One’s modus ponens is another’s modus tollens

Like the “proving too much” charge, philosophers tend to level the “begging the question” charge too hastily. I just had a paper—in which I argue that p is false (where p is a commonly held view in field F)—rejected on the grounds that my argument against p begs the question.

Strictly speaking, an argument begs the question only if it is a circular argument and the circle is vicious. That is:

            p => p

This is a viciously circular argument because the conclusion (= p) is assumed as a premise.

Now, if my argument were something like this:

~p => ~p

then it would have been viciously circular. But my argument actually goes like this:

(q -> ~p) & q => ~p

It just so happens, however, that q is a proposition that my opponent cannot accept. Does that make my argument viciously circular? I don’t think so. I suppose that is what Putnam had in mind when he said that “one philosopher’s modus ponens is another philosopher’s modus tollens.”

What do you think, my fellow pupae?

Comments

[Matt]

Does your opponent not accept q because he knows that it spells doom for his argument, or does he have other grounds for rejecting q?

[Richard Yetter Chappell]

I tend to think that it comes down to whether the premises are likely to seem antecedently plausible to your opponent. If your q very obviously entails ~p, then it would be a bit pointless to argue "q, therefore ~p", since that's never going to persuade anyone. But if it's a non-obvious implication, such that some current p-believers are probably also q-believers, then it's a worthwhile argument to make (and hence not "question begging" in any bad sense).

[Richard Yetter Chappell]

For a slightly more detailed development of that basic account, see:
http://www.philosophyetc.net/2008/05/assessing-arguments-and-begging.html

[David Morrow]

Supposedly, Schopenhauer once said that if you're debating someone who offers an argument whose conclusion follows so inescapably from its premises that you have no way of refuting it, the only thing left to do is accuse the other person of begging the question.

I was initially tempted to say that your argument begs the question if and only if there is no good reason to accept q unless you already accept ~p. But then it would always be open to a believer in p to object that any reason to accept q would be undercut by the truth of p, and so there can be no good reason to accept q unless you already accept ~p. That would make it impossible to argue for or against anything, so it can't be right. (Unless I'm begging the question about whether cogent argumentation is possible!)

So, I would say that your argument begs the question if and only if there is no positive reason to accept q unless you already accept ~p.

What if you added a section to the paper in which you discuss what attitude one should take toward q if one is agnostic about the truth of p? If you could show that a p-agnostic person would have good reason to accept q, I take it you could avoid the objection.

[Moti Mizrahi]

Thanks for the comments, everyone.

Matt: I assume that my opponent is arguing in good fate. So, perhaps s/he has other grounds for rejecting q, other than the fact that q entails ~p. In that case, I think that my paper can actually advance the debate, since it would challenge my opponent to articulate his/her reasons for rejecting q.

Richard: I agree that we need to say what makes an argument question-begging in a vicious way, but I am not sure that “obviousness” is the way to go. It is too relative for my taste. To me, it is obvious that q entails ~p, but I suppose that it is not so obvious to my opponent. And it also gives my opponent an opportunity to challenge the cogency of my argument simply by saying that the entailment is not “obvious.”

David: I like your suggestion to add a section about why p-agnostics should accept q. Thanks!

[Richard Yetter Chappell]

Moti - On my account, if the entailment is non-obvious to your opponent, then that's a good thing! It means that your argument may surprise them, and make them rethink their view. In short, it means that your argument is *not* question-begging.

It's true that my account introduces an element of contingency: an argument that's non-question begging in our philosophical community might be question-begging in the context of a different community. But that seems right to me. (After all, it's certainly true that the property of being dialectically effective is community-"relative" in this way. And I understand begging the question as simply a particular means by which an argument might fail to be dialectically effective: namely, because the connection between the premises and conclusion is too transparent to the relevant audience, rendering the reasoning trivial, or effectively vacuous.)

[Moti Mizrahi]

Thanks for the clarification, Richard.
Sorry I misconstrued your point about obviousness.

But I still worry that, on your view, an author who argues against a popular view p might be at a disadvantage more often than not. Suppose that p is a popular view in a philosophical community, and an author argues that q entails ~p. If the entailment is too obvious, the author’s opponents might complain that it is trivial. If the entailment is not obvious, the author’s opponents might complain that q doesn’t entail ~p after all. It seems like a no-win for the author. What is an author to do to avert that?

[Richard Yetter Chappell]

Well, hopefully the meat of the argument will illuminate / *explain* why q entails ~p. (Philosophy papers are usually more than 3 lines long, after all!)

Note also that it's okay to be obvious in *retrospect*. So it's not like a catch-22 where your very persuasiveness dooms you to failure :-). But I do think that a good, interesting argument needs to be capable of jolting its audience into thinking about things differently, which it can't do if its point is too obvious from the start.

[Fritz Warfield]

See 2 of Roy Sorensen's finest papers:

1. P therefore P without circularity

and

2. Unbeggable questions