Monday, February 23, 2009

Detecting Informal Fallacies

At someone asked the following question:

How does one _prove_ that an informal fallacy is a fallacy (instead of just waving a Latin name?)

It has three responses so far, but none of them answer the question, because all three responses explain ways in which you prove that someone has committed a formal fallacy. Peter Smith suggests counterexamples. But counterexamples won't help if the argument is not intended to be strictly deductive. William Rapaport recommends truth tables. But truth tables won't tell you if an argument is question-begging (as he fortunately recognizes). And Allen Stairs recognizes that the issue is about informal fallacies, but he simply recommends counterexamples again, although with qualification.

Here's a good argument:

The wall is made of stones.
The stones are grey.
Therefore the wall is grey.

It's not demonstrative, by any means, but everyone will recognize that it's reasonable, and if the conclusion is false, everyone will look for the reason. Here is an exactly parallel argument:

The body is made of cells.
The cells are small.
Therefore the body is small.

This is dubious at best. But the fact that this argument is exactly parallel to the other one doesn't lead to the conclusion that the other argument is dubious. They have the same formal structure, but the inference is licensed not by the structure but by assumptions about constitution and the relations between parts and wholes -- and these are different for each argument. The error, in fact, is that the latter treats size as if it could be used in an part-whole argument the way something like color can be used in part-whole argument; in a sense, the error is in the use of the parallel structure in a context where it is not viable. And obviously this won't be caught when you run it as a parallel argument.

So how do we detect informal fallacies? A problem with the question is that informal fallacies are simply anything that might be called a fallacy but are not formal fallacies; and this is an extraordinarily diverse group of errors. A further problem is that most accounts of informal fallacies are of very poor quality, consisting of bits and pieces of philosophical folklore that have been piled together in a mish-mash. I've argued before that one of the things that interferes with the development of an adequate account of informal fallacies is the consistent failure to distinguish rhetorical tactic from sophistical error. I think there are lots of other pitfalls in current wisdom about informal fallacies. So I think the only full answer to this question would be, "We'll call you back when we eventually figure out what's actually going on in informal fallacies." In the meantime, a bit of common sense is the best test: ask yourself whether the argument is irrelevant, whether it equivocates or changes the subject, whether the argument is smuggling in the conclusion as an assumption, whether it needs any obviously dubious assumptions to make sense, etc.

Incidentally, here is an argument of the form Stairs suggested that is not patently bad, contrary to his claim that any argument of this form is so:

Most arguments that have forms explicitly mentioned by Stairs in his comment use this argument form.
This argument uses this argument form.
Therefore this argument (probably) is an argument that has a form explicitly mentioned by Stairs in his comment.

In fact it's arguably quite good, and would stay good if we dropped the 'probably'; one could argue that it is, as we might say, 'materially valid', because although its form doesn't guarantee truth-preservation, the meaning of the premises does. The conclusion that any argument of this form is patently bad is a hasty generalization.

No comments:

Post a Comment

Please understand that this weblog runs on a third-party comment system, not on Blogger's comment system. If you have come by way of a mobile device and can see this message, you may have landed on the Blogger comment page, or the third party commenting system has not yet completely loaded; your comments will only be shown on this page and not on the page most people will see, and it is much more likely that your comment will be missed.