My colleagues will not like my saying this, but philosophers are strange creatures. Some of the things we do -- such as answer questions with questions -- puzzle or antagonize people. For example, if someone asks a philosopher whether taxation is theft, the likely response is, "What do you mean by 'theft'?" But the philosophical aim is not to puzzle, much less to antagonize. It is to go beyond platitudes - the stuff of too much political discourse - and get at the underlying logical structure of belief.

Some beliefs fit together, meaning that they can all be true. Others do not. The only leverage a philosopher has is the principle of noncontradiction, which says, simply, that no proposition can be both true and not true. This trite-looking claim is an amazingly powerful tool of analysis and criticism. Why? Because most people care about the truth of their beliefs. If you are one of these people, then you should be grateful to me, rather than annoyed or resentful, if I point out that you cannot believe both P and Q. I am helping you formulate a consistent, and therefore a coherent, body of beliefs. I am helping you avoid what Leon Festinger called cognitive dissonance.

Let's analyze the question I posed. Is taxation theft? Theft is both an everyday concept and a technical legal concept (although lawyers often use the term "larceny" instead). It is a mistake to assume that a term such as "theft" has the same meaning in different contexts. It may or may not. One has to look and see. (This is one of the lessons of Wittgenstein.)

According to the Oxford American Dictionary and Language Guide (1999), which, like any dictionary, purports to state the everyday meanings of words, "theft" means "the act or an instance of stealing." Alas, this helps us only if the word "stealing" is better understood than "theft," which is unlikely. So what is stealing? To steal is to "take (another person's property) illegally" or to "take (property, etc.) without right or permission, esp. in secret with the intention of not returning it."

As these definitions show, the essence of theft - that which makes it what it is and not some other thing, such as burglary, robbery, or borrowing - is the taking of someone's property without right or permission. If I ask for your old lawnmower and you give it to me, there is no theft. But if I ask and you say "No," then I commit theft if I take it anyway. Property is another concept that needs analysis, but for our purposes here we can think of it as that to which one is entitled by the prevailing rules, however unjust one thinks those rules are. So the question whether taxation is theft reduces to this question: "Is taxation the taking of someone's property without right or permission?"

Clearly, many or most people who pay taxes do not permit it. That is to say, the payment is coerced rather than voluntary. Coercion, which is itself a fascinating concept, is, to put it crudely: the imposition of a choice by one person (the "coercer") on another (the "coercee") with the intention of inducing that other to choose one way rather than another.

As the philosopher Onora O'Neill has shown in "Which Are the Offers You Can't Refuse?" [1991], coercion is an acquired -- and in some cases a highly developed -- skill. The coercer must know, among other things, the value structure of the coercee. Most people value their lives more than their cash or credit cards, which is why robbers feel confident in the success of "Your money or your life." If you are suicidal, however, my imposed choice may not have its desired effect, in which case I have bungled the attempt. Another thing coercers need is credibility. If your threat to kill me is incredible, because, for example, you are brandishing an obviously nonfunctional gun, you will not succeed in appropriating my wallet.

Governmental taxing authorities are coercers, and their threats - as anyone who has been audited knows - are credible. They say, in effect, "Your money or your liberty." Most people value their liberty more than their property, so they grudgingly comply. It does begin to appear as though taxation is theft (or, more precisely, robbery, which is a special case of theft).

But perhaps we have gone too fast. We said that taxation is the taking of someone's property without right or permission. The word "right" is notoriously ambiguous. We say both that so-and-so has a right to (do) X and that X is the right thing to do. Logically, these are distinct. I can have a right to do X (spank my children, for example, or cut the trees on my land) even though it is not right, all things considered, for me to do so. In this context, the question is whether taxation violates people's rights. If it does, then it is a straightforward case of theft. If it doesn't, then it isn't.

We have reached the issue that divides libertarians (classical liberals) and socialists or modern liberals. (I ignore the difference between socialists and modern liberals, since it doesn't affect my analysis. I use the term "socialist" to cover both camps.) Libertarians believe that people are entitled (i.e., have a right) to the wealth or resources they accumulate, provided that they violate nobody else's rights in acquiring them. What I would like to explore in the remainder of this essay is a familiar socialist argument in favor of taxation. It is not the only socialist argument, but it is a common one. Nor is it made only by the unreflective. I see it made by bright, thoughtful, even philosophically trained people. If it is a bad argument, as I think it is, then the socialist will have to come up with something better.

The argument begins with the observation (assertion) that some or much of the wealth individuals possess is undeserved. It might be said, for example, that the Kennedy children (and grandchildren) do not deserve their wealth. After all, it was given to them by their parents (or grandparents). They did not earn it. Some people, such as the Kennedys, are born to privilege; others, through no fault of their own, are born into poverty. Some people are born with disabilities; others are born healthy. As if this weren't bad enough, privileges and advantages tend to snowball. The rich, as they say, get richer; the poor get poorer. There appear, in short, to be both natural and social lotteries at work. Some of us are lucky; others are unlucky. Fortune and misfortune are rife.

This socialist argument introduces a new and interesting concept: desert. What does it mean to say that S deserves X (or doesn't)? Various theories of desert have been propounded and defended. But let us focus on the contrast between desert and entitlement. It might be thought that they are the same, that "S deserves X" and "S is entitled to X" have the same meaning. But careful attention to how we speak -- and think -- reveals that this is erroneous.

Suppose I have trained hard for a particular footrace. I am not a natural runner, but I have worked hard, made sacrifices, and overcome obstacles. Suppose my competitor is naturally gifted. He trains irregularly, eats poorly, and makes a halfhearted effort. Suppose that, despite my best efforts, my competitor crosses the finish line first (and that all the rules have been complied with). Is he entitled to the trophy? The answer seems clearly "Yes." Does he deserve it? Here the answer seems just as clearly "No." Desert has to do with things like effort, worth, and merit, whereas entitlement has to do with compliance with known rules, institutions, or practices. I deserve the trophy (arguably) but am not entitled to it. My competitor is entitled to the trophy but does not (arguably) deserve it. As my teacher, Joel Feinberg, pointed out four decades ago in "Justice and Personal Desert" [1963], we should not identify desert with justice while relegating entitlement to some other, inferior category. Desert and entitlement are two aspects of justice. They can -- and do -- compete with one another in cases such as the one I described.

The difference between desert and entitlement is not a philosopher's invention. It is reflected in how we speak and think. Those who are deserving but not entitled are said to have won a "moral victory," defined as "a triumph, although nothing concrete is obtained by it" (Oxford American Dictionary [1980], s.v. "moral"). Those who are entitled but not deserving are said to have won a "hollow victory," defined as one that is "empty, worthless" (ibid., s.v. "hollow"). Moral victories are not real victories, but they feel like it. Hollow victories are real victories, but, being empty, they don't feel like it. There is also such a thing as a "Pyrrhic victory," which is "a victory gained at too great a cost, like that of Pyrrhus, king of Epirus, over the Romans in 279 B.C." (ibid., s.v. "Pyrrhic victory"). This would occur, for example, if you earned a good grade as a result of cheating. The cheating stains your moral character, which, to my way of thinking, is far too great a cost to pay.

Now that we have clarified the relevant concepts to some degree, let us return to the socialist's argument. We saw that the socialist begins with the premise that some or much of the wealth individuals possess is undeserved. Whether this is so depends, we now see, on one's criterion (theory) of desert. I want to bypass this important issue and focus on another, so I will do something else that puzzles people, namely, assume it for the sake of argument. Let us assume, in fact, that nobody deserves anything. (Compare Judith Jarvis Thomson's famous assumption that all fetuses, at whatever stage of development, are persons [like you and me] with a right to life. She does not believe this, and said so, but she assumed it in order to see what follows (logically) from it. She argued, ingeniously, that it does not follow from the fact that fetuses are persons with a right to life that it is wrong to abort them. The value of this argumentative strategy is twofold: First, it focuses attention on previously unaddressed issues -- in this case, something besides the moral status of the fetus-- and second, it forces the opponent of abortion to reveal and evaluate hidden assumptions. This is how philosophical progress is made, and why Thomson is considered a great philosopher.)

What follows from the assumption that nobody deserves anything? To make the discussion concrete, suppose that I don't deserve my wealth. First, it does not follow that anyone else, much less people generally (or the state on behalf of people generally), deserves my wealth. This would be a blatant contradiction. For if nobody deserves anything, then people who are poor through no fault of their own do not deserve anything - including my wealth! Second, it does not follow that I'm not entitled to my wealth. We saw that desert and entitlement differ. There are cases in which one is entitled to X even though one doesn't deserve X. So the mere fact that I don't deserve my wealth doesn't entail that I'm not entitled to it. And if I am entitled to it, then taking it without my permission is theft.

But suppose (again, for the sake of argument) that I'm not entitled to my wealth (even if, as John Rawls says, I - like everyone else - am entitled to my "natural assets"). It doesn't follow that anyone else is entitled to it! As every law student knows, theft from a thief (i.e., from someone in wrongful possession of property) is still theft.

Finally, it doesn't follow from the fact that I don't deserve my wealth that it is morally permissible for someone (including the state) to coerce me into disgorging it. Nor does it follow that it is morally permissible, all things considered, to take it from me without my permission. If I give it away, fine. Doing so may even make me a good person. But unless one is a moral imperialist or totalitarian, there is logical space between (1) it being good to do X and either (2) it being obligatory to do X or (3) it being right for the state to coerce the doing of X.

I have obviously not canvassed all of the socialist arguments for taxation. I have addressed only one of them: the argument that, because wealth is (largely) undeserved, the coerced taking of wealth through taxation is not theft. I believe I have shown that the argument fails to accomplish what it purports to accomplish; indeed, that it never even gets started. (In philosophical jargon, it's a "nonstarter.") It may be possible to salvage the argument by adding one or more premises, but I suspect that the added premises will purchase validity (good form) at the expense of truth or acceptability (bad content). If this is correct, then socialists have their work cut out for them. They should not despair, however, at their inability to justify taxation, for if socialists were to pool their "undeserved" wealth, they could probably provide for the needs of those for whom they profess to care. And if they stop coercing people such as me into disgorging their wealth for the benefit of the disadvantaged and unfortunate, they might just increase the incidence of voluntary giving. This, in turn, would have an unintended but salutary effect: It would give the transfer a moral tenor that it presently lacks. Coerced actions, unlike their voluntary counterparts, are morally worthless.

Keith Burgess-Jackson, J.D., Ph.D., Associate Professor of Philosophy, Department of Philosophy and Humanities, The University of Texas at Arlington.

Editor's note: For more on this topic, read about the late Robert Nozick's argument that taxation is on par with forced labor.