"It is one of the happy incidents of the federal system," Justice Louis D. Brandeis wrote in 1932, "that a single courageous state may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country." Conservative and liberal justices have quoted Brandeis’s dictum in some three dozen cases. The metaphor invariably surfaces in any scholarly or public discussion of federalism and is accompanied by emphatic nods of approval. It conveys a pragmatic spirit that naturally appeals to a nation of compulsive tinkerers, and it connotes equally popular sentiments in favor of localism and decentralization.
Popular appeal aside, one can make a powerful theoretical case for the experimental, decentralized politics that the laboratory metaphor suggests. Political institutions should be capable of adapting to changing economic circumstances and social values. Much can be said for the piecemeal diffusion of new policies: when we do not know what we are doing, it is best not to do it everywhere, all at once. A state-based process facilitates gradualism and, therefore, feedback and institutional learning. Successful state and local experiments with airline deregulation, welfare reform, and school choice taught valuable lessons, built public confidence in innovative policies, and provided a testing ground for social scientists’ models and policy recommendations that might well have gone unheeded in a centralized political environment. State-based policy innovation also facilitates adaptation to local needs, circumstances, and preferences.
Political experimentation, however, does not operate with the efficiency of a disinterested, scientific process of trial and error. Such experimentation carries political risks, which arise principally from the fact that the governmental experimenters, and the interest groups that hang around them, have huge stakes in exploiting the test population of citizens. Legislative experimentation must therefore be constrained. To that end, we have a Constitution that allocates specified and limited powers to competing institutions and levels of government. The point of the constitutional arrangement is to limit what government may do to citizens in the way of experimentation; to inhibit rash, indiscriminate lawmaking; and to guard against the risk of factious, "partial" legislation, or what we now call rent seeking.
At some level, tension between constitutional constraint and political experimentation is inevitable. A prompt call for "balance" among those considerations, however, is a case of preemptive intellectual surrender. It is possible to harmonize constitutionalism and federalist experimentation while taking account of political risks. Also possible, however, is a view of experimentation that is inimical to constitutional purposes and oblivious to political risks.
Louis D. Brandeis’s view, unfortunately, was of the latter kind. His idea of experimentation was all trial—and never an error. Far from celebrating a genuinely diversified, experimental politics, Brandeis viewed state governments as a vanguard for the national administrative state.
New State Iceand Interests
The "laboratory" dictum appears in Brandeis’s dissent in New State Ice Co. v. Liebmann (1932), a case arising over an Oklahoma statute prohibiting the manufacture, distribution, and sale of ice without a certificate of public convenience and necessity. Then-extant constitutional doctrine permitted such conversions of private enterprises into public utilities vis-à-vis industries that were viewed as being affected with a public interest—for example, by virtue of a monopolistic market position. With respect to competitive industries, however, such measures were viewed as violating the Fourteenth Amendment—specifically, the liberty to apply one’s labor in an ordinary occupation, subject to reasonable regulation. Writing for the majority in New State Ice, Justice George Sutherland determined that selling ice was an ordinary private business and effectively enjoined the operation of the Oklahoma licensing statute.
Brandeis objected that intense public concern over the ice industry, "destructive competition," and the industry’s necessity to consumers and to Oklahoma’s economy might well warrant the licensing scheme. To reverse the state legislature’s judgment, Brandeis insisted, would "involve the exercise not of the function of judicial review, but the function of a super-legislature." The dissent culminates, and terminates, in the "laboratory" passage:
There must be power in the States and the Nation to remould, through experimentation, our economic practices and institutions to meet changing social and economic needs. I cannot believe that the framers of the 14th Amendment, or the States which ratified it, intended to deprive us of the power to correct the evils of technological unemployment and excess productive capacity which have attended progress in the useful arts.
To stay experimentation in things social and economic is a grave responsibility. Denial of the right to experiment may be fraught with serious consequences to the Nation. It is one of the happy incidents of the federal system that a single courageous State may, if its citizens choose, serve as a laboratory; and try novel social and economic experiments without risk to the rest of the country. This Court has the power to prevent an experiment. We may strike down the statute which embodies it on the ground that, in our opinion, the measure is arbitrary, capricious or unreasonable. We have the power to do this, because the due process clause has been held by the Court applicable to matters of substantive law as well as to matters of procedure. But in the exercise of this high power, we must be ever on our guard, lest we erect our prejudices into legal principles. If we would guide by the light of reason, we must let our minds be bold.1
The Supreme Court’s last decision to invalidate a state economic regulation as a violation of substantive due process, New State Ice represents an exceedingly deferential application of that doctrine. Oklahoma, like many other states, regulated prices and practices in the ice industry before turning it into a public utility. Justice Sutherland explicitly acknowledged the constitutionality of those regulations. He would have permitted even the licensing scheme had it rested on anything resembling a plausible rationale.
Brandeis’s dissent recites, at characteristic length, "facts" that could have justified Oklahoma’s experiment. (The "facts" were neither relied on by the Oklahoma legislature nor in the record before the Court; Brandeis cobbled them together from squawkings in the trade press and articles by protosocialist academics such as Charles A. Beard.) The low cost of entry into the ice market, Brandeis maintained, leads to "wasteful," "destructive," and "ruinous" competition. For the duration, consumers "suffer" [sic] because producers "go to extremes in cutting prices." Unable to recoup their fixed costs, Brandeis continued, some producers are forced out of business. Thus, "the business of ice . . . lends itself peculiarly to monopoly" and, presumably, monopoly pricing. That risk, though, apart from being unlikely in an industry with low entry costs and government price regulation, has nothing to do with the licensing scheme. As Justice Sutherland observed, and Justice Brandeis conceded, Oklahoma’s statute aimed to "create and foster," rather than to abate, "monopoly in the hands of existing establishments, against, rather than in aid of, the interest of the consuming public."
Pace Brandeis, Oklahoma was not a "single courageous state" whose "citizens chose" to stage a "novel experiment." It was rather the first state where beleaguered industry participants, threatened by intense competition and by a newfangled invention called the "refrigerator," managed to enact the oldest game in town—market entry and output restrictions in conspiracy against consumers and potential competitors. Justice Brandeis himself noted the industry’s "unremitting efforts, through trade associations, informal agreements, combination of delivery systems, and in particular through the consolidation of plants, to protect markets and prices against competition of any character." "The ice industry in Oklahoma has acquiesced in and accepted the Act and the status which it creates," he added—somewhat superfluously, since a member of that industry was the plaintiff in New State Ice and the defendant, a prospective competitor.
Neither those actual facts nor the plain absurdity of curtailing a threat of monopoly through monopolization gave Brandeis pause. That suggests not a "bold" mind but an unalterably closed one; not a healthy regard for facts but obtuseness to the political dynamics and dangers of state regulation.
The wholesale rejection of economic substantive due process in the wake of the New Deal earned the "Lochner Court" its notoriety and its contemporary opponents on the bench, Louis D. Brandeis and Oliver Wendell Holmes, their inflated reputations. The New State Ice dissent assumed the rank of a canonical statement of federalism’s innovative, experimental virtues. That interpretation, however—testimony to the lasting dominance of Brandeis’s progressivist ideology—is unsustainable. The New State Ice majority employed judicial review as a coarse screen to filter permissible, public-regarding experimentation from naked interest group dealing. What the dissent stands for is judicial abdication in the face of that spectacle.2
A credible commitment to state experimentation carries two necessary federalism implications. First, states must be limited to experimenting with citizens inside their own borders, rather than with out-of-state parties. Otherwise, the "laboratories" will routinely exploit outsiders for the benefit of their own constituents. Such mutual regulatory aggression will, ex ante, induce excessive experimentation and, ex post, obviate politically effective comparisons of costs and benefits.
Second, a plausible model of state experimentation requires limitations on national power. Without such limitations, states that are conducting costly social experiments will seek to prevent a deterioration of their competitive position and the flight of productive citizens to more hospitable jurisdictions by insisting that Congress induce all states to conduct the same experiment. Once that happens, we shall never know whether regulation or forbearance produces better results. Moreover, citizens and corporations lose the ability to move from one state to another in search of a regulatory package to their liking. The most reliable mechanism of institutional feedback and learning is lost.
By way of a current example, Texas, Pennsylvania, and Virginia, among other states, are conducting successful—if piecemeal—experiments with electricity deregulation. California, obviously, is not. Its citizens believe, and its politicians proclaim, that the state’s energy crisis stems from a corporate conspiracy, as opposed to the Golden State’s experiment with price controls and environmental restrictions on power plant construction. Those people deserve to sit in the dark. They will learn, though—perhaps when all the high-tech jobs and factories have moved to Nevada; probably sooner.
Learning from a failed experiment is the only benefit of the California fiasco and the most potent argument for sustaining state experimentation—even, in this instance, in the face of inevitable spillover effects on shareholders and energy users outside California. That benefit would be lost, and little could be said for federalism, if the state were permitted to impose the costs of its experiment on its neighbors or to engineer a federal bailout—to say nothing of contriving to prevent its citizens and companies from leaving the state. Effective experimentation and learning require institutional constraints that make the costs and benefits of political experiments hit home. Justice Brandeis did not believe in such constraints.
Across the Borders. Brandeis’s most fateful decision and opinion, Erie Railroad v. Tompkins (1938), wiped out the federal common law that governed private disputes among parties from different states and provided that such disputes must instead be governed by state common law. Three years later, in Klaxon Co. v. Stentor Electric (1941), the Supreme Court extended Erie’s "state-law-governs" rule to choice of law—that is, the question of which state law governs disputes among parties from different states.
While Erie looks like a profederalism decision, its effect is the opposite. To the Erie-Klaxon regime, we owe the fabulous "experiment" of systematic exploitation of out-of-state corporations in franchise disputes and, most egregiously, in products liability litigation. Plaintiffs’ attorneys shop for a hospitable court and jury, which proceed to apply their home state law and sock it to out-of-state defendants. That is not experimentation—quite the opposite: states lose their ability to experiment with varying liability regimes, since the most plaintiff-friendly jury in the country will determine product safety and liability standards for the entire country—provisionally, until plaintiffs’ lawyers find a still crazier jury.
Brandeis had left the bench before Klaxon and, in fairness, might not have joined that decision. For most of his career, he insisted on federal, constitutional limitations to prevent states from applying their own law to out-of-state parties and legal relations.3 But he would have approved of Klaxon’s spirit.
Brandeis burbled incessantly about the increasing complexity of the industrial age, which in his estimation necessitated a bigger government and "administrative machinery." "With the increasing complexity of society," he intoned in a 1918 dissent, "the public interest tends to become omnipresent."4 (In the ice industry, for example.) A serious consideration of complexity, however, should have produced the opposite result in Erie. One facet of complexity, the explosive growth of sales of products manufactured in another state, greatly increased the risk that parochial state courts might exploit out-of-state producers. For that reason, corporations liked federal courts and diversity jurisdiction. For precisely the same reason, Brandeis wanted to abolish federal diversity jurisdiction and, in Erie, committed corporate defendants to the tender care of state courts. His devotion was not to experimentation as a functional response to social complexity. What drove him was his animus against corporate capitalism.5
The States and the Nation. An early test of Brandeis’s commitment to state experimentation came in Hammer v. Dagenhart, a 1918 decision invalidating, as exceeding congressional powers under the commerce clause, the federal regulation of child labor. The purpose of the commerce clause, the Court explained, is to enable Congress to regulate interstate commerce, as distinct from flattening the diverse, experimental state world. None of the rationales that might warrant national intervention—say, a national emergency or an inability on the part of the states to tackle a social problem—was present. Every single state had at the time enacted prohibitions against child labor; the only question in the case was whether the federal minimum age (fourteen) should be permitted to trump North Carolina’s (twelve). Despite the evidence of competent, successful state experimentation, however, Louis "Laboratory" Brandeis joined Justice Oliver Wendell Holmes’s dissent in Hammer.
Almost two decades later, in Steward Machine Co. v. Davis (1936), the Supreme Court sustained a federal statute that enticed the states, through a "cooperative" system of taxation and funding, to adopt a uniform, federal unemployment insurance regime. The statute short-circuited an experimental dynamic that should have appealed to Brandeis. Wisconsin, on the initiative of Brandeis’s daughter, had adopted an unemployment insurance law shortly before the federal enactment, and other states were contemplating similar legislation. "We ought to get the full benefit of experiments in individual states before attempting anything in the way of Federal action," Brandeis had written in 1912.6 Heedless of that sensible exhortation, he joined Justice Benjamin N. Cardozo’s majority opinion in Steward Machine, which legitimated the federal statute on the grounds that state experimentation was not proceeding with sufficient speed.
State experiments, it turns out, are of no constitutional significance or consequence. Both experimentation, as in Hammer, and the lack (or lack of speed) thereof, as in Steward Machine, are a prelude and predicate for federal intervention.
While Brandeis mobilized federalism and state experimentation in opposition to the exercise of federal judicial power, what he actually believed in was social control and legislative supremacy at every level. Brandeis resolutely opposed Tenth Amendment limitations on the national government as well as judicially recognizable commerce clause or other enumerated powers limits to congressional authority. He generously read federal law to "preempt" state law, even when the two were not actually in conflict and, moreover, Congress had expressed no intent to override the states’ policies.7 If the states "wish to protect their police power," Brandeis argued, "they should, through the ‘state block’ in Congress, see to it in every class of Congressional legislation that the state rights which they desire to preserve be expressly provided for in the acts."8 His remark foreshadows the post-New Deal Supreme Court’s commitment to "process federalism," which holds that federalism’s only protection lies in the political process itself. That doctrine of judicial abdication, to which the Supreme Court adhered as late as 1985, is the credo of the centralized administrative state. It is the opposite of an experimental politics.
Speech, Trials, and Errors
Louis Brandeis’s enthusiasm for state experimentation was not only lukewarm; it was also selective. It was limited to "economic and social" matters, as his New State Ice dissent put it, and did not extend to speech, education, and "privacy." In Meyer v. Nebraska (1923), Brandeis voted to strike down, over a dissent by his soul mate Holmes, a statute that prohibited, to the untold relief of cornhusker kids, the teaching of German in public and private schools. In Pierce v. Society of Sisters (1925), Brandeis joined in striking down a state statute mandating public school attendance. In Near v. Minnesota (1931), he voted to invalidate a state statute prohibiting the publication of scandalous and defamatory materials.
Meyer and Pierce were substantive due process cases, just as New State Ice. Near incorporated the First Amendment into the due process clause. In each case the state invoked police power justifications that, however debatable, had greater plausibility than Oklahoma’s New State Ice scheme. Justice Sutherland, for one, saw the connection: his majority opinion in New State Ice cited Pierce and Near as precedents and wrapped them around Brandeis’s neck. The only reason for enjoining state experimentation in those cases while celebrating the Oklahoma racket, Sutherland suggested, was the contention that constitutional and common law rights count only when they are not economic.
Just so. Far from merely opposing substantive due process, Brandeis refused to recognize "economic" rights, even when they had a clear-cut textual basis—and, for that matter, when they ran against the federal government rather than against experimenting states.9 Meanwhile he detected, somewhere in the Fourth or Fifth Amendment, a constitutional aspiration "to protect Americans in their beliefs, their thoughts, their emotions and their sensations." (That blow of hot air appears in Brandeis’s celebrated dissent in Olmstead v. United States (1928), the perceived basis of a "right to privacy.") Such jurisprudence rested on Brandeis’s belief that free inquiry and "personal sanctities," in contrast to stultifying economic rights, were central to scientific management, social trial and error, and political and economic progress and must therefore be placed beyond the reach of experimenting governments.
Brandeis’s functionalism has the dynamics of rights and social change precisely backward.
Freeze! Brandeis rejected the judicial review of economic regulation as invariably premature. Regulatory experiments, he argued, should be allowed to prove their success or failure before being declared "arbitrary" or "unreasonable." The intended result of interest group politics, however, is often stasis rather than social experimentation. Oklahoma’s ice licensing scheme, for a perfect example, attempted to arrest the existing structures in a fiercely competitive, rapidly changing industry.
The conservative bias of political "experimentation" is systemic and utterly predictable: the emerging forces of change and progress are unorganized and underrepresented, whereas the dinosaurs have trade associations and friends in high places. Once a regulatory regime is in place, moreover, the beneficiaries will manage to sustain it regardless of its merits. The interest group detritus of the New Deal is with us to this day, from agricultural marketing orders to minimum wage laws. The price of uninhibited government experimentation is political and social ossification.
Toward Progress, March! Under the First Amendment banner planted in Near v. Minnesota, the Supreme Court later mowed down state libel laws, loyalty oaths, Pledge of Allegiance rituals in public schools, and state and local pornography regulations. The substantive due process of Pierce and the "privacy" right of Olmstead eventually became the substantive due process and privacy rights of Griswold v. Connecticut (1965) and Roe v. Wade (1973). Whatever one makes of the state regulation of birth control devices (as in Griswold) or abortion, though, it is most certainly experimental. If trucking and ice regulation present a case for diversification, gradualism, and institutional learning, so too do school choice, drug policy, and pornography regulation.
Social experiments enshrined in state legislation, moreover, are typically more easily reversed by political means than are economic experiments. Economic legislation is dominated by lobbyists and their well-heeled clients, who can easily defend their special interest schemes against a mass of rationally ignorant voters. In the social arena, political entrepreneurs can usually mobilize constituencies on both sides of issues that voters readily understand and, moreover, care about.10
Regardless of their merits in one case or another, then, universal rights to speech and privacy tend to reduce experimentation and diversity. That tendency is particularly pronounced when the commitment to universal noneconomic rights is instrumental and wedded, as it was for Brandeis, to an ideological notion of "progress." The judicial account of experimentation will then follow not the silly distinction between economic and noneconomic affairs but the more basic political commitments. Brandeis waxed about the right to "privacy" and suggested that the constitutional right to "life" must mean a meaningful life, including leisure time and a minimum wage. Having done so, he assented to state experiments with the forced sterilization of "imbeciles."
Louis D. Brandeis favored federalist "experimentation in things social and economic" as a means to progressive, statist ends. Even his hagiographers concede that Brandeis would have held a very different view of state economic experimentation and its judicial review had those experiments run against, say, trade unions.11
Modern justices have tended to overlook, or perhaps ignore, the instrumental and ultimately half-hearted nature of Brandeis’s federalist commitment. For example, they have quoted the "laboratory" dictum in the course of celebrating federalism’s virtues of diversity and attentiveness to local circumstances. Brandeis’s view of state experimentation, however, was entirely disconnected from those notions and instead emphasized its value as a step toward federal legislation. Similarly, profederalist justices have quoted the New State Ice dissent in opinions that reject, on Tenth Amendment grounds, federal impositions on state governments.12 Brandeis, as seen, did not believe in Tenth Amendment or any other constitutional federalism constraints.
One could easily live with an occasional out-of-context quotation. What distresses is the modern Supreme Court’s sustained Brandeisian tendency of subordinating federalism to progressive dictates and statist presumptions. The Court has empowered and protected state governments through creative interpretations of the Tenth and Eleventh Amendments. It has, however, refrained from resurrecting constitutional doctrines—foremost, a robust enumerated powers doctrine—that would discipline state governments by forcing them to compete for productive citizens. On the rare occasions that the Court has limited enumerated powers, it first reassured itself that the states can and will in fact regulate the problem at hand—gun possession on school grounds or sexual violence.13
On issues that we now call "social," the Court acts as a superintendent of experimentation. Untoward experiments, such as operating an all-male college, are verboten. Experiments of the right kind are not; in a way they are affirmatively required. If states fail to liberalize, with sufficient speed, laws governing sexual and life-and-death matters, the Supreme Court will move them along; witness Roe v. Wade.
Disagreements between the liberal and moderate justices chiefly concern the desirable rate of social progress, not the Court’s custodial role. If forty-nine states have abolished obstacles to terminating the lives of the permanently comatose, the speed-loving Justice John Paul Stevens views that as an excellent reason to seize the all-purpose due process club and to beat the last laggard—Missouri—into compliance. On the other hand (or perhaps the same hand), the First Amendment should not bar New Jersey from doing to the Boy Scouts what Oklahoma did to ice merchants—that is, turn them into a public enterprise. The Supreme Court, Stevens opined in a Brandeis-quoting dissent, should set aside constitutional concerns over an experiment so much in line with prevailing state efforts to overcome the "atavistic" sentiments that sustain private discrimination against homosexuals. The more patient Justice Sandra Day O’Connor explained in a Brandeis-quoting concurrence in a pair of "assisted suicide" cases that she would stay the Court’s hand in life-or-death matters—provided that the evidence shows sustained, conscientious state experimentation of the right kind.14
One cannot simultaneously believe in an open, experimental, federalist politics and in a comforting, social-democratic notion of progress; in institutional trial-and-error and in legislative supremacy and unconstrained interest group politics; in state experimentation and in the judicial enforcement of progressive moral sentiments. Real federalism requires confidence in the creative energies of a free society; a healthy suspicion of interest group schemes; and a willingness to tolerate indeterminacy and variegated results. The real Constitution allows for and in fact enshrines those premises. The Constitution according to Brandeis does not.
If ever we get a Supreme Court that trusts an experimental, federalist politics, that Court is unlikely to cite the New State Ice dissent. It is much more likely to cite George Sutherland’s majority opinion—and perhaps, that underrated justice’s cri de coeur about his progressive foe and brother: "My, how I detest that man’s ideas."
1. 285 U.S. 262, 311.
2. Funny that this should be true of many other decisions that form the foundation of the post-New Deal Court’s jurisprudence—for instance, the famous 1938 decision in United States v. Carolene Products, 304 U.S. 144 (1938), which established the Supreme Court’s "dual standard" of judicial review (none for economic rights, plenty for a few other rights and especially those of "discrete and insular minorities"). See Geoffrey P. Miller, "The True Story of Carolene Products," 1987 Supreme Court Review (1987), p. 397.
3. See Edward A. Purcell, Jr., Brandeis and the Progressive Constitution (New Haven: Yale University Press, 2000), pp. 151-53. Similarly, Brandeis was prepared to use the "dormant" commerce clause as a means of invalidating state interferences with interstate commerce.
4. International News Service v. Associated Press, 248 U.S. 215, 262 (1918).
5. For an excellent account of anticorporate sentiments as the basis of Brandeis’s Erie Railroad opinion, see Purcell, Brandeis and the Progressive Constitution, pp. 141-64.
6. Melvin I. Urofsky and David W. Levy, eds., Letters of Louis D. Brandeis, vol. 2 (Albany: State University of New York Press, 1972), p. 640 (July 8, 1912, letter to Mary McDowell).
7. Modern constitutional law requires a "clear statement" of congressional intent before the Supreme Court will infer federal preemption. Justice Brandeis, who seems to have coined the term preemption, held a far more expansive view. In Gilbert v. Minnesota, 254 U.S. 325, 334 (1920), for example, Brandeis dissented when the Court sustained a conviction for antiwar propaganda under a state statute. While the Gilbert dissent is widely read as a paean to the First Amendment, it actually rests on the proposition that Congress, in the exercise of its powers to raise armies and to declare war, had opted for a volunteer army—and that state injunctions against pacifist propaganda would interfere with the recruitment of fully informed volunteers. While Congress was free to bar state interferences with antiwar pamphleteering, it had not actually done so. To Brandeis, the deafening silence indicated comprehensive federal field preemption.
8. Urofsky and Levy, eds., Letters of Louis D. Brandeis, vol. 5 (Albany: State University of New York Press, 1978), p. 247 (January 6, 1926, letter to Felix Frankfurter).
9. For a good example of Brandeis’s restrictive view of textual constitutional guarantees of "economic rights," see his lone dissent in Pennsylvania Coal Co. v. Mahon, 260 U.S. 393, 416 (1922). For a Brandeisian endorsement of federal expropriation, see the so-called Gold Bond Cases, 294 U.S. 330 (1935), where five justices, including Brandeis, sustained the federal government’s scheme to repay the holders of gold-backed bonds in devalued dollars. The secretary of the Treasury proudly proclaimed that the U.S. government had managed to relieve the investor class of $2.8 billion, which—little is new under the political sun—would be deposited in a lockbox for future payment of the national debt. To Brandeis, it was just another good day at the federal lab.
10. To be sure, the dichotomy between interest group—driven economic policies and entrepreneurial social policies exists on a continuum. Notably, coalitions of "bootleggers and Baptists" may be able to enact policies with redistributive and "moral" aspects (blue laws, in the classic case) that neither constituency, acting alone, could obtain. See Bruce Yandle, "Bootleggers and Baptists—The Education of a Regulatory Economist," Regulation, vol. 7 (May/June 1983), p. 12. Those mixed and intermediate cases, however, do not invalidate the general observation in the text.
11. See, for example, Philippa Strum, Brandeis: Beyond Progressivism (Lawrence: University Press of Kansas, 1993), p. 89.
12. For miscites of the New State Ice dissent in support of diversity and local variation, see, for example, Roth v. United States, 354 U.S. 476, 505 (1957) (Justice John M. Harlan dissenting), EEOC v. Wyoming, 460 U.S. 226, 264 (1983) (Chief Justice Warren E. Burger dissenting), and Chandler v. Miller, 520 U.S. 305, 324 (1997) (Chief Justice William H. Rehnquist dissenting). For misciting Brandeis in support of Tenth Amendment limitations on national power, see, for example, FERC v. Mississippi, 456 U.S. 742, 789 n. 20 (1982) (Justice Sandra Day O’Connor concurring and dissenting in part) and Garcia v. San Antonio Metropolitan Transit Authority, 469 U.S. 528, 567 n. 13 (1985) (Justice Lewis F. Powell, Jr., dissenting). The charming misunderstanding evident in those passages is more readily explained than the winning entry in the misquote-Brandeis sweepstakes—Justice Harry A. Blackmun’s curious suggestion that federal impositions on the states somehow promote experimentation. Garcia v. San Antonio Metropolitan Transit Authority, 469 U.S. 528, 546.
13. See, respectively, United States v. Lopez, 514 U.S. 549, 581 (1995) (Justice Anthony M. Kennedy concurring) and United States v. Morrison, 529 U.S. 598, 618 (2000).
14. The cases and opinions mentioned in this paragraph are Cruzan v. Director, Missouri Department of Public Health, 497 U.S. 261, 330 (Justice John Paul Stevens, dissenting) (but see Justice Sandra Day O’Connor’s concurrence, which cites the New State Ice dissent, 497 U.S. 261, 292); Boy Scouts of America v. Dale, 120 S. Ct. 2446, 2459 (2000) (Justice John Paul Stevens dissenting); and Vacco v. Quill and Glucksman v. Washington, 521 U.S. 702, 737 (1997) (Justice Sandra Day O’Connor concurring). In Dale, Chief Justice Rehnquist responded to Justice Stevens’s venomous dissent that "Justice Brandeis was never a champion of state experimentation in the suppression of free speech." 120 S. Ct. 2446, 2457. While that is technically true, the Stevens dissent is true to the spirit of Brandeis’s jurisprudence.
Michael S. Greve is the John G. Searle Scholar and the director of the Federalism Project at AEI.