"How am I to get in?" asked Alice again, in a louder tone.
"Are you to get in at all?" said the Footman, "That's the first question, you know."5
A question with a false, disputed, or question-begging presupposition.
Why should merely cracking down on terrorism help to stop it, when that method hasn't worked in any other country? Why are we so hated in the Muslim world? What did our government do there to bring this horror home to all those innocent Americans? And why don't we learn anything, from our free press, about the gross ineptitude of our state agencies? about what's really happening in Afghanistan? about the pertinence of Central Asia's huge reserves of oil and natural gas? about the links between the Bush and the bin Laden families?6
A "loaded question", like a loaded gun, is a dangerous thing. A loaded question is one with a false or questionable presupposition, and that is what it is "loaded" with. The famous question "Have you stopped beating your wife?"―the example given in almost every discussion of this fallacy―presupposes that you have beaten your wife prior to its asking, as well as that you have a wife. If you are unmarried, or have never beaten your wife, then the question is a loaded one.
Since this example is a yes-or-no question, there are only the following two direct answers:
- "Yes, I have stopped beating my wife", which implies that I was beating my wife.
- "No, I haven't stopped beating my wife", which implies that I am still beating my wife, since you cannot stop something that you never started.
So, either direct answer implies that you have beaten your wife. Thus, a loaded question is one that you cannot answer directly without implying a falsehood, an unproven allegation, or something that you deny. For this reason, the proper response to such a question is not to answer it directly, but to either refuse to answer or to explicitly reject the question.
Aristotle was the first logician to both list fallacies and classify them by type. There were thirteen fallacies on Aristotle's list, including this one, making it one of the oldest recognized fallacies. He classified fallacies into two types: those dependent on language, and those not so dependent. Among the latter was "the making of more than one question into one"7. He describes the fallacy in the following passages:
[Fallacious arguments] that depend upon the making of two questions into one occur whenever the plurality is undetected and a single answer is returned as if to a single question. Now, in some cases, it is easy to see that there is more than one, and that an answer is not to be given, e.g. "Does the earth consist of sea, or the sky?" But in some cases it is less easy, and then people treat the question as one, and either confess their defeat by failing to answer the question, or are exposed to an apparent refutation. Thus "Is A and is B a man?" "Yes." "Then if any one hits A and B, he will strike a man" (singular), "not men" (plural). …
Those fallacies that depend upon the making of several questions into one consist in our failure to dissect the definition of "proposition". For a proposition is a single statement about a single thing. … If, then, the answerer has returned an answer as though to a single question, there will be a refutation; while if he has returned one not really but apparently, there will be an apparent refutation of his thesis.8
Later, he explains how to answer such questions:
To meet those refutations which make several questions into one, one should draw a distinction between them straight away at the start. For a question must be single to which there is a single answer….9
- Some systems of parliamentary debate provide for "dividing the question", that is, splitting a complex question up into two or more simple questions. Such a move can be used to split the example as follows:
- "Have you ever beaten your wife?"
- "If so, are you still doing so?"
In this way, the first question can be answered directly by "no", and then the second, conditional question never arises.
- Since a question is not an argument, simply asking a loaded question is not a fallacious argument. Rather, loaded questions are typically used to trick someone into implying something they did not intend. For instance, some salespeople are taught to ask such loaded questions as: "Will that be cash or charge?" This question gives only two alternatives, thus presupposing that the potential buyer has already decided to make a purchase, which is similar to the Black-or-White Fallacy.
Reader Steven Flintham asks the following unloaded question:
Q: "I've just been browsing your site and the page on loaded questions reminded me of something I came across ages ago without ever getting quite clear in my mind. Although it looks misleading, if I don't have a wife or have never beaten my wife, isn't it strictly accurate to answer 'No' to the question 'Have you stopped beating your wife?'? I haven't stopped, after allI never even started."
A: The answer to your question turns upon a subtlety about presupposition. Putting aside the unpleasant example of wife-beating, let's use as an example the question: "Have you stopped cobra-milking?" This is equivalent to saying: "You have stopped cobra-milking: yes or no?"
Consider the contained statement: "You have stopped cobra-milking". Clearly, this means: "You have milked cobras but you are not now milking them." However, these two conjuncts are not equal: the first conjunct is a presupposition of the question. A presupposition to a question is a statement which is normally known to be true before the question is asked.
Given that our example question is a yes-or-no question, there are two direct answers that we can give it:
- "Yes": "I have stopped cobra-milking" or, equivalently, "I have milked cobras but I am no longer milking them." Obviously, this implies "I have milked cobras."
- "No": "It is not the case that I have stopped cobra-milking" or, equivalently, "It is not the case that both I have milked cobras and I am not now milking them", which implies: "Either I have not milked cobras or I am now milking them." In other words, there are two bases for answering "no" to the question:
- You have never milked cobras.
- You are now milking them.
So, you are right, Steven, that you could answer the loaded question "Have you stopped cobra-milkinging?" with "No", because you have never milked a cobra. However, this answer has a kind of ambiguity, since it leaves it open as to whether you are saying that you have never milked a cobra or that you are still doing so. This is why it is misleading to simply answer "no" and leave it at that; one should at least say, instead: "No, I've never milked a cobra so I can't very well stop."
However, since the claim that you have milked a cobra is a presupposition of the question, we normally presuppose that it is true or the question would not arise. This leaves as the only possible reason for denying the question that you are still milking them. So, the second direct answer also commits you to milking cobras, though it does not logically imply it by itself, but it does imply it when taken together with its presupposition.
This is why loaded questions as a fallacy are sometimes classified as a type of question-begging. By loading some controversial or even false presupposition into the question, the unscrupulous questioner tries to sneak it in unchallenged.
Reader Doug Merritt raises the following objection:
You overlooked something important when you said, "Since a question is not an argument, simply asking a loaded question is not a fallacious argument." Loaded questions are, as you said, rhetorical devices, but in particular they are "rhetorical questions", and a rhetorical question is a form of statement that is merely phrased superficially as a question; it is not a true question. So I strongly disagree that a loaded question is not an argument. It is just as much of a (fallacious) argument as all the other categories in your taxonomy. This isn't merely splitting hairs; a debate can be won (in the view of the audience) by someone who does nothing at all but pose loaded questions; each serves to presume things for which no logical argument is offered, and it is very effective in practice.
Even if you're right, Doug, that a loaded question is not really a question, but is a statement, then it's still not an argument by itself. A loaded question is often a way of rhetorically making a statement, but it is grammatically a question. You will notice that as a logical fallacy it is in a category by itself, as can be seen in the Taxonomy of Logical Fallacies. This is because Loaded Question is a fallacy of questioning, as opposed to arguing. So, I don't disagree with you that asking loaded questions can be a powerful and fallacious rhetorical trick. Rather, I simply think that fallacies involving questions should be classified as a different type of fallacy than those involving arguments. I hope eventually to add further fallacies of questioning, so that Loaded Question will no longer be in a class by itself.
This is a series of loaded questions and it illustrates one of the common uses of the loaded question as a rhetorical device, namely, innuendo. The questions come at the end of the article, and presuppose the following controversial claims:
- The American government did something to bring about the terrorist attacks of September 11th, 2001.
- The public doesn't learn anything from the press about that government's mistakes.
- The public is not learning about what's happening in Afghanistan.
- Central Asia's oil reserves are somehow pertinent.
- There are some unspecified links between the Bush and bin Laden families.
No evidence is given in the article for any of these claims. Loaded questions are used in this way to slip claims into rhetoric without the burden of proving them, or the necessity of taking responsibility for unproven assertions.
- T. Edward Damer, Attacking Faulty Reasoning: A Practical Guide to Fallacy-Free Arguments (3rd edition, 1994), pp. 81-83.
- S. Morris Engel, With Good Reason: An Introduction to Informal Fallacies (6th edition, 2000), pp. 167-171.
- David Hackett Fischer, Historians' Fallacies: Toward a Logic of Historical Thought (1970), pp. 8-12.
- Madsen Pirie, How to Win Every Argument: The Use and Abuse of Logic (2013), p. 29.
- Lewis Carroll, Alice's Adventures in Wonderland, Ch. 6.
- Mark Crispin Miller, "Brain Drain", Context, No. 9.
- Aristotle, On Sophistical Refutations, tranlated by W. A. Pickard-Cambridge, §1, Part 4.
- Ibid., §1, Parts 5 & 6.
- Ibid., §3, Part 30.