« ELS Nuts and Bolts | Main | Is Religious Liberty Vulnerable, or is Religious Liberty Talk Too Insulated? »

Wednesday, July 11, 2012

The "If ___ means anything, it surely means ____" argument

We've all encountered this argument, I imagine:  "If ___ means anything, it surely means _____."  See, e.g., Central Hudson Gas & Elec. Corp. v. Public Service Comm'n, 447 U.S. 557, 575 (1980) (Brennan, J., concurring) ("If the First Amendment guarantee means anything, it means that, absent clear and present danger, government has no power to restrict expression because of the effect its message is likely to have on the public").  This is more of a move than a "modality," I suppose, but it's a common one.  (It evokes, in a way, Phil Hartman's "unfrozen caveman lawyer" tactics.)  Is there a name for it?  I'm sure there's a technical fallacy involved in the move, so maybe it has a cool Latin name?

Why yes, I am avoiding something I'm supposed to be doing . . .

Posted by Rick Garnett on July 11, 2012 at 04:11 PM in Rick Garnett | Permalink

TrackBack

TrackBack URL for this entry:
http://www.typepad.com/services/trackback/6a00d8341c6a7953ef01676866cc17970b

Listed below are links to weblogs that reference The "If ___ means anything, it surely means ____" argument:

Comments

I don't have time to look it up in the book right now, but check out Anthony Weston's, A Rulebook for Arguments. It has a great section on common fallacies and usually has the Latin name for them.
http://www.amazon.com/Rulebook-Arguments-Anthony-Weston/dp/0872209555

Posted by: James | Jul 11, 2012 4:21:09 PM

It´s a kind of petitio principii ("begging the question") since they are assuming a kind of indisputable core meaning for the concept.

Posted by: Gust Arballo | Jul 11, 2012 4:50:42 PM

I would think the form of the argument is fair if the example is in fact the paradigmatic example of what the law does. The trouble with the form of the argument is its misuse: Often people pair it to something contested, not at the core, often by playing wit the level of generality.

Posted by: Orin Kerr | Jul 11, 2012 5:50:22 PM

If X means anything, then it must mean Y.

In other words, if it does not mean Y, then X means nothing. But this is absurd. (i.e., e.g., absurd to believe that first amendment means nothing)

Therefore, X must mean Y.

I don't see anything wrong with the form. And I agree with Orin.

Posted by: Jae Lee | Jul 12, 2012 12:45:28 AM

The form of the argument may be valid but it does, as noted by Gustavo above, assume the question or issue at hand, so in fact it does beg the question. It may, or may not, be true, the case, etc.

Posted by: Patrick S. O'Donnell | Jul 12, 2012 2:01:08 AM

This plays on (or exploits) our belief that there must be _some_ meaning to X (If X means anything, it must mean Y, that is, something) and thus Y more accurately represents some core, fundamental, or minimal meaning of X (this captures, I think, its logical form). But that core, fundamental, or minimal meaning _might_ or _could_ be A or even Z; and thus it could be (Y)A, or (Y)Z, or…. What is more, X need not be timeless, invariant, absolute, perhaps it is subject to qualification or conditions, semantic drift with history, and so on, in which case there remains, as it were, a core, fundamental, or minimal meaning of X, that is, Y, but precisely what Y in fact is varies with purposes, situations, contexts, or history…. In other words, qualifications or conditions may play a role in the final determination of meaning, a meaning therefore that not invariant, timeless, absolute, even it there exists Y throughout the course of variations in meaning.

Posted by: Patrick S. O'Donnell | Jul 12, 2012 2:32:36 AM

correction: In other words, qualifications or conditions may play a role in the final determination of meaning, a meaning therefore that is not invariant, timeless, absolute, even if there exists a Y throughout the course of variations in precise meaning.

Posted by: Patrick S. O'Donnell | Jul 12, 2012 2:34:20 AM

Post a comment