The conservative attacks on “judicial activism,” a long-standing GOP target, have intensified in recent years. In his July 19, 2005, speech nominating John Roberts to the Supreme Court, President George W. Bush said that he wanted judges who will apply the law, “not legislate from the bench.” Earlier that spring, after the federal courts refused to order the feeding tube restored to the brain-damaged Terri Schiavo, then-House Majority Leader Tom DeLay promised reprisals against the federal judges involved. At around the same time, Senator John Cornyn of Texas gave an astounding speech in which he linked recent attacks against judges, including the murder of a federal judge’s husband and mother, to public frustration with judicial activism. Cornyn said that anger over “political decisions” by judges “builds up to the point where people engage in violence.” And this year, a bill was introduced into Congress to create an inspector general to investigate and oversee the federal courts.

Of course, there is an irony in that the conservative focus on judicial activism is coming at a time when seven of the nine justices on the Supreme Court and a majority of federal judges were appointed by Republican presidents. This fact aside, there can be little doubt that the attacks on judicial activism will continue. Such attacks may have much to recommend them as a political tactic, but is looking at our courts through the lens of activism/restraint useful, in substantive terms, when judging judges?

Kermit Roosevelt III, a former Supreme Court clerk and current law professor at the University of Pennsylvania, says no. In his new book, The Myth of Judicial Activism, Roosevelt argues that the criticisms of judicial activism are misguided because there are no clear criteria for determining what is activism and what counts as restraint. Both conservatives and liberals sometimes want courts to defer to legislatures, and sometimes to invalidate laws themselves. Both sides sometimes want courts to affirm precedents and, sometimes, to overrule them. Conservatives, for example, want no deference to government affirmative action programs, or government takings of private property for economic development purposes, or federal laws authorizing suits against state governments. On the other hand, they want judicial deference to laws regulating sexual practices between consenting adults, laws regulating abortion, and government actions advancing religion. For liberals, of course, the reverse is true. And thus the usefulness of “activism” as an organizing concept is limited.

Instead, what matters, argues Roosevelt, is not whether a court’s decision is stepping over its bounds or not, but whether it is “legitimate”–in other words an “appropriate exercise of judicial authority.” Legitimacy is thus a normative concept that can be used in evaluating particular decisions. It is not, as others have used it in the past, a measure of how the decision is regarded by the public. One striking aspect of Roosevelt’s definition is its binary quality: As he defines legitimacy, it exists or it doesn’t; a case either is or isn’t an appropriate exercise of judicial authority. To Roosevelt’s credit, he does not fall back on a Justice Potter Stewart–type standard of “you know it when you see it” when determining a ruling’s legitimacy. Instead, he contends that legitimacy is to be determined by whether the rulings show the appropriate degree of deference to the political process.

The book is written in a clear, accessible style that makes it understandable to non-lawyers. And its scope is impressive. In the process of advocating a theory for appraising constitutional decisions, Roosevelt offers his views on virtually every controversial issue in constitutional law: detention of enemy combatants, gay rights, abortion, separation of church and state, the death penalty, campaign finance reform, and abortion. However, although I generally agree with Roosevelt’s positions on these issues, I have serious reservations about his central thesis. Indeed, in his desire to chart a sweeping constitutional theory, Roosevelt unwittingly proves the difficulty–and even futility–of the exercise. This, in turn, raises the question of whether any sweeping theory of constitutional interpretation is needed at all by our judges.

Roosevelt believes that the legitimacy of a court’s decision can be ascertained by measuring its deference–or lack thereof–to the political process, as determined by a set of five criteria. The first factor is “institutional competence,” which he leaves problematically ill-defined. Roosevelt says, for example, that legislatures have more competence in making complex factual determinations and that legislatures generally are more competent in assessing the desirability of laws, which justifies the usual presumption of constitutionality applied by courts. But he says that other factors–especially the lessons of history and defects in democracy–provide instances where judicial deference is not appropriate.

The second criterion are the lessons of historymthemselves, a factor that is almost entirely focused on matters of equal protection. If there is a history of discrimination against a group, then it means that there should be less deference to the political process. Most obviously, the long, tragic history of racial discrimination justifies more aggressive judicial scrutiny of laws disadvantaging racial minorities. Interestingly, Roosevelt does not include within the lessons of history the role of judicial precedent, as some conservatives–including the present Chief Justice–would. Nor does he include “traditions” that often have been the basis for judicial analysis in constitutional cases, especially in deciding which rights are protected under the Due Process Clause. For instance, in Moore v. East Cleveland, the Court stressed traditions in striking down a zoning law that kept a grandmother from living with her grandchildren. In other words, it is not clear how precedent or tradition would fit within Roosevelt’s theory of judicial review.

Third, Roosevelt argues that “defects in democracy” are a basis for determining the appropriate degree of judicial deference to the elected branches of government. Here he echoes the approach taken by the late Stanford Law School professor, John Hart Ely. Yet, while Ely made this the centerpiece of his theory of judicial review, for Roosevelt it is only one of five factors to be considered. Under this factor, courts have a special role to play when they are dealing with imperfections in the democratic process, such as in curing malapportionment of congressional districts or protecting discrete and insular minorities, most notably racial minorities.

Fourth, there are the “costs of error” to be considered. Roosevelt says that if the Supreme Court errs in striking down a law as unconstitutional, it is very difficult to correct that mistake; only a constitutional amendment or an overruling by the Court can overcome the decision. On the other hand, if a court errs in upholding a law, there is still the chance for a fix through the political process. For example, when the Supreme Court held in City of Mobile v. Bolden that government actions with an adverse effect on racial minorities did not violate the Constitution, Congress amended the Voting Rights Act to prohibit this. But Roosevelt also acknowledges that there is a cost of error in not enforcing the Constitution, such as when innocent people are left confined in prison or groups needing judicial protection suffer needless persecution.

Finally, Roosevelt points to the difference between “rules and standards”: “A rule has sharp edges,” he writes, “it permits or prohibits a definite range of action. A standard is more open-ended; its application in particular cases will be open to dispute.” Simply put, rules are meant to provide a clear, bright-line criterion; standards are flexible. As an illustration, a rule would be that the government always must provide a free attorney to poor defendants facing parental termination proceedings. A standard, by contrast, would have a court balance competing considerations in each case to decide if an appointed attorney was constitutionally required. Roosevelt says that sometimes rules are better because they provide clear guidance to lower courts and to individuals as to what is required, though he acknowledges that sometimes standards are better because they provide more flexibility and less rigidity than rules.

Working with these five factors to determine a ruling’s legitimacy, Roosevelt then proceeds to review the landmark Supreme Court decisions and how his framework would have impacted them. He begins by examining what he regards as “easy cases”: Brown v. Board of Education, Miranda v. Arizona, and executive detention as part of the war on terrorism. He then proceeds to what he calls “hard cases”: gay rights, abortion, taxings, the Establishment Clause, the death penalty for crimes committed by the mentally retarded and juveniles, and campaign finance reform. Following this, he considers examples where the Supreme Court wrongly has refused to defer, focusing especially on affirmative action and Bush v. Gore. Finally, he looks to what he calls “reviled decisions”: Dred Scott v. Sandford, Plessy v. Ferguson, Lochner v. New York, and Korematsu v. United States. Using his framework, they all would have been decided differently.

Roosevelt could not have a more sympathetic reader than me. I share his view that the focus on judicial activism is misguided; it is political rhetoric without substance and gives a misleading impression about what courts and judges do. And, almost without exception, I agree with his position on each of these hot-button issues he covers. I would like to see him succeed in justifying the results he advocates. The problem is that the five factors he identifies don’t add up to a useful theory of judicial review.

First, it is unclear where these factors come from or why they are the appropriate measures of the desirability of judicial decisions. Two of them–the history of discrimination and the basis for distrust of the democratic process–are already used by the Supreme Court. In the famous footnote four in the 1938 decision in United States v. Carolene Products Co., the Supreme Court said that there is a general presumption of constitutionality, but that deference is not appropriate when the government is interfering with fundamental rights or discriminating against a discrete and insular minority, or where the political process has malfunctioned.

The other three factors–institutional considerations, costs of errors, and rules as opposed to standards–are novel. Yet Roosevelt does not adequately explain why they should be used. Part of the problem is that Roosevelt’s description of them is so cursory as to make it impossible to know how they are to be applied. For example, he acknowledges that sometimes rules are best, while other times standards are preferable. Yet he offers no criteria for determining when one is better than the other. Similarly, Roosevelt says that institutional considerations should influence the assessment of the desirability of a judicial decision. But what institutional considerations? How is it to be determined in a specific area whether institutional considerations favor courts or legislatures making the decision?

Perhaps most importantly, though, is that these five factors are so indeterminate that they likely could be used to justify any conclusion. Multifactor analysis inevitably means that there is a great deal of discretion and that almost any result can be reached, as some factors point one way and others in a different direction. Roosevelt does an excellent job of showing that the five factors he identifies can be used to justify liberal results as to the many issues that he discusses. The problem, though, is that these factors can be just as easily used to justify the opposite conclusions.

Roosevelt, for instance, argues that the Supreme Court was justified in protecting abortion rights, though he criticizes Justice Harry Blackmun’s opinion in Roe v. Wade. My views aside (yet I would defend Justice Blackmun’s opinion), the key is that the five factors Roosevelt identifies could be used to justify either affirming or overruling Roe. Institutional considerations might be used to argue that the legislature is in the best position to weigh the woman’s privacy interests against the interests in protecting fetal life. But institutional considerations might be said to favor the Court deciding because the fundamental right of privacy is implicated. Moreover, Roe has the advantage of providing “rules,” but so would denying all constitutional protection for abortions.

As a result, the reader is left with Roosevelt’s conclusions as to each area of constitutional law, but little more than that. Take his view of substantive due process: Roosevelt is very critical, even disdainful, of it. He writes: “The idea that judges should be deciding cases under the Due Process Clause by determining whether a right is fundamental is bad policy and bad law.” Little explanation is given. He suggests that due process is about procedures, but for at least a century, due process of law has included both a requirement that the government’s action be justified by a sufficient purpose and that the government follow proper procedures. Overruling the substantive component of due process would radically change the law, including overruling Supreme Court decisions protecting the right to marry, the right to procreate, the right to custody of one’s children, the right to keep the family together, the right to control the upbringing of one’s children, the right to purchase and use contraceptives, the right to engage in private consensual homosexual activity, and the right to refuse medical treatment.

In fact, Roosevelt seems to suggest that eliminating judicial protection of these rights would be a good thing. “And since I can think of no good way for judges to decide what these unenumerated rights might be, leaving them to the people seems a so under course,” he writes. However, the very factors that he identifies would provide a basis for protecting these rights. The Supreme Court is institutionally better suited to decide what “liberty” means and what aspects of liberty should be protected from the political process than the political process itself. History shows that often these rights are inadequately protected in the democratic process. The costs of errors are great if the political process fails to provide protection. Abandoning judicial protection of fundamental rights would be a radical and very undesirable change in the law.

Ultimately, Roosevelt’s attempt to expose the “myth of judicial activism” is successful, but in trying to offer an alternative, he reaches too far. At a time when confirmation hearings are full-blown battles with intense scrutiny of each nominee, the debate inevitably turns to the nominee’s judicial philosophy. This book, perhaps inadvertently, points to the futility of such an effort.

Indeed, throughout American history, the Supreme Court has interpreted the Constitution by looking to its text, its structure, its goals, judicial precedents, traditions, and contemporary needs and values. Yet few Supreme Court justices have ever espoused an overall theory of constitutional interpretation, and even those who did often have abandoned it. Perhaps, then, it is time to move past the judicial-theory obsession and focus instead on arguing over the best constitutional vision for each area of constitutional law. What is the best way to understand the president’s powers as commander-in-chief? What is the best way to understand what is “cruel and unusual” punishment? What is the best way to understand the protections of the free exercise clause? In each area, the discussion and disagreement will be about text and structure, purpose and precedent, and tradition and contemporary needs and values. Inescapably, constitutional law requires that Justices make value choices that cannot be determined by any constitutional theory. What is a “legitimate” or an “important” or a “compelling” government interest? No constitutional theory, including Roosevelt’s, can provide a coherent way of determining this.