored by Lori in a recent article and is currently being incorporated into a measure of activism by Frank Cross and me). Or, one might label as less activist a vote to strike a statute that creates a substantive result in conflict with a justice's attitudinal predisposition (assuming such an attitudinal predisposition can be validly measured, which I believe it can). But such an approach is transparent and objective--it involves no subjective judgments about whether a statute is truly
constitutional or not. In that sense, it can be critiqued and replicated by later social scientists.
In short, Robert begins with the right premise but goes too far, in my opinion, regarding his baseline normative standard for social science.