John Rawls, who died on November 24, raised modern political philosophy from the pit of Marxist and linguistic analysis and revived it as a serious subject for citizens of the real world. He believed that answering age-old practical questions about liberty and justice was the proper work of political philosophers. His own answers were ingenious, deeply thought through, and provocative. They aroused the passionate interest of other philosophers and drew scholars from other disciplines, including economics, into the discussion.

I think that Rawls lost the debate that ensued, but the intellectual battle was joined on the ground he had marked out, and that was a victory in itself. He was unquestionably one of the most influential thinkers of the 20th century.

Rawls is not exactly beach reading, though honesty compels me to admit that I have in fact been reading him this week under a palm tree by the Indian Ocean. National Journal is not light entertainment, either. Nonetheless, I recommend both: They go together well. The contest between Rawls and his critics should not be fenced off for academics. Anybody who wants to think more clearly about politics and ethics—about, say, the rights and wrongs of progressive taxation, or about how liberal democracies should order their relations with societies that do not share their values—will profit from grappling with this literature. If nothing else, it is an improving experience to see one's own flimsy prejudices give up without a struggle.

Rawls was a radical egalitarian-more radical than he seemed to realize. In A Theory of Justice, his most celebrated book, he famously proposes a way to crystallize our underlying ideas about justice. Imagine that you are asked to choose the basic rules by which society will be run, but without knowing anything about your personal circumstances—how intelligent you are, what talents you have, how rich your parents are, and so on. Rawls argues that from behind this "veil of ignorance" you would choose two key principles.

First is the liberty principle: You would insist on certain basic political and civil liberties to protect your autonomy as a person. Second is the so-called difference principle: You would want to live in a society that made improving the lot of the poor its highest priority. Why? Because the person at the bottom might turn out to be you. More precisely, Rawls argues that people stripped of self-knowledge would choose a system that, so long as it did not infringe basic political liberties, would strive to reduce inequality until cutting it any further would do such harm to the economy that the position of the poorest person would actually be worsened.

Rawls was often described as "left of center." That is putting it mildly. His idea of the just society goes far beyond what any mainstream American liberal would ever dream of proposing, though you would not think so from some of the articles that have appeared in the past two weeks. These articles praise the Harvard University professor and claim support in his writings for various timid liberal agendas. The Rawlsian idea of fairness actually goes far beyond what any European socialist party, to say nothing of the Chinese Communist Party, would nowadays propose.

Economies with elaborate and expensive welfare states and high taxes on the rich—Germany, say, or Sweden—still have very much further to go before they achieve justice, in Rawls's view. Taxes so high that they shrink the economy are fine, so far as the difference principle is concerned. Keep on raising them, channeling the proceeds to the least well-off. It is only when the aggregate loss from high taxation is so vast that redistribution cannot protect the poorest from sharing in the pain that the effort to reallocate income stops.

Wouldn't the liberty principle kick in long before this point was reached? Apparently not. Rawls excluded economic liberties such as property rights from its scope—otherwise, the liberty principle would, from the outset, have flatly contradicted the difference principle that calls for such far-reaching redistribution.

The assault on Rawls was led by Robert Nozick, another giant of 20th-century philosophy, who also died earlier this year. Nozick's theory put rights, including property rights, at the center. In a column in February, I said I thought Nozick made mincemeat of Rawls's theory. But Nozick ended up with a theory of his own whose conclusions were about as bizarre and extreme as Rawls's. Justice, according to Nozick, demands a minimal state. No safety nets at all—extremes of wealth and poverty that would make Saudi Arabia blush. So far as government is concerned, providing physical security is about it. No democracy is ever going to vote for a Nozickian society, any more than it will for a Rawlsian one.

So we have one theory proposing a brand of egalitarianism too extreme for any but the Far Left, and another defending a brand of libertarianism so pure it would give the Cato Institute nightmares. Each claims to be grounded in first principles. Each is irreconcilably opposed to the other. For all the use this is going to be, you might think, Rawls might as well have left the whole issue undisturbed.

Not at all. It is educational to see "liberal" and "conservative" positions located so carefully (for once) on their respective premises, then remorselessly developed to their unacceptable conclusions. If you regard economic freedoms as second-class liberties, and if you set out to judge fairness exclusively by reference to the incomes of the poorest—a position many liberals would be proud to call their own—then Rawls shows you where you end up: with a society organized somewhat along Cuban lines. On the other hand, if you reject every claim that considerations of fairness or social solidarity might have over individuals' rights, including property rights, as many conservatives apparently do, then logic drives you just as irresistibly toward a dystopian laissez-faire.

In politics, neither side accepts the implications of its own beliefs. This is a start. It suggests that their disagreement is partly a mere refusal to communicate. There might be more scope for consensus than either side wants to admit. This unspoken consensus, of course, is what Western democracy, despite the surface turbulence, manages to express.

Unfortunately, in nailing this agreement down and trying to get the underlying thinking straight, it won't work just to take a rough average of Nozick and Rawls. Nozick's system refuses to compromise. Rawls had a more elastic approach—above all in looking for an answer that balanced liberty and fairness, rather than insisting that one of these timeless human preoccupations be ignored altogether. Most can agree that he was right about that, even if A Theory of Justice then got the balance weirdly wrong. So I am waiting for a follower of Rawls, rather than a libertarian descended from Nozick, to come up with a believable theory of justice.

If any such thinkers are reading this, could I make a request? Please take economics more seriously. Any economist could have told Rawls that in deriving his "maximize the minimum" criterion for fairness, he assumed that people are far more risk-averse than they actually are. Which of these would a rational self-interested person, weighing things up behind the veil of ignorance, prefer: to be assigned a random place in the United States, or a place in some other country with the same political freedoms but where everyone receives an income equal to that of the poorest American? The latter, according to Rawls, because inequality is less and the position of the poorest is no worse. I don't think so.

Economics is helpful in another way. To reflect intelligently on what is just, you must also think about what is feasible. The first is no use without the second. Economics is especially interested in feasibility.

Social reformers such as Rawls are in a tradition that emphasizes the best over the possible. Conservatives do the opposite. They are suspicious of departing from the status quo. At least they know that that works, after a fashion—and the history of communism shows what happens when utopians take charge. Rawls emphasized liberty in his theory to check that dangerous tendency. But it was too narrow an idea of liberty, and he did not emphasize it enough. As a result, his system is forced to rely on an other-worldly degree of social solidarity.

Progressives tend to see solidarity, political action, and moral vision as the basis for all social progress. They are wrong. A social contract for modern industrial countries must recognize, as economics does, that cooperation among strangers based on mutual economic advantage is the necessary platform for any flourishing society. Social justice has to build on that economic reality and accept the constraints it implies. Give Adam Smith a seat at this table.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.