Read the case State v. Ransom (pp. 411-425). Consider your verdict. Prepare a document that expresses your deliberation that justifies your vote of guilty or not guilty. The goal is not simply a w

13 Begging the Question

Listen to the Chapter Audio on mythinkinglab.com

When a magician draws a rabbit out of her hat, she simply removes what she had managed to hide there earlier. It’s a great trick, and it is often done very convincingly—but it doesn’t create any new rabbits. The same applies to question-begging arguments. It is easy to pull a conclusion out of an argument if I have previously hidden the conclusion in the premises. And it often looks impressive, but it doesn’t really prove anything new.

The problem with begging the question is that it spins its wheels, it goes nowhere: It does not take up the burden of proof and go forward with it. Instead, a question-begging argument simply assumes as a premise what it purports to be proving.

Arguments are supposed to give us reasons for believing the conclusion. But you cannot support the conclusion by using the conclusion. If we are trying to build a bridge across a river, we must first find a solid foundation, then sink the pilings into that foundation, and finally conclude by placing the trestle onto the pilings. If our civil engineer proposed using the trestle to support itself, we would look for another civil engineer. And if someone proposes to support a conclusion by appeal to that same conclusion, then we should look for another argument.

The Problem with Question-Begging Arguments

Notice that question-begging arguments are certainly valid. A valid argument is one in which the conclusion follows from the premises; or more exactly, if its premises are true then its conclusion must be true. If the conclusion is contained in the premises, then the conclusion must follow from the premises. If the premises are all true, and the conclusion is contained among the premises, then the conclusion certainly will have to be true. So the problem with question-begging arguments is not invalidity. Rather, the problem with question-begging arguments is that they turn in a circle: They don’t give us any real reasons for believing the conclusion to be true. An argument cannot pull itself up by its own bootstraps.

When the magician pulls a rabbit out of her hat, we all know that it was done by sleight of hand, but it’s not always easy to tell how the magician managed to get the rabbit into the hat. That’s also true of question-begging arguments: It’s not always easy to recognize the question-begging premises. But take the simplest cases first. If our magician—in plain view of the audience—sticks a rabbit in her hat and then pulls it out, we aren’t likely to be very impressed by the trick. And some question-begging arguments are almost that obvious. They seem to rely on the hope that if the conclusion is repeated often enough—in a loud and confident voice—then people will believe the conclusion to be true (despite the fact that no evidence has been given in support of the conclusion). For example, “Socialism will not work as an economic system, because it is perfectly obvious that socialism is not workable as an economic system.” Such an “argument” is rather like the bellman’s “argument” in Lewis Carroll’s “The Hunting of the Snark”:

“Just the place for a Snark!” the Bellman cried,

As he landed his crew with care;

Supporting each man on the top of the tide

By a finger entwined in his hair.

“Just the place for a Snark!

I have said it twice:

That alone should encourage the crew.

Just the place for a Snark!

I have said it thrice:

What I tell you three times is true.”1

Strange as it may seem, many people seem to think that merely repeating the conclusion is a convincing argument. Thus the district attorney in her closing “argument” to the jury argues, “The defendant is guilty as charged! He knows it, I know it, and, members of the jury, you also know it! He is guilty of mugging that poor old lady to steal her grocery money! He is guilty, and thus justice demands that you speedily return with a verdict of guilty!” Such “arguments” offer no reason to believe in the defendant’s guilt or the snark’s proximity. And I trust that neither you nor the bellman’s crew will be deceived by such brazen uses of question begging.

A New and Confusing Use of “Begs the Question”

Unfortunately, the phrase “begs the question” has recently taken on a new use, a use that is very different from naming the fallacy described in this chapter. This new use is often directed at suspected half-truths, as in the following: “Professor Snodgrass told us that some students do very well in this class; that begs the question of what happens to the other students.” Or as in this: “The label says that this pesticide does not cause cancer; that begs the question of what other health problems it might cause, such as nerve damage.” Obviously there’s no way to ban the use of “begs the question” in this context, but be careful not to confuse the two very different uses.

No one should be misled by the blatant begging the question practiced by the bellman. Unfortunately, begging-the-question fallacies also occur in much trickier forms. One is through use of synonyms.

Consider the following brief argument:

Socialism is not a workable economic system, because an economic system in which the means of production are collectively owned cannot work.

But socialism is an economic system in which the means of production are collectively owned. So the premise merely repeats the conclusion, albeit in a slightly disguised (synonymous) form. The premise gives us no independent grounds for believing the conclusion to be true, and the argument commits the fallacy of begging the question (the synonymous form of that fallacy).

The synonymous form of begging the question may sometimes appear in more elaborate disguises:

Abortion is wrong; for it is always wrong to voluntarily and purposefully destroy a living and growing and developing human fetus.

But that is precisely what abortion is. Merely restating a synonymous form of the conclusion is not giving an argument for that conclusion.

In some cases of begging the question, the premise is a generalization that contains the conclusion as an instance of that generalization. For example,

John Hinckley should not be excused on grounds of insanity; for no one should escape punishment for an assassination attempt on the grounds that he or she is insane.

Certainly if no one should escape punishment on grounds of insanity then it follows that Hinckley should not. But the question at issue—the claim that the “argument” is supposed to be proving—is whether Hinckley should be excused on grounds of insanity; merely stating a general principle that contains the conclusion as an instance is to beg the question.

Suppose we are discussing what courses should be general requirements for students at Home State University, and someone offers the following argument:

A critical thinking course should be required of all students at Home State University, because obviously every university student should have to take at least one course in critical thinking.

True, if all university students should be required to take critical thinking, then that includes the students at Home State. But the premise is just a more inclusive, general claim that begs the question at issue.

Consider another example of the generalization form of begging the question. In 1983, a Korean airliner (Flight KAL 007) was shot down by a Soviet war plane over sensitive Soviet airspace. Why was Flight KAL 007 flying over a major Soviet naval center during a scheduled Soviet training exercise?

Was it a strange sequence of tragic errors? Or was it the deliberate use of a commercial airliner for spying? Or what? It is an important question, and there have been interesting arguments for a variety of views. Suppose someone offers the following argument:

South Korean Flight KAL 007 was not on a spy mission because the United States and its allies never use commercial passenger flights for spying.

Such an argument begs the question. The question at issue is whether this particular flight was used for spying. To assume—as a premise, and without further argument—that no such flights are used for spying is to assume the very point in question, and thus to beg the question.

Of course not all uses of general principles involve begging the question. Suppose that some mercury is dropped on the floor and forms into small balls. If someone inquires why the mercury is in small balls, it may be helpful to explain that mercury always forms into small spheres in such circumstances. (i.e., no one shaped the mercury into spheres; it is a natural property of mercury to arrange itself in that manner.) One may then want further explanation of mercury and its properties and why it behaves so differently from water; but while the explanation already given may be rather superficial, it is not a question-begging explanation. Explaining some specific event by subsuming it under a general principle is a legitimate and common explanatory device, both in science and elsewhere. It is a legitimate explanatory practice when there is agreement concerning the specific event and we simply want explanation of why it happened. But if the event or case is a controversial one, and there is disagreement about the particular event (Was Flight KAL 007 spying or was it careless? Should students at Home State University be required to take a critical thinking course? Is Hinckley’s behavior excusable on grounds of insanity, or is it not?), then trying to settle the controversy by appeal to a general principle that includes the controversial case commits the fallacy of begging the question.

Begging the question can be even more indirect, and even more insidious. Sometimes circular arguments travel such a convoluted circular path that it is difficult to detect exactly where the question begging occurs. Consider a standard, and relatively easy, example:

God exists. We know that to be true, since the Bible plainly tells us that God exists. And we know that what the Bible tells us is true, since the Bible is the word of God.

In this argument an essential premise (the Bible is the word of God) assumes the truth of the conclusion (for the Bible to be the word of God, then God must first exist).

But circular begging the question can be more subtle. There has long been controversy concerning whether the U.S. Constitution should be amended to allow residents of the District of Columbia representation in the Congress and Senate. One person (in a letter to the editor of the Greensboro Daily News) gave this argument against such a constitutional amendment:

This past August the U.S. Senate and House of Representatives voted on and approved the District of Columbia representation amendment [to the U.S. Constitution]. The amendment is now before the states and must obtain ratification by three-quarters of them by 1985 or it will die.

Basically this amendment would give the 700,000 citizens of the District of Columbia voting delegates in either house of Congress. In the words of the amendment, the District should have representation “as though it were a state.”

Article I of the Constitution directs the members of the House shall be chosen from the “several states,” and that two senators shall be chosen from “each state.” That should settle the issue once and for all. The District of Columbia is not a state and giving it representation as such is an outright and flagrant violation of the Constitution.

In order to see the question-begging trick, the first step—as always in critically examining arguments—is to determine exactly what the conclusion is. (Without knowing what the conclusion is, we can’t tell whether it is being slipped in among the premises; we won’t even know what to look for.) The conclusion here is clear enough: The amendment to the Constitution giving D.C. representation should not be passed. But why not? Because the Constitution clearly specifies that members of the House and Senate should be chosen from the states, and the District of Columbia is not a state; therefore, giving D.C. representation in the House and Senate “is an outright and flagrant violation of the Constitution.” That’s true: Giving the District of Columbia representation in the House and Senate would be a violation of the Constitution as it stands, and so would require an amendment to the Constitution. But the question at issue is whether there should be such an amendment to change the Constitution. By assuming that “violating” (changing) the Constitution must be wrong, the argument moves in a circle and begs the question. Obviously if “violating”/changing the Constitution is wrong, then it is wrong to change the Constitution, and the D.C. representation amendment must be wrong; but the question is whether this amendment is a good one, whether it is right in this instance to amend the Constitution.

The circular form of the begging-the-question fallacy can be found in recent discussions of child abuse testimony. Any case that depends heavily on the testimony of children poses serious difficulties. There are dangers that children may have trouble distinguishing fantasy from reality, but there are also dangers that children may be confused and frightened—or even intentionally intimidated—during their testimony. Such problems are exacerbated in child abuse cases; there is a danger that the child abuser may have threatened or physically abused the child in such a way that the child is afraid to testify against the abuser, particularly when the child abuser is present in the courtroom. On the other hand, one of our most important safeguards against false accusations is that the accused has the right to face his or her accuser. If you were accused of a crime you would certainly want to know who was accusing you and hear your accuser’s testimony and face that person in court. So there is a genuine problem here, and it is not at all obvious how it should be resolved. But one question-begging argument is sometimes given in favor of allowing children in child abuse cases to give their testimony without the defendant being present. That argument goes something like this:

It is doubly wrong to allow defendants in child abuse cases to be present when the abused children give their testimony. These child abusers often use terrible threats to keep the abused children from telling anyone about the abuse. To allow such vicious people to continue their intimidation of the children by being present when the children testify—and thus to perhaps avoid conviction—compounds their abuse of the children and extends that abuse into the courtroom.

There may or may not be good arguments against allowing the defendant to be present during all testimony, but certainly the above argument is fallacious. To assume that the defendant’s presence would terrify the testifying children because of the threats the defendant has made against the children is to beg the essential question: Did the defendant make such threats? Is the defendant guilty as charged? While it is important to make the

The trial of Carlyle W. Harris, charged with murdering his wife by administering morphine to her, was a famous trial of several decades ago. In his closing address to the jury, the prosecuting attorney included these remarks:

It is pretended by Mr. Taylor [defense attorney], in solemn mockery of your intelligence and manhood, that you should acquit his client because throughout all the nine months that Harris has been in prison and throughout this long trial he has always maintained his self-respect, and could look into the eyes of each juryman with a confidence and assurance which, in Mr. Taylor’s judgment, could only arise from a sustaining knowledge of innocence in his heart ... . Is he innocent because he can look a jury in the face? Why, could any man whose conscience wasn’t fairly choked in his breast look any honest man in the face after there had been proved against him what has been proven in this case against Harris? This is not innocence, to my mind; it shows an utter lack of conscience and of feeling.2

Let us grant the weakness of the defense’s claim that Harris’s ability to look the jurors in the face is evidence of his innocence. However, when the prosecution maintains that Harris’s ability to look the jurors in the face proves that the defendant is utterly without conscience and feeling—how else could a guilty man look jurors squarely in the face—that is a case of arguing in a circle, of begging the question at issue: Is Harris guilty?

situation as comfortable and secure as possible for children giving testimony, it should also be remembered that the testimony of children can be terribly and tragically wrong. The Salem witchcraft trials, in which several unfortunate and innocent women were condemned to death on the basis of the testimony of a group of children, should suffice to remind us that we must also safeguard the rights of defendants. And in particular, we must safeguard the presumption of innocence—an essential condition for both fair trials and accurate thinking.

Begging the question is always a fallacy; but not every case in which the conclusion is repeated is a case of fallacious begging the question. In a lengthy argument, it may be helpful to have the conclusion repeated several times—otherwise, those listening to the argument may forget exactly what the argument is supposed to be proving. It is often useful to start an argument by stating what you intend to prove, then giving your premises in support of your conclusion, and finally repeating the conclusion that your argument proves. That is not begging the question. Begging the question occurs when the conclusion is used as a premise. Simply repeating the conclusion is not using it as a premise. (Of course if there is nothing to the “argument” other than repetitions of the conclusion, that would indeed be begging the question, but don’t make false accusations of begging the question merely because the conclusion is stated more than once.)

In sum, watch out for the fallacy of begging the question, in all its treacherous guises. But don’t start seeing the question-begging fallacies when they aren’t really there. It’s a fallacy to beg the question; it is not a fallacy to restate the conclusion.

There are two important close relations of the begging-the-question fallacy, and they go by special names: self-sealing arguments and complex questions. Self-sealing arguments are perhaps the most subtle and misleading of the question-begging arguments. The self-sealing fallacy might also be called the no-true-Scotsman fallacy, in honor of the following delightful example of that fallacy (developed by Antony Flew3): A particularly proud Scotsman picks up the morning paper and reads that an Englishman has committed a horrendous sex crime. “No Scot would do such a thing!” our Scotsman exclaims. But the very next day he reads an account of the even more vicious and scandalous sex crime committed by Mr. Angus MacSporran of Aberdeen, Scotland. Confronted with such a clear counterexample to his claim that no Scot would commit such a crime, our proud Scotsman ought to withdraw his claim, or at least retreat to the milder claim that very few Scotsmen would commit such a crime. But instead he “saves” his original claim: “No true Scotsman would do such a thing.”

That manner of “saving” a claim turns it into a self-sealing fallacy. What begins as a substantive empirical claim is turned into an empty formula that is “true by definition.” The original claim was a strong, although doubtful, one: No individual of Scottish descent would ever commit a vicious sex crime. But when the claim is modified in order to preserve it, what sort of claim is it? What is a “true Scotsman”? Merely a person of Scottish descent who would never commit a vicious sex crime. So the claim ends up as: No person of Scottish descent who would never commit a vicious sex crime will ever commit a vicious sex crime. That’s true, but it doesn’t tell us much. In fact, it doesn’t tell us anything at all. Instead, it merely proposes a very peculiar and not very useful definition for “true Scotsman.” The truth of the claim is “sealed in” by making the claim empty, vacuous, a mere truth-by-arbitrary-definition.

The self-sealing fallacy is frequently used to make claims that sound important but are in fact vacuous. You have perhaps found yourself caught in the recurring late-night dorm discussion of whether people are selfish or generous. Some provocative individual asserts that everyone acts always and only for selfish purposes. And immediately that claim is contested. “My mother,” you say, “certainly is not always motivated by selfishness. I well remember the nights she sacrificed her own rest, staying by my bedside holding a cool cloth to my fevered brow for hours on end. And even as we speak, she is sending me all her extra money in order to pay for my college books.” But your opponent refuses to budge: “She is still motivated by selfishness; it’s just that she takes more pleasure in buying you books than in buying for herself, and more comfort in mopping your fevered brow than in sleeping. She is selfishly doing what she wants to do.” Unless we look closely, this will appear a strong argument. How can we answer? No matter what examples we propose—a self-sacrificing mother, a loyal and selfless friend—our opponent can handle them: The individual is pursuing his or her own interests, is doing what brings the most pleasure. But what appears a strong position is actually an empty one. This person is using a peculiar—and self-sealing—definition of “selfishness.” In the ordinary meaning of “selfish,” we do not count an individual as selfish who does good for someone simply out of a generous desire to help that person. That is precisely what we regard as unselfish behavior. If “selfish” behavior is redefined as “all behavior done from any motive whatsoever,” then certainly all behavior will be “selfish” behavior, in an empty sense. Indeed, under such a definition of “selfish” it would be impossible by definition to do any act that was not selfish. But that shows the emptiness of the new definition. The claim that all human behavior is selfishly motivated is supposed to be a real claim about humans and their behavior. But under the self-sealing definition of “selfish,” the claim has no empirical content, it tells us nothing about humans or about human motivation; instead, “All people are ‘selfish’” expresses an empty verbal formula that is true by definition.

This last point shows the way to answer self-sealing arguments. You cannot answer such an argument by attempting to find a counterexample, as you could if the claim being made were a genuine empirical claim about the world. After all, the tricky thing about self-sealing arguments is that they prevent anything from counting as a counterexample. Instead, when refuting a self-sealing argument you must make its actual structure clear; that is, you must show that its conclusion has been transformed from a factual assertion to a mere verbal formula. And the best way of showing that is to challenge the person offering a self-sealing argument to state what he or she would be willing to count as a genuine counterexample to the claim. If nothing can possibly count against it, then it is true by definition rather than being a claim about the observable, testable, empirical world we live in.

In fact, that is a good way to test some of your own beliefs: Possibly some beliefs you thought were about the real world have degenerated into comforting verbal vacuities. That can happen quite easily to our particularly cherished beliefs. I strongly believe that anyone who commits a murder must be psychologically unbalanced, and that therefore we should think more about how to reform or cure such murderers (while protecting society against such criminally sick individuals), and reject notions of vengeance against murderers. I can cite plenty of examples of mentally ill murderers: Loeb and Leopold, the wealthy Chicago youths who killed a playmate as a sort of bizarre experiment; David Berkowitz, the “son of Sam” murderer, who received messages from his dog; John Hinckley, who lived in a fantasy world inhabited by teenage movie stars; Albert DeSalvo, “the Boston Strangler,” who was driven mad by a sequence of events (including his daughter’s crippling illnesses). But suppose my claim is challenged: Here is a murderer whom all competent psychiatrists and psychologists believe to be sane, whose psychological test results show up normal, who does not exhibit any bizarre behavior patterns, who is not delusional, who has no neurological damage to his brain, has no history of mental illness; this murderer robs a bank and coolly kills a witness to prevent identification (and thus improve his chances of not being captured). Is that murderer a counterexample to my claim? If I admit that he is a sane murderer, then I must give up my claim that all murderers are insane (“Very well; I should say instead that most, but perhaps not all, murderers are insane”) but at least the claim will remain a significant factual claim about the mental condition of murderers. But suppose that instead I say (as I am very tempted to say): “No, I still believe he is insane; for anyone who murders another human being must be insane.” Then my claim is no longer about the mental condition of murderers; rather, it is an arbitrary verbal stipulation: As I am now using the terms, to murder someone is by definition to be insane. But then to say that a murderer is insane is no longer to say something significant: It is merely to say that someone who is insane is insane, because my special self-sealing meaning of “murderer” includes insanity as part of its meaning. If I cannot state what I would count as a sane murderer, then that is a sign that what was originally a significant empirical claim has become an empty verbalism.

Self-sealing arguments are also a means of concealing our prejudices and stereotypes from the light of critical examination. Bill is driving Jill to the airport, when a car pulls in front of them and forces Bill to slam on his brakes. The careless driver is a woman, and Bill exclaims vehemently: “Damn woman drivers. All women are awful drivers and shouldn’t be allowed on the road.” As they drive on, Bill enlarges on the subject: “See that guy driving ahead of us? Now he’s a good driver. He stays in his lane, keeps a steady speed, doesn’t talk on his cell phone or send text messages, signals when he’s about to change lanes. He’s a good driver, not like those lousy woman drivers.” At the next traffic light, however, Bill catches up to the car ahead, and as he pulls alongside, the good driver turns out to be a woman! Jill is delighted: “You see, women aren’t all bad drivers; you’re just prejudiced. That woman is certainly a good driver, as you yourself pointed out.” However, Bill can hang on to his prejudices, by swift use of a self-sealing argument: “Well, she’s not really a woman driver; she drives like a man.” “Woman driver” is now being used by Bill in a self-sealing manner: It means a driver who is a woman and is a lousy driver. So of course any “woman driver” must be, by self-sealing definition, a lousy driver. The problem is that once this challenge to his prejudices is past, Bill is likely to switch back to the larger original claim that all women are bad drivers. Thus he can maintain his prejudice in the face of almost any contrary evidence.

“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.”

It was, no doubt; only Alice did not like to be told so. “It’s really dreadful,” she muttered to herself, “the way all the creatures argue. It’s enough to drive one crazy!”4

Another fallacy from the begging-the-question family is complex question. “Are you still drinking too heavily?” Whether you answer yes or no, you seem to be admitting that at one time you did drink excessively. And that is how complex questions beg the question: They embed an assumption within a question. By answering the question, you seem to grant the embedded assumption.

Complex questions are standard stuff for Hollywood depictions of trials. The shrewd district attorney asks the villainous defendant: “Why did you leave the diamond necklace behind when you robbed Lady Bigbucks’s country estate?” “I didn’t see a diamond necklace,” replies the hapless defendant, thereby admitting that he indeed is the thief. But such loaded questions are not very likely in actual courtrooms. The defense attorney would certainly, and rightly, object before the witness had a chance to answer, and the judge would not only uphold the objection but would probably rebuke the district attorney for asking such a question. If complex questions occur at all in the courtroom, they are likely to be confined to the summation speeches made by the prosecution and defense just before the case goes to the jury, for lawyers are generally granted a good deal of latitude in their summations. At that point the defense attorney may rhetorically ask, “Do you really want to convict this person on such a flimsy patchwork of evidence?” And the prosecuting attorney may ask, “Do you want to have it on your conscience that you returned a dangerous criminal to the streets?”

Incidentally, complex questions are not the same as leading questions. A leading question leads in a specific direction, but allows the witness to decline going in the suggested direction: “You hate the defendant, don’t you?” That is a leading question, but the witness can simply answer: “No, I don’t.” A complex question commits the witness to an underlying assumption no matter how the question is answered: “Was it the investment you lost in the defendant’s business that caused you to hate him?” Whether the witness answers yes or no, the assumption remains that the witness hates the defendant; the only question left is why the witness hates him (was it because of a failed investment or is there some other reason?).

Complex questions may be unusual in court, but they are common in everyday life. When they escape detection, they insidiously control thought. For example, “What type of therapy is most effective in treating homosexuality?” “How do we gain our knowledge of right and wrong?” “What is the purpose of human existence?”

All of the above questions contain at least one unstated but controversial assumption. Failure to recognize those assumptions will fix the framework of discussion in a way that begs important questions. The most important issues may be settled before you even realize the discussion has started. “What is the most effective treatment for homosexuality?” If we accept that statement of the question, then the whole discussion will presuppose some very dubious assumptions: First, that homosexuality is a disease that should be treated, and second, that homosexuality is bad (like disease) and should be eliminated. Revealing those concealed assumptions may make the question irrelevant (we shall not ask how to treat homosexuality if it is neither a disease nor undesirable).

So be wary of complex (and loaded) questions: They may control discussion and constrict the range of possibilities considered. Don’t be in such haste to get to the answer that you overlook dangerous assumptions concealed in the question. In biased surveys, complex questions are often called loaded questions. They are a common trick in fake or manipulated “surveys” that are designed to give whatever result the surveyor desires, and which can then be publicized as “genuine public opinion.” For example, “Do you support government ‘make-work’ programs that will provide temporary jobs while greatly increasing the budget deficit?” is likely to yield a very different survey result than “Do you favor job programs that will put unemployed American workers to work on projects that will make our highways and bridges safer and better?” And, “Do you believe that every child in America should have good health care?” will get a different result from “Do you want the federal government interfering in your child’s health care?”

On the other hand, not all assumptions—and not all questions that contain assumptions—are question begging. A complex question is being used by the automobile salesperson who smoothly slides into the question of how you intend to pay for the car—“Shall we put this on our special 60-month super-low payment plan, or would you prefer to pay it off in only 3 years?”—when you have not quite decided whether you want to buy the car at all. But if it has already been settled that you are buying the car, then the same question may not beg the question. In a breaking or entering case, the defendant denies ever having been in the building he is accused of entering; the prosecuting attorney then asks him: “What were you doing inside the building?” That is a complex question that includes the assumption that the defendant was inside the building—and whether the defendant was inside the building is a disputed question that should not be begged by embedding it in a complex question. But suppose that in a different breaking or entering case, the defendant admits that she was inside the building (but claims she had the owner’s permission). In that case, if the prosecuting attorney asks the same question (“What were you doing inside the building?”), it will not be a complex question; for in this case, whether the defendant was in the building is not in dispute. The moral of the story is this: A question that contains an assumption is not always question begging. If the assumption is not controversial—but is instead an assumption that all the parties to the discussion recognize and accept—then no questions are begged by making the assumption. (Of course the assumption may be false, but the assumption does not beg the question.) The complex question fallacy occurs only when the assumption being made is a controversial one.

Subtle complex questions sometimes do occur during cross-examination. (They must be so subtle that the opposing attorney does not catch them.) The following is a particularly good—at least a particularly tricky—example. Seymour Wishman was the defense attorney for Johnny Sayres, charged with murdering a man named Leander. The defense would claim that Johnny acted in self-defense. Wishman describes what happened:

The first witness the D.A. called was a fat lady who said she had seen Johnny and the victim arguing. When their words had gotten “really angry,” she testified, she had sensed danger and jumped over the bar.

“I saw him take out a gun,” she said, pointing at Johnny, who was sitting next to me at counsel table, “and I ducked down behind the bar. Then I heard these shots, four or five of them. Some people screamed, ‘He shot Leander! He done shot Leander!’ When I looked up, three people was standing over him. And the man was gone.” ...

I began my cross-examination by asking this woman how much she weighed. When she said “two hundred pounds, more or less,” I had her step down from the witness stand and walk to the edge of the jury box. I stood next to her, in front of the waist-high wooden wall separating us from the jury... .

“How high was the bar you say you jumped over? Hold up your hand to show the height of the bar in comparison to the jury box.”

The woman looked at me suspiciously, and then held up her hand.

“Let the record reflect,” I said, “that the witness is indicating a height of approximately four feet. Now how much do you really weigh? Something closer to two hundred fifty than to two hundred pounds, wouldn’t you say?” Actually, she looked closer to a thousand pounds to me and probably to the jury.

“Well, maybe that’s a fact.”

“And you’re telling us you jumped over this bar?”

“That’s a fact.”

“You must have been mighty scared.”

“Ain’t that the truth.”

“They must have been arguing pretty bad, the two of them, for you to have been that scared.”

“And how! I ain’t ordinarily what you’d call no high jumper.”

The jury and I laughed.

“So I guess being so scared, you didn’t see, with your jumping over this bar and hiding behind it and all, whether Leander reached for his gun before Johnny reached for his?”

“That’s right.”

Before reading on, examine the above cross-examination and try to pick out exactly where the complex question occurs, and what important assumption is hidden in that complex question. (That is not an easy task, and it is not surprising that the district attorney failed to notice the use of the complex question.)

The defense attorney continues the story:

I ended my questioning there, leaving the woman standing in front of the jury. There hadn’t been any evidence that Leander had a gun, and the witness’s answer let stand the existence of the gun assumed in my question. The prosecutor should have objected, but he didn’t.5