WHEN HE nominated Judge David Souter for the Supreme Court, President Bush was asked if he was certain "how Judge Souter will vote if Roe v. Wade comes before the court." The president responded by simply saying, "What I'm certain of is that he will interpret the Constitution and not legislate from the federal bench."

The unstated premise of Bush's answer is that any judge who is committed to interpreting and not legislating would necessarily find Roe an illegitimate decision and vote to overrule it: I.e., since "abortion" is not mentioned in the Constitution, any judge who finds a constitutional right to abortion must be legislating.

The issue, however, is not so easy. As the questioning of Souter at his Senate hearing this week will likely demonstrate, a simple statement that a judge should "interpret and not legislate" does not resolve the difficult issues raised by Roe v. Wade. The Constitution does not contain the word "abortion," but it does expressly provide in the 14th Amendment that "No state shall deprive any person of life, liberty or property without due process of law." Holding that some state restrictions on abortion violate this clause is a defensible judicial interpretation.

It is now well established that a state deprives persons of "liberty" without due process if the government places a restraint on liberty that serves no legitimate public purpose. In this century, moreover, the Supreme Court has held that a few liberties are so fundamental that a routine public purpose is an insufficient basis for overriding the fundamental right at stake. Government may invade a fundamental liberty only if it clearly establishes a compellingly important reason for doing so.

Some conservative scholars would restrict the content of fundamental liberty under the 14th Amendment to those few items specifically contemplated by the members of the 39th Congress who framed the amendment. Such a narrow reading of the Constitution would mean, among other things, that discrimination against women did not violate the equal-protection clause, that the Bill of Rights limited only the federal government (and not the states) and that legislatures could be grossly malapportioned. None of these matters were specifically contemplated by the Reconstruction Congress in 1866.

A court may not, of course, disregard the text of the Constitution: it is the source of judges' authority. A judge must, moreover, interpret a constitutional provision in light of the meaning intended by its authors. This does not mean, however, that judges must limit the content of "liberty" to those specific items that the framers would have listed had they chosen to enact a specific list. In fact, the framers of the 14th Amendment did not choose to limit it Amendment to any itemized menu of liberties.

Consider this analogy: You would agree that if you were given a grocery list that called for "potatoes" and purchased "tomatoes," you would not be interpreting the author's list but substituting your own. Yet suppose your grandmother bequeathed $500 and a note instructing you to use the money when you grew up to "buy nutritious food" (a phrase closer in its scope to "liberty" and "equality").

Applying this instruction 20 years later, a "strict construction originalist" would engage in detective work, poring over grandmother's diaries and cookbooks in an attempt to find out what specific items she in fact ate. ("Eat fatback twice a day," she might have written, "it greases up the veins.") But did she really intend us to uncover what specific items she would have included if she had chosen to make a list? Is it not more likely that she meant that you were to use her money to buy foods that, after careful consideration, you determined to be nutritious? The framers of the 14th Amendment chose to employ broad language guaranteeing "liberty," "equal protection" and "privileges and immunities" against state interference. The Congress that proposed this language, moreover, was fully aware of a tradition of broad interpretation of Constitutional terms that had been established during the first half century of the republic. And they knew that judicial review had been firmly established as part of our constitutional tradition. An attempt to reduce the broad principles of the 14th Amendment into a specific, historically contingent list is by no means faithful to the intended function of these clauses. (I sometimes suspect that hard-right conservatives have never been reconciled to the adoption of the 14th Amendment and wish it would either go away or be interpreted out of existence.)

Respected jurists like John M. Harlan and Lewis H. Powell understood that the obligation of the court is not to ignore the 14th Amendment's broad guarantee of liberty but rather to determine on a cautious, case-by-case basis whether some laws are so intrusive into fundamental liberties that they cannot be justified by routine governmental objectives.

As Harlan wrote, in upholding the right of married couples to use birth control, "the full scope of the liberty guaranteed by the due-process clause cannot be limited by the precise terms of the specific guarantees elsewhere provided in the Constitution." He noted that "certain interests require particularly careful scrutiny of the state needs asserted to justify their abridgment."

Powell continued this tradition, noting, "This court has long recognized that freedom of personal choice in matters of marriage and family life is one of the liberties protected by the due process clause of the 14th Amendment." When the government intrudes upon an intimate and fundamental matter, "this court must examine carefully the importance of the governmental interests advanced and the extent to which they are served by the challenged regulation."

Powell acknowledged that the process of giving content to liberty "has at times been a treacherous field for the court. There are risks when the judicial branch gives enhanced protection to certain substantive liberties without guidance from the more specific provisions of the Bill of Rights." Principles of federalism, respect for other branches of the government and humility about one's own judgment should lead a judge to exercise great care in determining which few liberties are fundamentally important. Constitutional history, Powell properly warned, "counsels caution and restraint" by the court. "It does not," he added, "counsel abandonment" of judicial protection of liberty. The current debate is over which liberties are of fundamental importance, not over their existence. The right of an individual to make her own choices about critical matters of reproduction and family formation is a constitutional liberty of fundamental importance. From the time of Meyer v. Nebraska in 1923, the court has recognized that a couple's decision about how to raise their children is an especially important liberty. From the time of Skinner v. Oklahoma in 1942, the Court has recognized that a state needs an extraordinary justification when it acts to control a person's reproductive capacity.

Government-compelled childbirth is an extraordinary intervention into a person's life. It is difficult to imagine a meaningful concept of fundamental liberty that does not encompass protection of a person's reproductive choice. If the choice to continue or terminate a pregnancy is anything less than fundamental -- if it is, as Chief Justice William H. Rehnquist would hold, a merely ordinary "liberty interest" -- state and local governments would need only minimal justification to require women to undergo compelled abortions as well.

Imagine the passage of a state law that prohibited needed hysterectomies. Even though the 39th Congress was not thinking of reproductive surgery when drafting the 14th Amendment, a court that invalidated such a law would not be "legislating a new constitutional right to hysterectomy" but merely applying the established constitutional rule that deeply intrusive state laws must be justified by compelling governmental reasons.

Once it is established that a woman has a fundamental right to terminate a pregnancy, the state may still attempt to override her right by asserting that protecting fetal life constitutes a compelling justification for invading the woman's liberty. This is the most difficult point in Roe. But as Harlan observed, "The mere assertion that the action of the state finds justification in the controversial realm of morals cannot justify alone any and every restriction it imposes." While many individuals believe that human life begins with fertilization, there is no compelling national consensus or tradition on this point. As Gene Sperling and I wrote recently, "An interest pursued only by some states in some decades, and even then in an inconsistent and economically discriminatory fashion, can hardly qualify as a compelling governmental interest."

The framers of the 14th Amendment understood that it would subject state legislation to federal judicial review under broadly worded provisions protecting equality, privileges, immunities and liberties. Judges who apply this provision to determine the scope of an individual's right to be free of state control of reproduction are not "legislating" but performing a duty assigned them by the Constitution of the United States.

When the Judiciary Committee questions Souter, senators should ask not only whether he will "interpret the Constitution" -- of course he will -- but how he will go about that difficult task. It is important for the Senate to learn whether a nominee takes a narrow, historically confined view of liberty and equality, and to decide whether confirming a nominee who takes such a view would serve the nation's best interest.

Walter Dellinger is a professor of law at Duke University.