The First Amendment was once thought of as the law of the political underdog. Its archetypal beneficiary was the humble pamphleteer whose unpopular ideas eventually gain majority support. From this narrative arose the original and still most powerful justification for First Amendment review: the protection of political debate and the democratic process, which powered Oliver Wendell Holmes’ famous dissent in Abrams v. United States as well as New York Times v. Sullivan’s assertion that “debate on public issues should be uninhibited, robust, and wide-open.” And today, in tough cases, the Court continues to rely on democratic process as a justification for striking down enacted laws.”

Today, however, the First Amendment’s role in the American political process has changed decisively. It can longer be described as a law that protects unpopular speakers or other politically weak actors in the Carolene Products sense. If the First Amendment could once be described as a remedy for defects in the political process, it has now as often become the cause of such defects. For today’s First Amendment is regularly deployed not to promote or facilitate political debate but to end it. Across broad areas of public regulation, the judiciary has intervened to shut down active political debate in the fields of privacy, telecommunication, securities, false advertising, and health and safety regulation, among others.

In such cases, in process terms, the First Amendment is playing a very different role than envisioned in cases like Sullivan. It cannot be described with a straight face as enabling political debate. It is, instead, a tool of regulatory leverage used by politically powerful groups. And, at the risk of stating the obvious, this newer role played by the law is in tension with its goal of promoting democracy; yet there is no doctrinal tool specifically designed to prevent the First Amendment from biting its own tail.

As a means of addressing the gap, this paper suggests an “anti-circumvention” principle that may be usefully invoked in certain First Amendment cases. The principle would be used by the judiciary, in its discretion, to limit the use of the First Amendment to circumvent an ongoing and functioning political debate. More precisely: In cases where the underlying law does not censor political speech, nor arise from majoritarian prejudice against a despised or unpopular speaker, and particularly where the political debate is in progress, the judiciary should avoid using the First Amendment to give one side of the debate a judicially granted circumvention of democratic politics.

“First Amendment opportunism,” Fred Schauer’s phrase, well describes the cases that call for application of the anti-circumvention principle. Practically speaking, this doctrine would call for a different outcome in close cases like Sorrell v. IMS Health or United States v. Playboy Entertainment Group, Inc., where the defendants sought to overcome their defeats in highly contested political processes at the state and federal levels, respectively. Far from the classic model of a politically disenfranchised speaker with an unpopular message, these are cases where politically well-represented subjects lost legislative battles, and turned to the First Amendment as an alternative.

Some of the cases I’ve mentioned might also be criticized as “First Amendment Lochnerism,” so it might be useful to clarify what a process-based approach adds. I am sympathetic to both the label and the critique but believe that a process approach offers different insights. The Lochnerist critique rightly points out that using the First Amendment to protect economic liberties parallels the judicial abuses of power practiced by the Supreme Court in the early 20th century. But the process approach offers a broader assessment of when First Amendment review is justified or not, and is less tethered to the commercial-speech doctrine. It asks courts to broadly consider if their own judicial intervention threatens the democracy-promoting values of the First Amendment itself. And while this short essay cannot hope to fully describe the interactions of democratic process and the First Amendment, I hope it may make a start.

I want to stress that what I am suggesting here is not a rejection of constitutional review but a discretionary limitation based on an overriding interest in democratic process. In New York Times v. Sullivan, the Court asserted that the First Amendment should be enforced to promote “free political discussion to the end that government may be responsive to the will of the people.” This paper asks reviewing courts to examine carefully whether a litigant’s claim would actually facilitate “free political discussion” or do the opposite. Reduced to a sentence, what this paper asks a reviewing court to do is something familiar: examine the political context in which the case arises and then ask whether it stands on stronger or weaker grounds when it decides whether to intervene.

Should this seem too vague, consider the following three situations in which a court might be asked to strike down legislation or a regulation under the First Amendment. In the first, the judge is asked to enforce the Amendment to protect a litigant’s expression of political ideas or views. In such a case, the judiciary is at the zenith of its legitimacy, engaged in an Abrams- or Sullivan-based defense of democratic process by protecting the expression of ideas and opinions that might not otherwise be heard. Such heightened review may also be justified if the judge is being asked to protect the speech of a widely despised or poorly represented group expressing unpopular ideas (in shorthand, a Carolene Products context ). Overbreadth claims can fall into this category, especially if they affect bystanders to the political process.

A second category includes judicial intervention that is not directly related to the enhancement or abrogation of the political process. Here the judiciary is typically protecting social, aesthetic, moral, commercial, or other ideas and experiences. Judges may rely on justifications that are unrelated to protecting the democratic process but are instead related to the speakers’ liberty interests or the listeners’ interest in receiving information.

The third category — the focus of this paper — comprises cases that are an evident effort to use the First Amendment either to abrogate an ongoing political debate or to circumvent a political defeat. A prerequisite is the presence of a fully and fairly contested policy debate by well-represented parties, possibly on a state-by-state basis, where the judiciary is asked to intervene and put an end to the political process. Here the judge should understand that he or she is at a nadir of legitimate reviewing power, given that his or her own action will terminate the kind of political debate the First Amendment was designed to promote.

Such considerations are not always easy: far simpler for a court just to pretend that cases arrive without a political history or that every party before the court is in the position of the oppressed pamphleteer humbly seeking justice and vindication in the courts. But that is to ignore political reality. In most First Amendment cases, the courts now face plaintiffs whose legislative losses are not silencing but are part of a larger, ongoing political process where the judicial intervention simply creates new winners and losers. Given that reality, it is willful blindness to pretend that enforcement of the First Amendment is always an aid to democracy.

I. The Majoritarian Justification for an Anti-Majoritarian First Amendment

A major premise of this paper is so obvious as to be hiding in plain sight: that enforcement of the First Amendment has long been justified as a remedy for failings in the American political process. I don’t want to suggest that this is the only reason courts enforce the First Amendment, but it has historically been the most important reason and one that retains a commanding currency in our time.

The link between the First Amendment and political process originates with the well-known judges who invigorated the First Amendment in the first half of the twentieth century. These judges — Holmes, Brandeis, Stone, and Hand — viewed it as a way to protect the democratic political process from becoming warped by government censorship or punishment of disfavored speakers. That is what made it possible for strong believers in majoritarian democracy like Holmes and Hand to author the founding opinions in the First Amendment tradition. “It is noteworthy,” writes Vince Blasi, that these judges “went out of their way to avoid recognizing a free-standing individual right of expressive liberty that exists apart from and thereby limits the principle of majority rule.”

The majoritarian defense of First Amendment review is best captured in Holmes’ famous dissent in Abrams, a case where anti-war leaflets were distributed by being thrown out of a window. Holmes’ dissent sought to justify judicial review on strong grounds: as the protection of the democratic process itself. Holmes believed that political truth is both unknowable and changing; that majority opinion is essentially dynamic in nature, for “time has upset many fighting faiths.” The best approximation of truth, he believed, came out of a marketplace of ideas; hence the First Amendment should protect an open competition of ideas that might help a nation find the answers to its most pressing problems.

These ideas remain a mainstay of First Amendment doctrine. They were strongly affirmed in New York Times v. Sullivan, where the court protected a strident and accusatory political advertisement from a libel suit so as to enable “free political discussion to the end that government may be responsive to the will of the people.” And decades later, the country remains on an ongoing search for political answers to its hardest questions, and the Supreme Court continues to rely upon a process justification in many of its hardest cases. Take, for example, the Citizens United decision, which struck down a restriction on political speech by corporations. Facing a strong dissent, the majority insisted that the decision was justified to afford the strongest possible protection for the democratic process, given that “speech . . . is an essential mechanism of democracy.” Moreover, “the right of citizens to inquire, to hear, to speak, and to use information to reach consensus is a precondition to enlightened self-government.”

To be sure, Abrams/Sullivan–stylefacilitation of the democratic process does not exhaust the reasons the judiciary has used to justify its interventions. Even among the early cases, the famous Whitney concurrence presented a broader assessment of how freedom of thought and speech serves democratic values. And over the years, the court has found reason to give protection to speech having nothing to do with political debate but instead touching on artistic, commercial, or cultural expression. There is no use or reason here to dismiss the many other reasons, highlighted by scholars like Robert Post and Jack Balkin, that freedom of speech might be a valuable part of a democracy broadly considered. But in this paper I want to focus not on the periphery of the First Amendment but its core and point out that the current view of how the First Amendment interacts with the political process is both dated and two-dimensional.

The point can be put this way. The broadening of First Amendment protection to matters such as artistic or commercial speech certainly has its own set of justifications, best described as rooted in concerns for liberty. But the under-recognized danger is that such broader protections can set the liberty- and democracy-promoting justifications of the First Amendment against each other, as judicial intervention comes to undermine the larger goal of promoting democracy and political debate. I’m not saying that the First Amendment’s goal of promoting democratic debate should automatically trump the protection of liberty, but it cannot be that it counts for nothing. This is what I mean when I say that the First Amendment needs to be prevented from biting its own tail.

There is no better way to demonstrate that danger than looking more carefully at a few cases. As we shall see, there is a striking dissonance in a Court’s willingness to glorify open political debate as it intervenes to stop such debates in their tracks.

II. From Enabling Debate to Ending It

Privacy Regulation

In the early 2000s it was first widely reported that commercial chain pharmacies — Rite-Aid, CVS, and so on — were selling private prescription records without the consent of doctors or patients. Pharmacies are generally required by law to store an enormous amount of private prescription data. The laws banned the sale of patient data but not the prescribing information of doctors. Sensing an opportunity, the pharmacies began selling their records to marketing firms, who in turn created marketing profiles to help drug companies sell to doctors.

Doctors were generally unaware that they were being profiled, but when knowledge of the practice became public, a number of state and national doctors’ associations (but not the American Medical Association) condemned the practice as both unethical and a violation of privacy. The states would later claim that the practices raised their public insurance costs by encouraging doctors to prescribe more expensive drugs. New laws seeking to ban the resale of prescribing data were introduced in 29 states around the country. Publicity campaigns and outreach surrounding the practices generated phone-in and email campaigns.

Facing the prospect of new regulation, the pharmacies, the pharmaceutical industry, and data-mining firms fought back. They portrayed the privacy concerns as overblown and instead stressed the benefits of data mining. The influential American Medical Association intervened on the side of the drug companies arguing that “restrictions on the use of prescription information could disrupt health care research and its corresponding benefits for patients, government agencies, health planners, academicians, businesses and others.” Across the country, lobbyists for industry actively discouraged passage of the laws as a misguided attack on the future.

There was, in short, a robust political debate, one that could be used to illustrate how the American democratic and federal processes are supposed to work. The press discovered and publicized the practice. Agitated citizens and affected groups turned to their local representatives, who proposed legislative action. The industry and other parts of civil society, like the AMA, fought the laws with strong arguments stressing the benefits of practices that looked unattractive on their face. The debate, moreover, proceeded on a state-by-state basis, following Brandeis’ dictum that “a 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.” States that enacted new privacy laws were engaged in an experiment that might inspire others to follow — or, in time, prove a costly folly that others could learn from.

A few other matters are relevant to the analysis. No one could claim that the new laws were designed to silence political debate. Nor can it plausibly be argued that the pharmacies or data miners were a group unable to use the political process to defend themselves; there was nothing that might “curtail the operation of those political processes” meant to protect them, in the words of Carolene Products. In fact, with their large lobbying budgets, the opponents were arguably overrepresented, even if they faced worthy opponents in the New England medical societies.

So what happened? On a state-by-state count, the pharmacies won the political battle in more states than they lost, managing to stop or at least delay passage of a new privacy law in most of the states. But such laws did successfully pass in New Hampshire, Vermont, and Maine, thanks mainly to the efforts of local medical societies. The Vermont society would later write that it “spent much of the 2007 Vermont legislative session championing the law.”

However, something of real importance did happen during the legislative process. Vermont and New Hampshire each decided to include exceptions to the ban on the sale of private prescribing data, as a response to some of the benefits stressed by the AMA, industry groups, and universities, among others. Most importantly, the Vermont legislature put in exceptions for “health care research,” for “educational purposes,” and as made necessary for law enforcement or other laws. Introducing such exceptions is, of course, commonplace in legislative process and might be considered good practice; yet, as we shall see, the exceptions ended up being a poison pill for First Amendment purposes.

Medical privacy is obviously a matter of great and ongoing public concern, but this particular political debate was interrupted by the courts. For after losing in the New England legislatures, the opponents of the laws, with the help of counsel, reimagined themselves as politically disadvantaged speakers and, in their federal complaint, argued that their speech was burdened by the new laws. They made two arguments: first, that the sale of prescription data was a form of speech, unfairly burdened by the new law, and second, that the new privacy laws were a discriminatory punishment that should be subject to the strictest constitutional scrutiny.

The First Circuit threw out the New Hampshire case, with Judge Seyla, for the court, asserting that the pharmacies were really in no different a position than the sellers of any other commodity. “The plaintiffs, who are in the business of harvesting, refining, and selling this commodity, ask us in essence to rule that because their product is information instead of, say, beef jerky, any regulation constitutes a restriction of speech. We think that such an interpretation stretches the fabric of the First Amendment beyond any rational measure.” A second panel of the First Circuit agreed. But the Second Circuit accepted the simple formula described above: that the sale of private data by pharmacies was “speech” that was burdened by the new law. The state’s interests, the court decided, were insufficiently “advanced” to justify the laws ; hence, the law was unconstitutional.

The Supreme Court took the Vermont case and struck down the state’s law based on a different legal theory: discriminatory censorship. The Court seized on the exceptions put into the law for “research,” “education,” and “law enforcement” to claim that the data miners were being selectively punished based on the type of speech they were engaged in (i.e., marketing to doctors, as opposed to research). More precisely, the Court declared the law a “content- and speaker-based” burden that was “aimed” at pharmacies, pharmaceutical companies, and data miners.

It is not very hard to criticize the Supreme Court opinion. Any regulatory process, as Justice Breyer noted in dissent, necessarily involves making distinctions among regulated parties. The legislature had included the exceptions for academic research based on the industry’s own arguments. Oddly enough, a total ban on all transfer of data would have probably been a worse law, yet might have survived scrutiny, at least under the Supreme Court’s approach.

But in this essay I focus on a different problem, which is the Court’s complete indifference to its own role in the political process. Nowhere in the Supreme Court’s elevation of data miners into speakers and the privacy laws into censorship is there any hint of concern that the judiciary was about to put an ongoing political debate on ice. Nor did the court seem to feel any discomfort about a federal judiciary interfering with what might be thought of as a matter of state regulatory prerogative. Nowhere, finally, is it mentioned that the pharmaceutical companies were, in fact, the losers of a fairly contested democratic debate, and that the exceptions in the law were a byproduct of that process

A large and embarrassing gap between rhetoric and result is a byproduct of this deliberate indifference. The Supreme Court relies on New York Times v. Sullivan, a case that stands for the idea that public debate should be “uninhibited, robust, and wide-open” as it proceeds to end a debate that was, in fact, quite robust. A case that upheld the prosecution of protestors burning draft cards is recruited, counterintuitively, to suggest the relevance of censorial intent. The Court also manages to suggest that nude dancing or even a flag burned in protest — an unpleasant, though undeniably political message — might be more readily regulated than doctors’ privacy.

The dissent, authored by Justice Breyer, was not wrong to stress that the law should have been reviewed by a standard appropriate for economic regulation, nor to point out the dangers in the “constitutionalization of economic theories preferred by individual jurists.” But it could have been made even stronger by stressing the themes developed here. It could have returned to Abrams and Sullivan for the viewthat the First Amendment is meant to promote robust political debate and then observe that the Court, by its own actions, had just ended such a debate. It might have emphasized the Court’s indifference to its own endorsement of the states as “laboratories of experimentation” for difficult matters of public policy. The dissent, in other words, should have submitted into evidence the fresh corpse of the democratic process to establish the severely anti-democratic nature of the majority’s actions.

Regulation of the Pornography Industry

If Sorrell represents First Amendment interference with state lawmaking, consider more briefly two cases involving the same federal legislation: Reno v. ACLU and Playboy v. Federal Communications Commission. In each case the Court confronted “family values” laws passed during the far-reaching legislative process that resulted in the 1996 Telecom Act. As a part of that Act, Congress passed new statutes governing Internet pornography sites (then new) and also adult-entertainment cable channels like Playboy and Spice. The democratic process approach described here suggests that the Court got the decision right in Reno but wrong in Playboy.

At issue in Reno was the Communications Decency Act, which required that all indecent internet content be placed behind an effective age-verification screen. Pornography sites, under the law, would be required to make certain that their users were over the age of majority. In Playboy, a new law required that adult entertainment cable channels (like Playboy, Spice, and so on) “fully scramble or fully block” their channels for non-subscribers, replacing a flickering screen offering occasional glimpses with a plain blue screen. The law was a response to parental complaints of “signal bleed” — that is, sexual content becoming visible to non-subscribers.

A superficial take might see the laws as similar: both purported to regulate new pornography industries by creating clearer and stronger barriers to prevent access by children. But, from a political perspective, there were important differences. The Court in Reno, upon careful examination of an extensive district-court record, concluded that the law was hopelessly overbroad. It ruled that the law might require the lockdown of scientific sites, educational sites, and so on: the law “unquestionably silences some speakers whose messages would be entitled to constitutional protection.”

Reno might have been a closer case were it somehow crafted only to deal with the Internet pornography industry. But in striking down the law, the Court was actually defending the rights of parties unrepresented in the political process: the unknown multitude of speakers who might be chilled by an overbroad law. Under a Carolene Products logic, such unknown and disparate groups are parties who may lack the ability to seek the protections of the political process.

In contrast, the law in Playboy was far narrower in its scope. There was little or no question of accidental censorship of unrelated speech. And Playboy and other members of the pornography industry were the ones who fought the law in Congress, and it was their rights as speakers that were at issue. The Court nonetheless struck down the law based on the idea that there were “less restrictive alternatives” to what Congress did. The opinion is weak for reasons described in the dissent (most clearly that the less restrictive alternative, home filtering, is obviously far less effective ). But the dissent could have been buttressed by pointing out that the industry was seeking to use the First Amendment as an alternative to politics, that they had obtained a legislative compromise but wanted more. For the Playboy case was not a case of political censorship, nor one where the regulated parties were unable to use the legislative process. Nor was Playboy’s speech actually being banned or censored; the industry was being asked to install better filters (as over time, as technology changed, they did). This is why the Playboy case, if superficially similar to Reno,was quite different, politically speaking. For at bottom, it represented a well-represented industry’s effort to knock out an unwanted regulation, as opposed to a law that actually threatened the democratic process in any serious way.

To be sure, in its favor, companies like Playboy might have stressed that they historically have been subject to a prejudice, one that has weakened their ability to get a fair hearing in the legislature. Yet it should have also mattered that the law was a democratic compromise: It didn’t ban pornographic cable channels but sought to restrict their viewing to actual subscribers. Given that fact, the Court should have recognized its own weak position in terms of reviewing power. Instead, the Playboy corporation, just like the data miners, became the humble pamphleteer, simply hoping to catch the ear of an uncaring world.

III. Elaboration and Questions

Doctrinal Elaboration

There can be little disagreement that the role originally imagined for the First Amendment in political debate does not resemble the role that the Amendment now plays in the political process. Suggestive of this fact is an empirical study by John Coates finding that the anti-regulatory usage of the First Amendment by business now exceeds its usage for any other purpose. In this last section, the goal is to further specify the approach here introduced and to answer some of the most obvious questions.

The introduction specified a three-category hierarchy of scrutiny under a process-driven First Amendment review. In brief, the first category, the highest degree of scrutiny, are cases of political speech, Abrams/Sullivan cases, already widely understood to be at the core of First Amendment protection. A second category includes cases where judicial intervention represents neither a protection of the political process nor its abrogation. Heightened review might be driven by the observation of obvious defects in the underlying political process or some inherent reason to doubt that those whose rights were burdened had the capacity to avail themselves of a political remedy (Carolene Products). The third and final category are cases like Sorrell, where a court is asked to intervene on behalf of parties fully and fairly represented in the underlying political process and by its intervention interrupt an ongoing political debate.

This hierarchy raises a series of obvious questions. First, let me specify that it does not mean that judges would never be justified in using the First Amendment to strike down a law that was the byproduct of a fairly contested political process. As already stated, a law that bans political speech is still interfering with the democratic process even if it was fairly arrived at. Hence, a full debate that led to a law banning members of the Communist Party from presenting their views would still be subject to strict scrutiny. A law could be overbroad and thereby censor parties outside the political process, as in Reno v. ACLU. And a law that emerged from a fully contested process could nonetheless be invalidated if it was irrational — the court still has the duty to “determine the reasonableness of the legislature’s belief in the existence of evils and in the effectiveness of the remedy provided.”

Might this give parties reason to deliberately avoid participating in politics so as to preserve the strength of their First Amendment claim? The point is an interesting one, for it would seem counterproductive to discourage political participation. Yet as a practical matter it seems it would be relevant only at the margins. True underdogs — those entirely unrepresented in the political process (like random citizens) -- would, by definition, be unaffected. We might also distinguish the cases of groups like the EFF or the ACLU who generally promote the rights of others, as opposed to their own interests.

However, the politically powerful and well represented would sometimes face a choice — arguably a fair choice — to devote their efforts to politics or litigation instead of taking advantage of both. The law is full of doctrines designed to prevent similar forms of opportunism, such as estoppel and preclusion doctrines, or those that require the exhaustion of state or administrative remedies. Those doctrines respond to the unfairness of allowing a second bite at the apple. There is a similarly evident unfairness in allowing well-financed interest groups to hash things out in the legislature, blocking legislation or gaining whatever concessions they can, and then, only if they lose, asking the courts to intervene — using, without irony, a part of the Constitution meant to facilitate political debate.

Would this proposal give the judiciary the impossible task of assessing the validity of an underlying regulatory or legislative process? The task is hardly impossible: The judiciary is already in that business. Courts are constantly assessing the process due and also assessing whether parties are trying to re-litigate previously decided matters under a new heading. This is particularly true when a case involves both the Administrative Procedure Act (APA)— or state equivalents — and First Amendment review. In such cases, the judiciary is already called upon to evaluate the underlying process that yielded the rule in question.

There is reason to think that the D.C. Circuit has in fact already shown some impatience with First Amendment challenges tacked onto APA challenges. For example, in the telecommunications field, where every regulated party is plausibly a “speaker,” the opponent of any regulation can always bring some kind of First Amendment claim. However, the D.C. Circuit has come to reject such challenges, perhaps because it’s so obvious that they are an effort to circumvent the underlying regulatory process. The most recent example was the effort to negate the net neutrality rules enacted in 2015, in which sellers of broadband services claimed that the right to block or discriminate among speakers was a form of editorial speech. The Court dismissed the argument without difficulty. The unstated reason, perhaps, was that given that this was the FCC’s third effort to write a net neutrality rule in the midst of a very public debate over the rules, it was obvious that deciding the matter on First Amendment grounds would be an extreme arrogation of power to the judiciary over a matter of widespread public concern.

Beyond Lochnerism

Over the last decade, scholars and dissenting judges have critiqued parts of the Supreme Court’s First Amendment jurisprudence as “First Amendment Lochnerism.” The critique suggests that the protection of commercial speech has become a means for the judiciary to strike down economic regulation at will, creating a contemporary equivalent to the substantive due process theories relied upon by the Court in New York v. Lochner.

I am very sympathetic with the criticisms that fall under the Lochnerist label. But I want to specify how a process-driven approach to the First Amendment might yield complementary insights. A shared insight, already described in some detail, is that overzealous First Amendment enforcement is actually at war with the democracy-promoting purposes of the Amendment. To be sure, some of the Lochner critics, like Amanda Shanor, have argued that the Court may “render self-government impossible.” But the process approach affords is a more precise means of condemning interference with democracy on the First Amendment’s own terms.

The outcomes produced are also slightly different. The critics of Lochnerism would essentially call for more relaxed scrutiny of economic regulation that can be labeled, by a clever attorney, as a burden on speech or a form of viewpoint discrimination. But the process-driven approach embraces a broader set of speech categories, including the non-economic. It also differs by offering the government an affirmative defense: that the plaintiff had a political remedy that it did not avail itself of. The results may often be the same — namely, the upholding of the law in question. But the reasoning is different, less grounded on the idea that there is an intrinsic and natural difference between economic regulation and everything else.

Unlike the Lochnerist critique, the process approach also provides further insights as to when the judiciary should feel it is empowered to strike down legislation using the First Amendment. It is tied to the idea of pointing to obvious defects in the underlying political process, the original rationale for enforcement of the First Amendment. This isn’t to say that process is all that matters in First Amendment cases, but it should not be ignored either, as it usually is.

In the end, the Lochnerist and process approaches have the same villain, namely, unjustifiable judicial intervention to strike down democratically enacted legislation. As critiques of the Court’s jurisprudence, they are fellow travelers. But the Lochnerist is more inclined to suggest, based on historical reasons, that economics should be left to the people, while the process approach demands that a court justify its interventions in the absence of political failure.

More Fundamental Objections

So far I have presumed that the legitimacy of judicial intervention is at least partially a byproduct of its role in protecting the political process. But there are some who might think the entire project of grounding First Amendment as a defense of political process has always been misguided. A lesser version of this critique, based on Laurence Tribe’s well-known critique of John Hart Ely, would suggest that process-driven theories must ultimately become substantive anyhow. The broader critique, rooted in natural law theory, takes the liberties protected by the First Amendment as absolute and essentially free-floating, making the discussion of process irrelevant.

These objections take us into a much broader debate surrounding Constitutional interpretation than I am able to engage in this essay, but I wish to offer at least two responses. The narrower response is doctrinal and historical. It points to the specific history of the First Amendment, which was activated by the judiciary in the twentieth century in order to protect the political process. In other words, the Abrams/Sullivan concern for open political debate is too fundamental to First Amendment jurisprudence to be ignored. What I’m insisting is that the foundational concern for political process does not just compel judicial action but also forbearance.

For those who take the First Amendment as a recognition of a natural right, the thoughts of Justice Holmes and the reasoning of Sullivan might as well be so much background noise. But those who have rejected the principles of majoritarian democracy and the consent of the governed must at least recognize the inherent fragility of their approach. As John Hart Ely implicitly predicted, only those constitutional rights explicitly enumerated or reinforcing of representative democracy have tended to survive over the longer term. A broad and free-floating First Amendment lacks any democratic justification better than Lochner’s. Hence it is not hard to predict that the parts of the First Amendment that will survive over the long term are only those consistent with the idea that the Amendment is an aid to majoritarian democracy, not its enemy. For the Amendment was, as the Court put it, "fashioned to assure unfettered interchange of ideas for the bringing about of political and social changes desired by the people."