« More from Renee Lerner on the Second and Seventh Amendments
Michael Ramsey
| Main | Donald Drakeman on Garrett Snedeker on Donald Drakeman on Originalism
Michael Ramsey »

10/24/2021

Cass Sunstein: Analogical Reasoning
Michael Ramsey

Cass R. Sunstein (Harvard Law School) has posted Analogical Reasoning (64 pages) on SSRN.  Here is the abstract: 

In law, the process of analogical reasoning appears to work in five simple steps. (1) Some fact pattern A—the “source” case—has certain characteristics; call them x, y, and z. (2) Fact pattern B—the “target” case—has characteristics x, y, and q, or characteristics x, y, z, and q. (3) A is treated a certain way in law. (4) Some principle or rule, announced, created, or discovered in the process of thinking through A, B, and their interrelations, explains why A is treated the way that it is. (5) Because of what it shares in common with A, B should be treated the same way. It is covered by the same principle. It should be clear that the crucial step, and the most difficult, is (4). Often analogical reasoning works through the use of incompletely theorized agreements, making (4) tractable. Some of the disputes about analogical reasoning reflect contests between Burkean and Benthamite conceptions of law.

(Via Larry Solum at Legal Theory Blog, who says "Highly recommended.  Download it while it's hot!")

I'm skeptical that (most) analogical reasoning is anything more than moral/policy analysis.  Not that there's anything wrong with that -- but in most cases it cannot be elevated to a logical imperative that provides a distinctly legal solution (as opposed to a moral/policy solution).  In Professor Sunstein's step (4), how is a "principle or rule, announced, created, or discovered"?  And why that "principle or rule" and not a broader or narrower one?  True, sometimes the "A" and the "B" are so close -- distinguished by differences that all would agree are immaterial -- that the treatment of B does follow as a logical matter from the treatment of A.  But usually not, or at least not necessarily.