Charles Murray is the author of two books that have become touchstones (or lightning rods) of a sort in American political debate. Losing Ground: American Social Policy 1950-1980 argued that postwar welfare programs have actually made people worse off; The Bell Curve: Intelligence and Class Structure in American Life (written with Richard J. Herrnstein) argued that antidiscrimination policies like affirmative action are futile, since black Americans have lower intelligence than white Americans. These were, in form at least, empirical arguments. What It Means to Be a Libertarian is an attempt to sketch out a complementary political philosophy.

This was not an absolute necessity. A person might believe that welfare policies have been counterproductive, or that affirmative action programs are a wasted effort, without concluding that government is pointless as such. But that is what Murray claims he does believe. “It is not that government intervention hasn’t done as much good as people think,” he says, “but that it has not made any perceptible change in the outcomes of life that matter. It is said that roosters think the sun rises because they crow. Politicians are much the same.” He calls the philosophy that he derives from this conviction “libertarianism.”

As Murray presents it, libertarianism is the belief that the fewer the number of laws there are to abide by, the more law-abiding people will become. This paradox is founded on a common-sense notion, which is that if you understand the risks of following a certain course of conduct, you will conform voluntarily to the most responsible pattern of behavior possible, and you will seek to avoid contact with people whose patterns of behavior you consider dangerous or distasteful. People, in this view, are imagined as fully competent calculators of costs and benefits, and society is imagined as a self-regulating organism which does not require a government to dictate its rules or to correct its tendencies. Requiring individuals to bear the full risks of their actions is thought to instill virtue; permitting society to solve its problems by the give-and-take of the market rather than by legislative fiat is thought to promote efficiency.

These are by no means uncommon or eccentric views, although, as Murray admits, not all the people who share them would choose to call themselves libertarians. Murray’s philosophy, as he spells it out, is really a hybrid of several political theories, including libertarianism (in the emphasis on the decriminalization of most types of personal behavior, such as prostitution and drug use), nineteenth-century, or “classical,” liberalism (in the emphasis on the efficiency of free markets), republicanism (in the emphasis on the inculcation of civic virtue), and communitarianism (in the emphasis on the moral self-determination of local communities).

These various theories are not ordinarily run together in this way. Libertarians and classical liberals tend to celebrate the autonomous individual, who is a figure republicans and communitarians tend to contemplate with loathing. But they all do share something that makes them congenial to Murray’s purposes, and that is a deep distrust of the modern state. The usual objection to them all is that they fail to take account of the scope and complexity of technologically advanced and culturally pluralistic societies like the contemporary United States. The objection is not that liberty, efficiency, virtue, and self-determination are undesirable or obsolete; it’s that in modern societies a strong state is precisely what is required to secure them.

One response to this objection has been: then so much the worse for modern societies. If the scope and complexity of contemporary life require authority to be taken out of the hands of individual citizens and local communities and placed in the hands of a centralized government bureaucracy, let’s cut down on the scope and complexity. If we have to make a trade-off between social progress and civic virtue, or between economic growth and community, or between universal rights and moral values, we should sacrifice some of the former to get more of the latter. This was roughly the view taken by Christopher Lasch, for example, in his last works, particularly The True and Only Heaven (1991), and, more temperately, by Robert Bellah and his co-authors in their influential study of American society Habits of the Heart (1985).

But it is not the view taken by Charles Murray. He thinks that no trade-offs are necessary. He believes that we can have all the small-town virtues—faith, family, voluntarism, moral self-determination—plus all the modern conveniences—big paternalistic corporations, lots of consumer choice, high-speed travel and communications, unregulated freedom of expression—at the same time. He thinks that we can live like moderns and feel like farmers. All we need to do is reduce the size of government to a bare minimum, and every good thing will follow.

How plausible is this proposal? What It Means to Be a Libertarian is composed in a style somewhere between the homiletic and the Op-Ed. It is, in other words, a mostly rhetorical exercise. A good number of its discussions begin with a challenge to disprove a negative (for example: “It is difficult to think of a case of authentic economic coercion in a free economy”) and are clinched by a flat assertion (for example, as proof that laws against racial discrimination are unnecessary: “Deprived of the use of force, human beings tend to cooperate”).


In keeping with this hortatory spirit, the book includes, Murray warns in his introduction, only one graph—one empirical demonstration, which is presented as a kind of synecdoche for what a full-dress analysis of the failure of government action generally would look like. (Readers are encouraged to construct similar graphs dealing with other issues themselves.) The graph is what Murray calls a “trendline.” It charts the decline in the rate of death from traffic accidents between 1925 to 1993, with the purpose of exploding conventional wisdom concerning “the one case in which everyone knows that a government law absolutely, without question, saved thousands of lives: the imposition of the 55-mph speed limit in 1974.” This is not much of a bone, but since it is the only one Murray has thrown, we might as well chew on it a little.

What the graph shows, according to Murray, is that

the period of the greatest change had nothing to do with government regulation. The steepest downward slope occurred in the period 1934-1949, an era when regulation of automobile safety features was nil and highway speed limits, where they existed at all, were high. The only thing that government did was build better highways as traffic increased [oops—one skate just went through the ice]. Meanwhile manufacturers were building safer cars—not because the government said they must but because greater safety generally goes hand in hand with improved technology…. By 1974 no change in automobiles, roads, or speed limits was going to make a huge difference…. As it happened, the passage of the 55-mph speed limit made no visible difference at all.

There are a number of things wrong with this lesson in policy analysis. The first is the validity of the conclusion. Did the passage of the 55-mph speed limit make a difference in the motor vehicle fatality rate or not? A number of factors are involved in determining a change of this kind—not only, as Murray suggests, highway conditions and engineering improvements, but also, as he somehow neglects to mention, the imposition of seat-belt laws, the adoption of passive-restraint systems for children, the installation of air bags, the enforcement of laws against drunk driving, the price of fuel, the relative cost and availability of alternative means of travel, and what might be called attitudinal changes in the culture of driving. The only way we can come close to measuring the impact of a single factor like a reduction in the speed limit is to look at the years immediately before and immediately after the change was introduced.

Having asserted that the 55-mph speed limit made “no visible difference at all,” Murray doesn’t trouble us with the actual numbers. They are as follows. In 1970, 54,600 Americans died in traffic accidents; in 1973, the number was 55,500. In 1974, after the passage of the 55-mph speed limit, traffic deaths fell to 46,400. In 1975, they fell again, to 45,900. In 1976, there were 47,000. Measured as deaths per 100 million vehicle miles, which is the form Murray prefers: in 1970 there were 4.7 deaths per 100 million miles traveled; in 1973 there were 4.1; in 1974, following the passage of the 55-mph law, there were 3.5; in 1975, 3.4; and in 1976, 3.3. These decreases correlate with a decrease in the average speed of cars on interstate highways, which fell from 65 mph, with fifty percent of cars exceeding 65, in 1973, to 57.6 mph, with only 9 percent of cars exceeding 65, in 1974.

It is hard to see how Murray can claim, short of pointing to another cause, that the 55-mph speed limit made “no visible difference” in the number or the rate of traffic deaths. What is notable about the decline in the fatality rate, in fact, is how decisive it was. For the effect of reducing the speed limit to 55 cannot be meaningfully measured against all motor vehicle travel, which is how Murray measures it. The effect is obviously limited to travel on roads where the speed limit was previously higher than 55.

These were, for the most part, interstate highways, which are built, maintained, and regulated by federal and state governments and which are statistically the safest roads in the nation. The least safe, as it happens, are locally built and maintained rural roads. The fatality rate on interstate highways in 1994 was .74 per 100 million vehicle miles—higher (.99) on rural sections, where speed limits have generally been raised, than on urban sections (.58). On noninterstate rural roads, it was 2.66. Murray naturally does not get around to pointing out that by 1994, following aggressive government efforts to require the use of seat belts and air bags, the total number of traffic deaths had dropped to 40,676, or just 1.72 per 100 million miles traveled.


Murray’s whole discussion of this matter starts, of course, from a false premise, which is that Congress voted to reduce the top speed limit in 1974 in an effort to decrease the rate of traffic fatalities. It did not. The purpose of the law was not highway safety; it was energy conservation, a response to the threat of an oil embargo by OPEC. And in this respect the law was also effective (though a steep rise in the price of gasoline no doubt helped). The average number of gallons consumed annually per vehicle fell from 851 in 1973 to 788 in 1974 and 790 in 1975. The 55-mph speed limit was an extremely modest political measure. It added a few minutes to long drives on interstate highways; in return, it helped to save a little oil and, as an unanticipated bonus (and contrary to Murray’s explicit claim), thousands of lives. It was slightly nerve-wracking back in 1974 driving at 55 miles an hour on roads where one was accustomed to doing 75 or 80, but there was also a weird sense of solidarity about the experience, as though people, by driving in what at the time seemed like a farcical slow motion, were pulling together, doing the right thing to meet a national crisis. It was possible to resent the loss of time, but I don’t think very many people experienced it as a loss of freedom.

The complete misrepresentation of the data isn’t what is most interesting about Murray’s discussion of the 55-mph law, though. For he returns to the subject many pages later, where, this time stipulating that “some non-zero number of lives were saved” because of the law, he asks whether its repeal in 1995 was therefore a bad idea. “The key to thinking about the answer,” he suggests, “is this: Even though the speed limit has been raised, you still have the option of driving at 55 miles per hour yourself…. For practical purposes the net effect on your safety of raising the speed limit can be zero if you unilaterally choose to drive at 55 miles per hour. The difference is that now you may also choose to drive faster, getting where you are going sooner and spending your time… doing something more satisfying than sitting in a car” (for example, dying?). And he draws an analogy with seat belts and childproof caps on medicine bottles: in an unregulated society, he says, we would be free to obtain the advantages of safety features like these or not as we chose, which would represent an increase in liberty. “The same logic applies,” he concludes, “to most of the regulations that try to protect us from ourselves.”

The obvious question raised by this is whether, on Murray’s libertarian view, the state should be allowed to set any limit to the speed at which citizens can drive. The logic would appear to be that it should not, since any speed limit is a limitation on personal choice, a way of taking responsibility for the assumption of risk out of the hands of individual drivers. For that matter, would Congress, on Murray’s theory, have had the authority to pass the 55-mph law for its original purpose—to address a national economic problem by restricting people’s “liberty” to drive at 75 mph? Murray does speak of a highly restricted concept of the “public good” as the only valid excuse for government regulation, but if that concept does not extend to the protection of human life, it is hard to see how it can cover a mere dip in the supply of oil.

But of course the whole notion of the individual assumption of risk is virtually meaningless in this case. Most people regard speed limits not as a way of protecting themselves from themselves, but as a way of being protected from other people. How the net effect for you of raising the speed limit can be zero if you don’t choose to drive faster while everyone else does so is mysterious. The only thing more dangerous than driving twenty miles an hour faster than everyone else on the road is driving twenty miles an hour slower. You can’t control against the effects of the behavior of other people on the highway by adopting a completely different mode of behavior yourself. You are, as every driver knows, almost always better off conforming to the standard practice.

And as with traffic, so with life. Murray is right to think that the alternative to regulation by the government is, in most cases, simply regulation by other means. But he’s wrong to assume that voluntary or informal regulation makes people “freer” than government regulation does. Every social setting has its explicit and implicit rules for conduct which individuals ignore at their peril. We are at liberty to flout the generally accepted conventions of etiquette and decorum—in dress, in deportment, in the proper way of asking someone for a date or of applying for a job—so long as we are prepared to suffer the humiliation and rejection that may ensue. But we are similarly at liberty to drive in excess of the speed limit at the risk of paying a fine or of being sued if we cause an accident. It is hard to see how the first sort of liberty is a higher, or “freer,” kind than the second. It’s true that customs don’t always carry legal sanctions, but neither is there legal recourse for the sense of injury that may be the result of one’s failure to adhere to them. The idea that we are perfectly free to be different so long as no law forbids us rests on an extremely limited conception of forbiddance.

The test case for any anti-state political theory, libertarian or communitarian, is the American South before the 1964 Civil Rights Act. Here is a situation in which a customary pattern of behavior in local communities was forcibly overturned by the central government in the name of a value—racial equality—which a majority of the members of those communities did not share. Murray’s treatment of the actual historical circumstances of desegregation is brief and almost unworthy of comment. The Civil Rights Act was, he asserts, otiose. “Go back to the newspaper indexes for the years immediately preceding the passage of the Civil Rights Act of 1964,” he explains, “and you will find a steady sequence of stories about hotel chains and restaurants and other service providers renouncing racial favoritism or segregation in their operations.” The notion that the movement toward desegregation between 1954, the year of Brown v. Board of Education, and 1964 was the result of a voluntary and natural trend in human relations below the Mason-Dixon line, unaffected by federal court orders, Supreme Court decisions, and the deployment of federal troops, is of a piece with the notion that a decrease in the number of highway fatalities from 55,500 to 46,400 does not, despite mathematical appearances, represent a saving of thousands of lives.

Actual history aside, how does libertarianism address racial discrimination as a theoretical matter? Murray offers an argument on behalf of something called the “freedom of association” (which was, of course, the old rallying cry of the Southern segregationists). “Does unabridged freedom of association permit people to engage in racial, ethnic, religious, or other kinds of bigotry?” he asks. “Yes, as long as it is done passively, without the use of force or fraud.” This is because “a free society is most threatened not by uses of government that are obviously bad, but by uses of government that seem obviously good. Antidiscrimination law has been a leading case in point. Freedom of association is too precious to be sacrificed for any particular social goal—precious as a hallmark of a free person, but precious also as a resource for making a free society work.”

The whole idea of “discrimination” as a bad thing gives him trouble, he says, because although the term has acquired an opprobrious connotation, it really names something honorable and precious. “To discriminate is to perceive differences,” he explains; discrimination is “the basis for the thousands of affiliations whereby we gravitate toward things we like and away from things we don’t like.” And he goes on to enumerate the ways in which employers legitimately discriminate against dishonest employees, schools and parents discriminate against bad behavior, audiences discriminate against bad art, and churches discriminate against sin. What a terrible thing it would be to make discrimination illegal! Of course, bigotry is a bad kind of discrimination, he admits, but outlawing it would mean outlawing all the good kinds of discrimination as well. “You do not have the option of excising the bad kinds of private discrimination and keeping the good ones,” Murray warns. “They are of a piece.”

This is really a little triumph of illogicality. The argument is not only specious on its face—it essentially claims that prohibiting discrimination against people on account of the color of their skin is the equivalent of saying that all poems are equally good. It is also nonsense on its own terms. First we are told that discrimination is a vital faculty of individual and social life, one we must exercise all the time, and then we are told that we are incapable of discriminating between good kinds of discrimination and bad kinds. Too good to lose, apparently, but not good enough to use.

“Freedom of association,” sometimes expressed as “property rights,” is a concept that ends up covering a multitude of cases for Murray. For in every issue he treats, his idea seems to be that the people who have a property interest in a particular mat-ter—food and drug regulation, air traffic safety, social insurance—will get together and decide the rules of the game among themselves. People who don’t like the rulesor who come out on the short end are free to live elsewhere. (This is a good reason, he points out, for requiring all government action to be local: “It is much easier for the average person to move out of Detroit than it is for him to move out of Michigan, and infinitely easier than to move out of the United States.”) It’s this privatization of everything that makes the civil libertarian side of Murray’s libertarianism meaningless; for though government cannot discriminate or regulate personal behavior, property owners—landlords, employers, colleges, and clubs—can, in Murray’s view, discriminate and regulate as much as they like. Public space is still “free,” but there is hardly any of it left.

We will still have, in a libertarian society, a Food and Drug Administration, for example. But it will be operated by the food and drug industries, which will see such an agency as being in their own best interest—a kind of Good Housekeeping seal of approval on their products—and its regulations will be optional. People who find a thumbtack their Twinkie will still have an avenue of redress open to them. They can sue; and if they can prove that the manufacture of that Twinkie was not in accordance with standard industry practices (however these are determined), they can recover. Tort law, Murray thinks, can take the place of product regulation. This is apparently intended as an argument for efficiency, on the grounds that it creates more options for consumers. But the idea that it is more efficient to enhance product safety by an endless series of expensive individual lawsuits for faulty products rather than by imposing universal standards seems completely wrong-headed. Whatever else one can say about it, libertarianism will evidently be very good for lawyers.

Two things are peculiar about the notion of privatized regulation, and about Murray’s view of social and political life generally. The first is a common feature of a certain kind of contemporary conservative thought. Abstract philosophical reasoning is used to establish a set of norms for human life, and it is then discovered that, by some wholly unforeseen coincidence, these norms correspond almost exactly to the way things were in the United States circa 1958. Murray’s obsession with the 55-mph speed limit is a classic instance of this way of thinking: he writes as though 70 mph were simply the speed at which God intended people to drive. Thus Murray doesn’t really propose to get rid of any of the functions the federal government has exercised, or the investments it has made, or the human rights it has established—almost always over the fierce and sometimes violent objections of local communities and private interests—since the Great Depression. He wants food and drug regulatory agencies, air traffic controls, interstate highways, social insurance, welfare agencies, freedom of expression, environmental protection, universal access to education, and virtually every other benefit of modern life. He just wants them all privatized. Having granted every wish, the genie of big government can now return to the bottle.

A more idiosyncratic peculiarity of Murray’s book is the dissociation it persists in making between people and some entity called “government.” What It Means to Be a Libertarian does not contain a single discussion of—it barely contains a single reference to—voting, elections, or the principle of political representation. There is one disparaging reference to “a voting majority of some legislature,” as though the very phrase were synonymous with arbitrary authority. The word “democracy” hardly appears. Complete faith is placed in the ability of human beings to make their own rules for running their lives, without the slightest hint that government is just a lot of human beings, granted provisional power by other human beings to make decisions about certain aspects of the common life, subject to recall or repeal. No doubt there are many cases in government today in which an interest-group tail wags the democratic dog—in which a minority finds a way to influence legislation and dictate terms to the majority. In Murray’s world this won’t be a problem, though, because the interest group will be the whole dog. The concept of “the public,” in his theory, is replaced by the concept of “the customer”: you can buy what they’re selling, or you can get the hell out of Detroit.

Libertarianism has two roots. One (which Murray pays his acknowledgements to) derives from Adam Smith and classical liberalism. The other derives from Rousseau and Romanticism. Homo economicus and the noble savage, the rational benefit-maximizer of laissez-faire economics and the virtuous primitive of Romantic art and literature, represent rather divergent models of humanity’s essential nature, but they both reflect the conviction that freedom is our natural state. “Freedom is first of all our birthright,” as Murray puts it in the opening pages of his book, and his whole philosophy is built up from that premise.

If we look at the actual things we call freedoms, though—the freedom to speak and to believe what we choose, the freedom to peaceably assemble and to petition the government for a redress of grievances, the freedom to travel—we find that they have been established only by the exertion of centralized political power against the natural tendency of some groups of human beings to stifle and control other human beings, particularly human beings who look or think differently from themselves. Libertarianism of Murray’s variety is a philosophy for winners, for the same reason that “Born Free” is a song about lions, not about the animals they prey upon. Not many people who genuinely feel vulnerable or oppressed will find much use for it. What those people will want is a public body vigilant in protecting their interests, including their interest in personal freedom, against the tendency of majorities to dominate and exploit them, and they will be willing to abide by the decisions of such a body so long as they are democratically reached. Groups make laws, formally or informally, by consent or by fiat. Democracy is a way of trying to do it by consent. There is a law of the jungle, but it’s not particularly big on “liberty,” and the animals don’t get to vote.

This Issue

February 20, 1997