Beg Your Pardon, What is Begging the Question?

We’ve all met them. Usually they are fresh off of a critical thinking, or informal logic course. They are the fallacy mongers. Taught to identify informal fallacies in headlines and textbooks, they begin to “see” fallacies at every turn. And suffering them in any conversion is nearly intolerable. For those unfamiliar, I am talking about people who behave like this. Now, I am not saying that it isn’t important to be able to know and be able to identify informal fallacies. It is. But it can also become a hammer that turns all arguments into nails. This is especially dangerous because informal fallacies tend to be vaguely defined, and often resemble perfectly good methods of reasoning. Pro-tip: When you encounter such people, inform them that it is not sufficient to merely burp up fallacies at you. Ask them to explain to you what the fallacy means, and specifically how your argument matches the description of the argument. If they misrepresent your argument, resist the urge to shout “strawman” or any other fallacy term at them, as this only aggravates the situation.

As of late, however, I’ve noticed that one fallacy-accusation has become ubiquitous in debates, reviews, conferences, and blogs. I think of it as the last refuge of the philosophical scoundrel since, if all else fails, you can be sure that they will tell you that you’ve committed this particular fallacy. Surprisingly, it is not uncommon to see professional philosophers smugly appeal to it, brush the dust off of their shoulders, and walk off as if they’ve done you a huge favor. Which fallacy is this? It is none other than Begging the Question. So here, I want to discuss the fallacy: what it is, and what it is not.

Begging the question is an informal form of reasoning in which the question under discussion (typically the conclusion of the argument being discusses) is assumed in one premise or another. I say one premise or another because, in a sense, every conclusion is contained in the premises of a valid argument. A valid argument just takes two or more premises together and shows that another proposition is logically entailed by taking those premises together. This can confuse some people, because they will see a series of premises, note that, jointly, they logically entail the conclusion, see that all the premises are uncontroversial except for one, and then cry that the one premise is question begging. This may not be the case if that one premise can be a) justified by considerations that are independent from the conclusion, and b) the premise is not an obfuscated restatement of the conclusion.

So when has an argument committed the fallacy of begging the question?

  1. When one premise or another is logically equivalent to the conclusion.
    • Example: 1. It is not the case that if God exists, then God does not exist. Therefore, 2. God exists.
      • Premise one is logically equivalent to the conclusion, even if it seems reasonable. One could literally “prove” the existence of anything by swapping out “God” for another term.
  2. When one premise or another is semantically equivalent to the conclusion.
    • Example: 1. Morphine has dormative virtues. Therefore, 2. Morphine has the power to induce sleep.
      • This example comes from Molière’s Le Malade Imaginaire. There is a semantic equivalence between the premise and the conclusion because “Dormative” just means “sleep inducing” and a “virtue” just means a “power”. So if there is any question as to whether morphine has the power to induce sleep, this sort of argument would be useless as a proof.
  3. When the only justification one can provide for accepting one premise or another is the conclusion of the very argument where the premises appear.
    1. Example: 1. The Bible is the inerrant Word of God. 2. Whatever is inerrant reports no factual errors. 3. The Bible reports that God exists. Therefore, 4. God exists.
      • This is where things get a bit tricky. Technically, I would call this circular reasoning rather than begging the question, but some want to treat circle reasoning as a subset of begging the question. The idea is that A and B imply C. You might defend the truth of A by appealing to D and E, but you say that B is true because of F and C. Well, wait a minute, that means that the only reason you gave to accept B is that C is true. But that is precisely what you are trying to prove! So,an argument is circular if the justification for one premise or another somehow works back to the conclusion. In this example, the reason to think that (1) the Bible is the Word of God will eventually work back to some justification that depends upon the assumption that conclusion, God exists, is true. Since that is the question under dispute, it cannot be used to justify a premise that leads to that conclusion.

Now here are some circumstances where I don’t think one has necessarily begged the question. I say not necessarily because they may have done so in other ways along with meeting the criteria of these circumstances:

  1. When the only people who are willing to accept the truth of the premises are those who are willing to accept the conclusion, or, to put it the other way, those who are willing to accept the truth of the conclusion are only those who are willing to accept the premises of the argument in question.
    • I find that this often leads to the charge of “begging the question” because one has reached the conclusion of the argument he or she is incredulous about the conclusion. So he or she declares that one premise or another must be more dubious than initially thought. Perhaps such a person reasons that everyone else already accepts the conclusion and so have not made the effort to pay the cost of abandoning one of the premises. Such a person would be incorrect to say that the argument begs the question, though, unless a premise can be doubted for the expressed reason that the justification for the premise depends upon conclusion in the ways that I outlined above. The fact that one finds a premise to be dubious, or is more willing to deny a premise than admit a conclusion, is not sufficient to establish that the given premise is question begging in and of itself. In such a case, one should simply deny the premise(s) that he or she finds dubious, and live with the consequences of that commitment.
    • Also, it is not enough to simply note that only people accept the conclusion are those who accept the premises. Suppose, for instance, that there is a certain proposition, p, for which there is only one argument that believers in p find cogent. Let’s also say that no believer in p happens to believe in p without justification (i.e. all believers in p depend on the same set of premises to justify their belief and no one believes p on faith or because p is properly basic). All other arguments for p are universally panned by believers and non-believers alike. That there is only one accepted argument to justify belief in p, and that it happens to be the case that all believers accept p on the basis of the premises contained in that argument is really just a report of the psychology of believers and non-believers with respect to p and arguments in support of p. It is perhaps an accident of history, or something to do with the nature of p that has led to this circumstance. But simply noting the fact that people who accept p also happen to be the same people who accept the those premises that lead to the conclusion p does not sufficiently establish that the argument for p is question begging.
  2. When an argument is made without stating independent justification for the premises.
    • Simply observing that a premise has not been justified independently of the conclusion does not establish that the premise somehow begs the question. At best, one can say that the premises are unjustified or unwarranted. Assuming that the justification or warrant for the premises is somehow found in the conclusion is itself fallacious. One has made an argument from ignorance. The best course of action is to ask one’s interlocutor to give some justification for the premises, if the justifications are not apparent. It may then be the case that the argument is question begging, but only if the premises assume the conclusion in the ways I’ve noted above.
  3. When an argument raises other questions that need to be answered.
    • The idiom “this begs the question” really just means that some statement or fact raises other sets of questions that need to be addressed. That a conclusion of an argument raises other perhaps deeper questions is not a mark against the argument. Something akin to this occurred in Richard Dawkins’ The God Delusion when he used Dennett’s distinct between “skyhook” and “crane” explanations to complain that “God” is merely skyhook explanation given that God raises more questions, in Dawkin’s mind, than answers (like who designed the designer, which is not a question my mind happens to raise). This is not to say that Dawkins actually said that “skyhooks” beg the question, but others made this blunder. For instance, David Thompson writes:

The only thing the evolutionary algorithm is guaranteed to deliver is design: it produces organisms that are “fit” for their environment, i.e., that are well designed. Dennett rejects the idea that we can explain design by appealing to a pre-existing Design (the Platonic option) or a conscious Designer (the theistic option). He describes such explanations as “skyhooks,” –devices hanging from the sky with no visible means of support that still do the heavy lifting in design space. If we are to explain how design comes to be in the first place, we get nowhere by assuming that design already exists. Design and designers are what need to be explained; minds are products of evolution, not (at least initially) producers of it. The Darwinian, algorithmic approach is the only one that is not question-begging (2009, 62).1

What is extraordinary about this passage is that it not only mistakes “raising new questions” with the fallacy of begging the question, it actually commits the fallacy of begging the question in the process of committing this error. An argument that concludes with “A designer initiated and produced the evolutionary algorithm” may raise the question (for some) of who designed the designer, but such a question isn’t question-begging. Rather, it reveals a certain assumption in the questioner, that designers need to be designed! Intelligent Design theorists think that things like minds require explanations that go beyond evolution. But to show that intelligent design is an inadequate explanation for certain design features, Thompson begs the question by assuming that the evolutionary algorithm is the only way to explain minds. He essentially argues that God’s mind cannot explain design because all minds are the products of evolution and no minds initiate evolution. But that is precisely the question at issue in the debate. To put this more pointedly, Thompson reasons that it is not the case that there exists a mind that initiated evolution because there is no mind that initiated evolution. And we can show that his one premise is logically equivalent to his conclusion, where Mx is ‘x is a mind’ and Ix is ‘x initiates evolution’ :

1. (∀x)(Mx ⊃ ~Ix) (assumption for CP)
2. (∀x)(~Mx ∨ ~Ix) (1 Impl)
3. (∀x)~(Mx & Ix) (2 DeM)
4. ~(∃x)(Mx & Ix) (3 QN)
5. (∀x)(Mx ⊃ ~Ix) ⊃ ~(∃x)(Mx & Ix) (1-4 CP)
6. ~(∃x)(Mx & Ix) (assumption for CP)
7. (∀x)~(Mx & Ix) (6 QN)
8. (∀x)(~Mx ∨ ~Ix) (7 DeM)
9. (∀x)(Mx ⊃ ~Ix) (8 Impl)
10. ~(∃x)(Mx & Ix) ⊃ (∀x)(Mx ⊃ ~Ix) (6-9 CP)
11. [(∀x)(Mx ⊃ ~Ix) ⊃ ~(∃x)(Mx & Ix)] & [~(∃x)(Mx & Ix) ⊃ (∀x)(Mx ⊃ ~Ix)] (5,10 Conj)
12. (∀x)(Mx ⊃ ~Ix) ≡ ~(∃x)(Mx & Ix) (11 Equiv)
13. ◻[(∀x)(Mx ⊃ ~Ix) ≡ ~(∃x)(Mx & Ix)] (12 NI)

Since the equivalence is necessary, there is no model where the truth conditions would fail. That is just to say that the premise is logically equivalent to the conclusion. No premises are required to know that “necessarily, no minds initiate evolution if and only if it is not the case that some mind initiates evolution”. Of course, no theist would agree that all minds are the product of evolutionary process, and that no minds can initiated evolution. Theism is the thesis that there is at least one uncaused mind, and that mind could design aspects of evolutionary processes and guide them. The answer to what explains God’s mind is found in the nature of his own necessity. But even if the premises or conclusion of an argument raise questions with no readily apparent answer, the argument is not invalidated.

So, in sum, the fallacy of begging the question can be tricky to spot. When arguments rely on one relatively controversial premise, ask yourself this: a) does my interlocutor provide justification for the premise independently from the conclusion, and b) is this premise merely a logical or semantic restatement of the conclusion. If the identified premise is logically or semantically equivalent to the conclusion, then pay attention to how that premise is justified. If independent justification is provided for the premise, you can help your interlocutor by noting that the real argument is the one that went into independently justifying the premise that you’ve identified as equivalent to the conclusion. That is a far more constructive way to handle a situation where you think an argument has been begged.

Cheers!

1D. Thompson. 2009. Daniel Dennett. New York: Continuum.

  • VQ

    Reblogged this on vexing questions and commented:
    Here is my recent contribution to Attack of the P-Zombies. Enjoy!

  • Hi there,

    Nice piece — I agree with the spirit of it wholeheartedly. Fallacies are rarely obvious and there are almost always other things in play which complicate matters.

    I have a couple of comments on question-begging. To me it’s an interesting one, and I think it plays a special role in philosophy. Used in academic contexts (or in discussions between anyone who knows vaguely what they’re talking about) I’ve found that it’s most often used as a way of tracking interpretive slack. So for example someone advancing a correspondence theory of truth might say, at a first approximation, that a proposition is true if and only if it corresponds to the facts. A critic might then wonder how this is not a question-begging analysis, since what is a fact if not a true proposition? But this needn’t be taken as an attempt to identify a fallacy so much as a way of bringing to light the new question which is raised by the correspondence theorist’s answer to the first, namely, how can they cash out truth in terms of correspondence without tacitly appealing to the notion of truth itself?

    Similarly, people are always accusing each other of begging the question in philosophy of mind. But this makes sense if we think of these attributions not as trying to identify particular instances of people advancing fallacious arguments so much as the possibilities of and constraints on certain strategies of conceptual analysis. For example, a key premise in the causal argument for physicalism is that every physical event which has a sufficient cause has a sufficient physical cause. People often worry whether this begs the question by depending on some version of physicalism or another to support it. Whether it does or not on any particular occasion will of course depend on what the person making it means by ‘physical’ and how they justify the premise. But wholesale rejections of the causal argument as begging the question can be thought of as raising concerns about the possibility of justifying such a claim — that any definition-justification pair will ultimately *have to* appeal to the thing it’s trying to argue for.

    Philosophy is not, in my view, a case of analysing arguments made with terms whose meaning everyone has already agreed on (not in general, anyway — philosophy of religion may be like this, but if so it is unique in that respect). Usually it is about assessing the explanatory trade-offs (expressed as arguments) between different analyses of terms whose meaning (or reference) we grasp intuitively but have not yet found suitable articulations (or explanations) of. So it’s like a huge game of conceptual whack-a-mole, and the discursive role of ‘question-begging’ is to keep track of the mole, i.e. the circularities that move around on different analyses, thereby raising questions in new places.

    Sam