From left to right, not many people have had a kind word for Roe v. Wade, the 1973 Supreme Court decision that partially legalized abortion. Radical feminist Catharine MacKinnon called it “a study in male ideology” that “reaffirms what the feminist critique of sexuality criticizes: the public/private split.” Michael Kinsley, editor of “The New Republic”, called it “one of the worst things that ever happened to American liberalism” and warned that “there is a time bomb ticking away” inside it. Conservatives have, of course, registered disapproval often and loudly (sometimes explosively).

Roe v. Wade struck down a Texas law that made abortion a crime except when “procured or attempted by medical advice for the purpose of saving the life of the mother.” The court’s logic was not transparent (Kinsley calls it “a mess,” and many agree), but was essentially this: (1) A woman’s right to make reproductive decisions is part of a “right of privacy” implicit in the 14th Amendment to the constitution, which says that government may not “deprive any person of life, liberty, or property without due process of law; nor deny to any person within its jurisdiction the equal protection of the laws.” (2) A fetus does not have a “right to life” or to “equal protection” because, according to the court, “the word ‘person,’ as used in the Fourteenth Amendment, does not include the unborn.” (3) Constitutionally guaranteed rights may only be infringed when a “compelling” societal interest is at stake. In the case of abortion, society has two legitimate interests: to protect the health of pregnant women and to protect “potential life.” But these interests are not “compelling” from the moment of conception. The former becomes compelling only when the medical risks of abortion become comparable to those of normal childbirth, i.e. (as of 1983) during the second trimester. The latter becomes compelling only when the fetus becomes “viable,” or “capable of meaningful life outside the mother’s womb,” i.e. (as of 1973) during the third trimester. (4) What follows from all this, the court said, is that the state may regulate abortion during the second trimester, but only “in ways that are reasonably related to maternal health,” and may prohibit abortion during the third trimester “except where it is necessary . . . for the preservation of the life or health of the mother.”

Roe raised a host of questions, of two kinds. First: is the court’s reasoning persuasive? Does the constitution really recognize a “right of privacy,” even though it’s not mentioned anywhere in the text? Why doesn’t the 14th Amendment apply to fetuses, even though they are not mentioned anywhere in the text? What is “potential” life? Why does the state have a “compelling interest” in protecting it, even at the expense of actual persons? And why does that interest become compelling at viability rather than at conception or birth? Second: even if the court was right on all counts, is this how social policy should be made in a democracy? Insofar as the national political debate over abortion since 1973 has been a debate and not a jihad, these questions have defined it.

The guild of constitutional lawyers divided sharply over Roe. John Hart Ely, possibly the leading contemporary theorist of constitutional adjudication and a longtime (and continuing) supporter of legalized abortion, called it “frightening…a very bad decision.” Archibald Cox deplored the court’s “failure to confront the issue in principled terms.” But other eminent jurists disagreed. The issue turned on what may be the thorniest question in American constitutional law: how to interpret the “due process” clause of the 14th Amendment. What does “life, liberty, and property” include—only those rights enumerated elsewhere in the constitution (e.g., religion, speech, assembly) or all rights the court considers “implicit in the concept of ordered liberty” (Justice Harlan’s famous definition)? This is roughly the shape of the present-day dispute between advocates of judicial ‘ (or “strict construction”) and advocates of judicial “activism.” Obviously the latter position would make it easier to find constitutional support for a “right” to reproductive autonomy.

As the academic debate flared and the court’s Nixon and Ford appointees (“strict constructionists,” for the most part) gained confidence, the Roe majority seemed to lose its nerve. In the late 1970s several state legislatures and then Congress eliminated Medicaid funding for all except “medically necessary” abortions. The lower courts invariably struck down these regulations as unwarranted interference with a constitutionally guaranteed right; but to everyone’s surprise, the Supreme Court upheld the rules. The court did not reverse Roe. It did not over look Roe. In each case, the court explicitly reaffirmed Roe and then handed down a decision plainly inconsistent with it.

The court’s reasoning in the abortion funding cases—that Roe did not prevent government “from making ‘a value judgment favoring childbirth over abortion, and... implement[ing] that judgment by the allocation of public funds’”—was lame. The Medicaid program had been established to provide medical services for the poor. To discriminate among the services provided for no reason related to the overall purpose of the program, but solely in order to express the legislature’s moral disapproval, was clearly unconstitutional, especially when the activity discriminated against had the status of a constitutional right. Constitutionally protected rights may not be interfered with merely in order to express the majority’s moral disapproval; that is the definition of a constitutionally protected right. And there was no question that moral disapproval was the sole motive of the regulations: Congressman Henry Hyde, author of the measure, said so again and again.

The Hyde Amendment was the right- to-life movement’s most successful political gambit, but not its most ambitious one. In 1981 the human life bill was introduced in both the House and Senate. It stated that “for the purpose of enforcing the obligation of the States under the fourteenth amendment not to deprive persons of life without due process of law, each human life exists from conception for this purpose ‘person’ includes all human beings.” If it passed, any state legislature would be free to outlaw abortion, as many of them were eager to do.

The Human Life Bill was a direct response to Roe v. Wade. Its proponents regarded Roe as the founding charter of what Senator John East, the bill’s Senate sponsor, kept calling “the abortion-clinic society”; and they would evidently try anything to undermine the decision. But Roe appeared to invite just such an attempt. The court had ruled that fetuses were not “persons” in a 14th Amendment sense. But it had also asserted, without explanation, that the state had a compelling interest in protecting “potential life.” And in an apparent gesture of humility before this “most sensitive and difficult question,” the court committed a fateful ambiguity: “We need not resolve the difficult question of when life begins. When those trained in the respective disciplines of medicine, philosophy, and theology are unable to arrive at any consensus, the judiciary, at this point in the development of man’s knowledge, is not in a position to speculate as to the answer.”

This curious remark seemed to leave open a legislative loophole. The 14th Amendment provides that “Congress shall have power to enforce, by appropriate legislation, the provisions of that article.” In American constitutional tradition, fact finding for the purpose of policy making is the function of Congress, interpretation the function of the Supreme Court. But was the question “is the fetus a person?” a matter of fact or of interpretation? The proponents of the human life bill claimed that Congress’s enforcement power gave it the authority to define key terms in the 14th Amendment by making a “finding of fact” as to “when life begins.” The legal profession responded pretty much with one voice that this was a radical and dangerous challenge to the separation of powers and thus to the constitutional structure or American government. In the end a majority of Congress agreed and the bill failed.

Apart from the constitutional challenge, there were two ominous aspects to the debate over the human life bill. For one thing, it revealed a certain lack of fair-mindedness on one side. A platoon of extremely distinguished scientists and physicians trooped through the Senate committee room, explaining, sometimes with considerable eloquence, that questions like “when life begins” or what a “person” is are not factual and scientific questions but moral and philosophical ones. The result of all this eloquence was that the initial draft of the bill was changed from “Congress finds a significant likelihood that the life of each human being begins at conception” to “Congress finds that the life of each human being begins at conception.” The phrase “a significant likelihood” was dropped because, according to Senator East, pro-choice witnesses had offered no “objective evidence” to the contrary, but only “value judgments.”

Even more ominously, both sides demonstrated a stark indifference to the interests of born women. Pro-choice witnesses convincingly portrayed the bill’s threat to the professional discretion physicians and vividly evoked various hard cases—raped adolescents, genetically defective fetuses, etc. But in 200 pages of testimony—admittedly a small fraction of the voluminous Senate record—I came across not a single reference to the economic and sexual independence of women as a group. No one asked why the protection of potential life is more compelling than the enhancement of actual lives.

Of course, Roe v. Wade, the Medicaid funding restrictions, and the human life bill were debated and decided largely by men. Far from the mainstream, feminists also voiced doubts about Roe. The court had affirmed, with some reservations, that “the right of privacy, however, based, is broad enough to cover the abortion decision.” But whose privacy and whose decision? Near the end of their opinion, the justices summed up: “This decision vindicates the right of the physician to administer medical treatment according to his professional judgment up to the points where important state interests provide compelling justifications for intervention. Up to those points, the abortion decision in all its aspects is inherently, and primarily, a medical decision, and basic responsibility must rest with the physician.” So much for the rights of women; two steps forward, one- and-a-half steps back.

One particularly penetrating critique, Catharine MacKinnon’s “Roe v. Wade: A Study in Male Ideology,” took this objection still further, MacKinnon identified the key assumption of the “privacy doctrine” that underlies much liberal support for abortion rights: government has no right to interfere in the sphere of the “private” because that is the sphere of autonomous, equal, freely contracting selves. (This is also—though MacKinnon does not raise the point—the precise form of the liberal defense of market relations.) Now, feminist investigations have shown that the realm of the private is anything but a realm of freedom—is, in fact, a realm of pervasive inequality that requires, if anything does, public remedies. But it is just such public action that the privacy doctrine precludes. From this MacKinnon argued, plausibly and disturbingly (though in the end, I think, mistakenly) that the Hyde Amendment was logically and ideologically consistent with Roe.

Notwithstanding such reservations, most feminists and other radicals consider Roe, at least provisionally, a victory. No doubt it is, at least provisionally. But in the long run, it may prove a Pyrrhic victory. In Roe the court broadly construed the “due process” clause, on the theory that the constitution implies many more rights than it enumerates. At first glance this interpretive strategy (known as “substantive due process”) may seem more appealing, more libertarian, than its opposite number, strict constructionism. But substantive due process also has illiberal implications. Before Roe, the best-known application of the doctrine occurred in Lochner v. New York, a 1905 case in which the Supreme Court struck down progressive labor legislation. The court held that limiting the work day to 12 hours violated the worker’s and employer’s “liberty of con____” a right it claimed to derive from the due process clause. Subsequent decisions struck down laws setting minimum wages for women, restricting “yellow dog” contracts, and prohibiting the sale of shot-weight loaves of bread—all in the name of economic “liberty.” These decisions were gradually abandoned, but their lesson for both sides remains: granted discretion, judges are just as likely “to discover uncongenial “rights” as congenial ones.

Roe was the last in a series of cases, beginning with Brown v. Board of Education (1954), that rekindled intense debate over the proper scope of judicial review. Unlike the Lochner series, most of these cases—dealing with civil rights, school integration, criminal law, prison reform, electoral reform, and so on—had “progressive” outcomes. But should radical democrats be happy with the way these reforms became social policy? The courts are, after all, a counter-majoritarian institution; and there is an uncomfortable irony about “progressives” appealing to constitutional tradition.

George Will, as might be expected, scored heavily off this irony:

“Intelligent persons of good will can disagree as to whether the radical revision of abortion laws since 1973 was wise. But it is impudent of those who think so, and who worked tenaciously to bring this about by judicial fiat, to argue that those who disagree and want to restore the rights of states to regulate abortions are guilty of wanting to “impose” their “personal” values on society. What do the triumphant advocates of the social policy of abortion-on-demand think they have done? They have used judicial power to strike down laws that embody the democratically expressed judgments of all fifty states.”

Of course it is impudent of Will to equate forcing an involuntarily pregnant woman to bear a child with forcing Will to live in a society where she can have an abortion if she wishes. These are hardly parallel forms of “imposing” one’s personal” values. Still, there is something troubling about that final phrase: “the democratically expressed judgments of all fifty states.” Democratically expressed judgments do not always deserve obedience, but they deserve respect, at least from democrats. Even the diehard Senator East declared—sincerely or not—that he’d be “more inclined to accept [Roe v. Wade] if it had been a decision arrived at in the legislative body over a period of time through public discussion.”

There are two answers to Will’s argument, neither of them conclusive. For one thing, the majority vote of a legislature does not always “democratically express” the majority sentiment of the electorate. Single-interest constituencies—the National Rifle Association, the tax-shelter industry, the right-to-life movement— can seriously distort the democratic process. This is not a debater’s point. In 1972, vetoing an attempted repeal of New York’s comparatively liberal abortion law, Nelson Rockefeller explained that “the extremes of personal vilification and political coercion brought to bear on members of the Legislature raise serious doubts that the votes to repeal the reform. . . represented the will of a majority of the people of New York State.” To the extent that they concentrate on browbeating legislators rather than persuading voters, single-interest groups may thwart, or at least preempt, democratic decision making.

For another thing, the popular will is not univocal. A majority may want incompatible things, like enforcing its sexual morality and adhering to the constitution. It is true that all morality is ultimately conventional, and also true that a shared morality is indispensable to genuine community. But part of every democratic society’s morality is a tacit, traditional agreement marking the limits of majority power. The philosopher Ronald Dworkin calls this our “constitutional morality”; it is the closest thing we have to a social contract. A society’s conventional morality may be in conflict with its constitutional morality; resolving such conflicts is what courts are for. Legislatures attempt to enforce current mores; courts decide whether these attempts are compatible with our “higher,” permanent morality. Courts may be wrong, of course, or we may want to alter the terms of our social contract. In these cases, a large, authoritative majority can amend the constitution, bringing conventional morality and constitutional morality into line. Seen this way, Roe v. Wade was a prediction by the Supreme Court that, once we’ve thought about it, we’ll recognize that reproductive autonomy is implicit in our traditional “concept of ordered liberty.” Only the passage of a Human Life Amendment can refute that prediction.

So matters are not as straightforward as Will’s confident references to “society” and its “democratically expressed judgments” would suggest. Nonetheless, he has a point: “power to the people” is a complicated notion, and radicals don’t own it. In February of this year, overturning congressional spending limits on political action committees, the Supreme Court noted that the National Conservative Political Action Committee had received 100,000 contributions, averaging $75 each. The court said that looked a lot ‘like democracy, and it had a point.

Michael Walzer has put the left’s predicament well: “We have won victories far beyond our political base—and this over a very wide range of issues. Civil rights, school integration, school prayer, affirmative action, criminal law, capital punishment, prison reform, electoral reform, abortion, pornography: in all these areas, with only the partial exception of the first, victories have come not because of the political strength of the left but because of the persuasiveness of its lawyers. The case of the right has been exactly the opposite. Whereas we have won in court, again and again, and failed to build a movement, they have lost, again and again, and organized outside... committees, alliances, even single-issue parties, at local as well as national levels: an extraordinary proliferation of groups, which participatory democrats might well envy.” All this is embarrassing as well as inconvenient. So much so that we may be tempted to wish the trolls would return to their holes. But that would be wrong: the only cure for the evils of democracy is more of it.

It is now common to talk glibly about the “tragic dilemma” of abortion. There is, in fact, an aspect of tragedy about this issue, though it’s not the one that agitates middle-aged male legislators and columnists. The “tragedy” of abortion consists not in some imaginary “slaughter of the innocents,” but in the willingness of a near-majority of the electorate to use the American constitutional machinery—with all its flaws, as enlightened and emancipatory a political arrangement as presently exists—to subject themselves and their fellow citizens to a primitive and barbarous superstition.

Roe v. Wade is not, strictly speaking, part o that tragedy. It is part of the prologue, now comes the play.