Academic

Vermeule on Many Minds Arguments

By Lawrence Solum

Adrian Vermeule (Harvard University - Harvard Law School) has posted Many-Minds Arguments in Legal Theory on SSRN.  Here is the abstract:

Many-minds arguments are flooding into legal theory. Such arguments claim that in some way or another, many heads are better than one; the genus includes many species, such as arguments about how legal and political institutions aggregate information, evolutionary analyses of those institutions, claims about the benefits of tradition as a source of law, and analyses of the virtues and vices of deliberation. This essay offers grounds for skepticism about many-minds arguments. I provide an intellectual zoology of such arguments and suggest that they are of low utility for legal theory. Four general and recurring problems with many-minds arguments are as follows:

(1) Whose minds?: The group or population whose minds are at issue is often equivocal or ill-defined.

(2) Many minds, worse minds: The quality of minds is not independent of their number; rather, number endogenously influences quality, often for the worse. More minds can be systematically worse than fewer because of selection effects, incentives for epistemic free-riding, and emotional and social influences.

(3) Epistemic bottlenecks: In the legal system, the epistemic benefits of many minds are often diluted or eliminated because the structure of institutions funnels decisions through an individual decisionmaker, or a small group of decisionmakers, who occupy a kind of epistemic bottleneck or chokepoint.

(4) Many minds vs. many minds: The insight that many heads can be better than one gets little purchase on the institutional comparisons that pervade legal theory, which are typically many-to-many comparisons rather than one-to-many.

And some more from the text:

Here a useful idea stems from the 19th-century legal theorist James Bradley Thayer, who argued that courts should defer to legislatures unless the statute is clearly unconstitutional or irrational.109 Thayer?s rather vague suggestion was that this regime would be beneficial because of legislators? superior ?judgment.? We can put analytic backbone into Thayer by reading him in epistemic terms, more particularly in Condorcetian terms.110 Under plausible conditions, legislative minds will be more likely than judicial minds either to identify preference-independent right answers, or to track what a fully accurate poll of public preferences would reveal. Structural features of modern legislatures give them powerful Condorcetian advantages, especially the sheer number of legislators; their powerful tools for acquiring information, which raises average competence; and the relative diversity of the legislators along dimensions of class, profession, and geography, which reduces the correlation of biases as compared to the more homogeneous judiciary.

To be sure, it may be that the legislators are motivated to achieve ideological aims or partisan advantage or some other political end, rather than to get the constitutional question right in some sense. But this shows why the possibility of a polling-model interpretation of Condorcet matters; in that interpretation, the aim is precisely to track public preferences through an accurate poll, in which case legislators? relative political accountability and superior information about public preferences is a good, not a bad; it makes them a more accurate sample group than the judges.

Once again, a supercool paper from Vermeule.  Highly recommended.

Full post as published by Legal Theory Blog on January 30, 2008 (boomark / email).