Darrell A. H. Miller (The University of Chicago Law School; Duke University School of Law) has posted ORIGINALISM'S SELECTION PROBLEM on SSRN. Here is the abstract:
Originalism is called a “family” of theories. If that’s true, they all suffer from a congenital malady: the problem of selection. This vulnerability afflicts every branch of the family, whether we speak of original intentions originalism, original public meaning originalism, original methods originalism, or original law originalism. To the extent originalism of any variety purports to make descriptive, falsifiable claims about language, history, tradition, practice or law, all of them must confront problems common to any empirical project—what data to collect, how to code that data, over what time period and how long to collect data, and what conclusions to draw from that data.
This essay examines the problem of selection with any originalist theory and what it means for constitutional adjudication. Part I surveys the major branches of originalism, explains how they each share a commitment to the pursuit of objective, knowable, falsifiable fact, and then use that fact as a fixed point to constrain discretion. Part II discusses how each of these theories are vulnerable to methodological challenges familiar to all empirical projects. Part III explores the implications of these methodological challenges for originalism in particular and for constitutional doctrine and theory more generally.
I agree with much of this. There may well be some issues where methodological rigor is especially important for determining the original public meaning of the constitutional text. From my perspective, the "counting" metaphor is a bit misleading. The fundamental enterprise for public meaning originalists is recovery of the communicative content of the constitutional text. There may be cases where quantitative methods are very helpful: for example, the work I did with Thomas R. Lee, James Phillips, and Jesse Egbert or the original public meaning of "incomes" in the Sixteenth Amendment applied corpus linguistics methods. In other cases, gleaning the original meaning is relatively straightforward, because of linguistic continuity. Determining original meaning is always about inference to the best explanation: what hypothesis about the original meaning of a term or sentence provides the best explanation for all the evidence that is available? Sometimes we will have a very high degree of confidence in our judgments about original meaning. Other times and more rarely, there will be significant doubt. If I have a bone to pick with Miller's argument, it is that he seems to assume that originalist methodology must guarantee accuracy and eliminate uncertainty, but no serious originalist of whom I am aware has made the claim that originalist judging is or must be perfect.
