# Complex question triple play

Many are familiar with the fallacy of the complex question, perhaps in the form of its most well known example:

when did you stop beating your wife?

The trick consists in cramming two questions into one such that a response to one of them (you can after all only answer one question at a time) looks like a response to the other. So if you answer "I haven't" to the above question then you admit to beating your wife, but  you thought you were denying beating your wife.  Chris Wallace of Fox News tried this out on Bill Clinton a few years back.  He asked: why didn't  you do more to stop al Qaeda?  Clinton, smart guy that he is (whatever else you may want to say about him) attacked the question.  Why this is called a "fallacy," by the way, is really beyond me, since no inference is really drawn.  Perhaps there's an inference drawn at the end when the person responds to the trap.  It seems to me to be more of a trick than a fallacy.

In any case, most examples of it that I have seen involve two questions.  My informal sense is that the structure forces a negative answer to the trick part of the question which looks like an affirmative answer to the assumption.  But I'll have to think about that a little more.

But it doesn't seem to me by the way that one needs to be restricted to two questions.  Why not three?  I can only get as far as three in a complex question.  But I fear I may have not thought hard enough about it.  Here's my example:

Why must you persist in doing that?

That's three questions: (1) why do you do that? (2) must you do that? (3) why do you persist in doing that?

Why can't anyone come up with more?

## 11 thoughts on “Complex question triple play”

1. Gary Herstein says:

“Why this is called a “fallacy,” by the way, is really beyond me, since no inference is really drawn.  Perhaps there’s an inference drawn at the end when the person responds to the trap.”

My best guess would be that it seems that the inference has already been drawn in the asking of the question, since the circumstances that make the implied question meaningful are already assumed to be the case — that you’ve actually been beating your wife, etc.

2. hR says:

Yes, the inference is the accusation contained in the question.

3. Maybe.  But the problem concerns the nature of the question, not the assertions implicit in it (which aren’t inferences–though they may be the product of other inferences).  The problem is that the question doesn’t allow for a clear answer.  Perhaps the fallacy might occur by claiming the response to the question (that is, the response that falls for the trap) has some kind of probative value (when it doesn’t).

4. Gary Herstein says:

But the question, by the very fact that it is being posed, asserts itself as a meaningful question. This in turn requires the assertion of the propositional assumptions that must necessarily be taken as true for such meaningfulness to occur. If these claims are themselves not legitimate, then the logical structure of the question is itself flawed.

Somehow it managed to get mispacked in one of my  moves, else I’d be referring more directly to Belnap’s The Logic of Questions and Answers for this one. There’ve been a couple of recent articles in JPL on the subject as well. But since ASL terminated their relation to Springer, I can’t look it upon on JSTOR. (Living out of a garage sucks.)

But the fallacy has already been committed in the way the question is posed. The respondent might be guilty of permitting the interlocutor to get away with a bad form of reasoning, but that can occur with any type of fallacy.

5. John,
Re: the question why complex question is a fallacy.
Fallacies, I think, are generally undertheorized, and complex question is a perfect case in point.  Here’s a dialectical-epistemic story. Any question, really, is complex, since there must be a background of some facts taken for granted for the question to be about.  Even the most general-style questions: asking ‘why is there something rather than nothing?’ presumes that there is something, not nothing; asking ‘what’s the meaning of life?’ presumes that there is at least a class of living things (though it may be empty at some time — God, perhaps, could ask himself what’s the point of life before he, ahem, created it).

It’s not the complexity of questions that’s objectionable, but the unwarranted assumptions behing complex questions.  So, asking ‘how does ESP work?’ or ‘why do aliens abduct farmers and their lucious daughters?’ presumes certain facts that are illicit, at least given the standing evidence.

So at first blush, what’s fallacious about complex questions is that in posing them, one (as I think Gary points out rightly) also proposes the assumptions behind the quesiton as acceptably presumed.  It’s a smuggler’s tactic, really.

What’s also so difficult is that all questions have a certain grammar for answers.  If I ask what time it is on the sun, the answer’s frame is presumed to be in an “o’clock” idiom.   When people change the idiom of the answer, they must break the flow of the exchange.   And this is the trick of the complex question, in that if you answer in any of the accepted forms for answering it directly, you concede the suppressed/assumed illicit propositions.  And so it is not really an inference, but a performance that licenses further inferences, since a failure to object is often taken as acceptance.   That’s not a problem if the complex assumptions are broadly acceptable, but it is when they are either (a) not epistemically supported or (b) ones your interlocutor rejects.

6. BN says:

“Why can’t anyone come up with more?” đź™‚
jcasey, I always thought that a fallacy has to be an argument, but I guess not…

7. Thanks for all of the comments.  I’m convinced, by the way, of the wrongness of the complex question.  I just don’t think, as I think Aikin confirms, that its wrongness consists in faulty inference drawing (like most or all of the other fallacies).  I would also agree with Aikin that there is much wanting about fallacy theory.  Since Hamlin (1971) or so this has been a subject of much philosophical research.  I don’t have (as I think I say elsewhere here) a unified theory of fallacy.   The most I usually say (as BN points out) is that a fallacy is some kind of faulty inference (implicit or explicit).  The inference, if any, in the complex question might occur well after the question, if at all.

In any case, I was really hoping for people to come up with the most complex question they can, just for fun.  Have at it.

8. John,
Right that, if complex question is to be called a fallacy on the basis of an inference, it’s not the inference in the argument of asking the question, but the inferences licensed by its being unchallenged.  So complex questions aren’t arguments, but a epistemic-dialectical perspective on fallacies doesn’t require that fallacies be arguments or inferences, but contributions to discussions that distort the discussion’s product.

And so a triple play with complex question: “Are complex questions legitimately classified as fallacies on the basis of their comprising inferences?” (1) Complex questions have been are classified as fallacies on the basis of their comprising inferences. (2) Fallacies are fallacies on the basis of their comprising inferences. And (3) There is presumptive reason to believe that complex questions are not composed of inferences, or that, at least, there is reason to doubt the classification of fallacies as being exclusively inferential.

9. Thanks Scott.  Speaking of non-inferential fallacies, I’m at a loss (too much beer?) to think of any at the moment.  Thoughts?

10. Colin says:

Off the cuff, it seems to me that the fact that complex questions seem like fallacies might best be understood on the sort of “logical pragmatics” (dialectics) that Walton prefers. Situating argument in “persuasion dialogues” or in a dialectical context, lets us broaden the notion from “fallacy” to something like “illicit moves” in a dialogue. If the goal of the dialogue is persuasion, then there are rules of logic that determine right and wrong moves. Not all of these are able to be represented syllogistically, without often a bit of violence. (I’d place the “appeals” (baculum, populum, misericordiam) in an similar group. They’re often a bit forced when restated as syllogisms.)

I think complex questions are included, perhaps, because traditional conceptions of fallacy assumed the dialectical context. We reifying moderns want a form or a structure to measure against. The best I can come up with is that the conclusion of the inference is  implicit or understood by context and one of the premises (a Toulminy warrant) is taken to be obviously true. The complex question serves to secure the Toulminy data, by a sort of trick.

People who beat there dogs are bad.

Because the only controversial claim is the “data,” the argument hangs on tricking someone to admit the premise.

So although the complex question is not inferential as such, it is inferential in the sense that the conclusion and warrant are implicit in the dialogue, or context, and the arguer takes the answer to the complex question to be sufficient for the conclusion.

Perhaps a way of putting it is that a complex question fallacy is a rhetorical-dialectical manipulation of a falsely dichotomous disjunctive syllogism.

Either you beat or you used to beat your dog.