My exchange with ‘Ramanan’ in the comments thread of my last post helped me to appreciate the discipline I’m in.
In philosophy, I deal regularly with people who disagree passionately with my views and aren’t shy about voicing their disagreement. But I can almost always expect certain standards to be maintained.
For example, if somebody argues that Q because P, and I respond by arguing that Q doesn’t follow from P, I can expect one of three responses:
- an argument to show that Q does, in fact, follow from P,
- an alternative argument for Q, not based on P, or
What I do not expect is for my interlocutors to accuse me of denying P. I expect my interlocutors, whether they know any formal logic or not, to understand that ¬(P→Q) doesn’t entail ¬P. And even if I had to point this out, I wouldn’t expect to be met with repetitions of the original accusation: ‘But you denied that P!’, ‘Why should we listen to somebody stupid enough to deny that P!’, ‘Look everybody! This moron denies that P! Now he’s trying to say other stuff! Let’s ignore him!’
Or suppose I give two independent arguments for the same claim. If one argument is shown to be weak, I expect attention to become focused on the other. I expect my interlocutors to know that it only takes one good argument to justify a claim. I do not expect them to keep harping on about the rejected, weak argument, as though continually drawing attention to that in any way detracts from the alternative, independent argument.
Or suppose that I say something ambiguous. I mean X, but my interlocutor thinks I mean Y. I expect to be able to explain myself in less ambiguous terms and then have the conversation focus upon the claim I actually mean to make. I do not expect to be accused of inconsistency, duplicity, ‘shifting the goalposts’, or anything else that depends on the assumption that I changed from asserting Y to asserting X rather than simply having spoken ambiguously. Nor do I expect to keep hearing new arguments against Y, so that I have to keep explaining that it isn’t what I mean, only to face again the charges of inconsistency and so on.
All of these things that I do not expect I encountered in my dispute with ‘Ramanan’ and I encounter often when I argue outside philosophy. They and other similar operations (in the past I mentioned Gregory Mankiw’s Circle of Hell and William Lane Craig’s flooding strategy), are cheap rhetorical tricks, aimed at winning arguments without having to legitimately justify a position (Craig’s trick, I hasten to add, is far less cheap than Mankiw’s, and both are considerably less cheap than the tricks mentioned above). They’re brute force strategies: by repeatedly refusing to acknowledge what your opponent is actually arguing, you bore or frustrate her into giving up on arguing it. Then you ‘win’.
If people generally respected basic norms of reasoning then the outcome of this strategy wouldn’t look like a victory at all. But by no means do all intelligent people respect basic norms of reasoning.
I don’t know how to argue with somebody who doesn’t respect such basic norms. Arguing with such a person while respecting the norms yourself is like trying to fend off a rabid pit-bull while obeying the Queensberry Rules. Using the cheap tricks yourself is like trying to fend off a rabid pit-bull by imitating it; the dog’s been doing it for longer than you have. From Fox News to the House of Commons we’re all being pushed around by people willing to use the cheap tricks. If we ignore them they go on to bully and influence others. A counterstrategy would be nice, but I can’t think of one.
Notice that these cheap tricks aren’t the same as the ‘logical fallacies’ that pop logicians like to talk about: the ‘strawman’, the argumentum ad hominem and ad populum, question-begging, post hoc ergo propter hoc, etc. These ‘fallacies’ are usually presented as faulty inferences rather than deliberate attempts to cheat in an argument.
The ‘logical fallacies’ are also mostly a sham. People who talk about them rarely know anything about logic. One internet source I examined seemed to think that any inference-pattern with a Latin name is a fallacy: reductio ad absurdum was classified as a fallacy! (perhaps the author is a dialetheist) Several others claim that circular arguments are invalid! (surely even a dialetheist accepts A→A)
Many of the ‘logical fallacies’ have little to do with logic (e.g. strawman). Of course they appeal to standards of argument outside of basic inference rules. But on their own they carry no information about those standards, and so on their own they are generally useless for assessing arguments.
For instance, post hoc ergo propter hoc or ‘confusing correlation with causation’ can be hurled against a perfectly acceptable inductive inference. The fact is that correlation in many circumstances is, rightly, taken to indicate causation. A certain type of correlation is the best evidence we have of the presence of a causal relationship. In other cases the same type of induction would be patently invalid. But it remains a great philosophical mystery to explain precisely what makes the difference between the valid and the invalid cases. The problem with treating post hoc ergo propter hoc as a blanket fallacy is that it reduces this complex issue to a slogan that sounds for all the world like a bare rejection of inductive reasoning in general.
Likewise, one can cry ‘strawman’ when somebody criticises any argument that isn’t identical to the one being made. But the argument criticised might be formally equivalent to the one being made. Or it or might be simply an improved version of it. In this case, the strawman ‘fallacy’ is in fact extremely informative: it demonstrates that since the ‘strawman’ argument is subject to criticism, so must be the argument originally made. Crying ‘strawman’ only distracts us. It covers over genuine logical insight with pseudo-logical sloganeering, as does mention of the ‘logical fallacies’ in general. An argumentum ad hominem might be a perfectly legitimate impeachment against the credibility of a witness. A ‘question-begging’ argument might informatively reveal the semantic equivalence of certain terms or propositions. The retreat into slogans simply dulls our receptivity to such sources of information. Or, more often, it indicates that those thus retreating are already not very receptive.
Often, invocation of the ‘logical fallacy’ terminology is combined with usage of the cheap rhetorical tricks mentioned above. Look what Mr. Spock does to the poor helmsman in the video below. He distorts his perfectly valid argument into an invalid one by ignoring a crucial premise – ‘They can’t all be wrong’. Then he explains the proposed invalidity in terms of a ‘logical fallacy’ (argumentum ad populum or the ‘bandwagon fallacy’) that would appear to concern only soundness. THIS IS HIGHLY ILLOGICAL.