Law and 'The Argumentative Theory'

18 Pages Posted: 20 Jan 2012 Last revised: 19 Apr 2012

Date Written: December 2, 2011


Cognitive scientists Hugo Mercier and Dan Sperber recently introduced what they call "The Argumentative Theory." Scientists such as Steven Pinker and Jonah Lehrer have hailed it as an important breakthrough in understanding human behavior. Jonathan Haidt believes that the new theory has solved "one of the most important puzzles in psychology."

The puzzle is this: Why are human beings so good at reasoning in some situations and so hopelessly wrong in others? Mercier and Sperber provide an elegant answer: It is because the function of reasoning is not to arrive at the "right answer," but rather to find support for a conclusion the reasoner has already intuited. In a nutshell, reasoning is not intended to discover truth; rather, its role is to win arguments with other people.

Mercier and Sperber's empirically-supported argument reflects what theorists such as Jerome Frank, Richard Posner, and Stanley Fish have long contended about legal reasoning: that conclusions drive reasoning and not vice versa.

Yet ironically, recently-appointed Supreme Court justices such as Elena Kagan, Sonia Sotomayor, and John Roberts have adamantly contended - at least at their confirmation hearings - that simple application of law should drive legal decisions.

This Essay examines why, at the very time cognitive scientists are offering increasingly sophisticated analyses of reasoning, the views offered by these recent appointees to the Supreme Court should be so simplistic. It goes on to suggest what both judges and lawyers can learn from the Mercier-Sperber theory.

Suggested Citation

O'Neill, Timothy P., Law and 'The Argumentative Theory' (December 2, 2011). Oregon Law Review, Forthcoming, Available at SSRN:

Timothy P. O'Neill (Contact Author)

John Marshall Law School ( email )

315 South Plymouth Court
Chicago, IL 60604
United States

Do you have negative results from your research you’d like to share?

Paper statistics

Abstract Views
PlumX Metrics