The Yale Law Journal

March 2016

Professional Speech

Constitutional LawFirst Amendment

abstract. Professionals speak in the course of exercising their profession. At the same time, the state can regulate the professions. What is the permissible scope of regulation of the professions as distinct from regulation of professional speech? This Article provides a comprehensive account of the doctrinal and theoretical bases of professional speech and its application to controversial First Amendment questions.

First Amendment protection for professional speech rests on distinctive theoretical justifications, and the key to understanding professional speech lies in understanding the character of the learned professions. This Article suggests that the professions should be thought of as knowledge communities. Conceptualizing the professions as knowledge communities not only informs the justifications for First Amendment protection but also the limits of that protection, the permissibility of regulation of the professions, and the imposition and extent of tort liability for professional malpractice.

author.Associate-in-Law, Columbia Law School. Many thanks to Vince Blasi, Caroline Corbin, Bill Eskridge, David Fontana, Katherine Franke, Philip Genty, Kent Greenawalt, Sasha Greenawalt, Jamal Greene, Philip Hamburger, John Inazu, Renee Knake, Chip Lupu, Gillian Metzger, Henry Monaghan, Doug NeJaime, Burt Neuborne, David Partlett, Robert Post, David Pozen, Bill Simon, Alex Stein, Allison Tait, Kendall Thomas, Larry Tribe; and participants in workshops at Columbia Law School, University of North Carolina School of Law, Wayne State University Law School, Yale Law School, the 2015 Annual Conference of the Law & Society Association, and the Lavender Law National LGBT Bar Association Annual Conference 2015 for insightful comments and discussions.


Professionals speak; some speak a lot. Lawyers use verbal communication to exercise their profession. So do psychologists. Medical advice is dispensed via such communication as well. The list goes on. The content of these communications, we intuitively assume, is protected. The scope of protection, however, is elusive. At the same time, the state can regulate the professions. Traditional forms of regulation include licensing requirements, advertising regulations, and the imposition of professional malpractice liability. But new forms of regulation go further: they target the content of the communication between a professional and her client. Sometimes, such regulation aligns with professional insights, but sometimes it contradicts them. The resulting tension between state regulation of the professions and professionals’ free speech interests remains underexplored.

Recent cases involving professional speech1 have made this tension apparent. Can the State of California and the State of New Jersey ban sexual orientation change efforts (SOCE)?2 Can the State of South Dakota require that abortion providers read to their patients a legislatively drafted statement that does not correspond to the current state of medical science?3 In other words: do psychologists have a First Amendment right to engage in conversion therapy? Do physicians have a First Amendment right not to be compelled to make state-scripted, erroneous claims about abortion to their patients? These examples represent potential infringements on a professional’s right to free speech. But federal appellate courts have taken opposing approaches to indistinguishable questions.4

What is strikingly—and perhaps somewhat surprisingly—still absent from the case law and the legal literature is a comprehensive theory of professional speech.5 The Supreme Court has never identified, with any clear boundaries, the category of professional speech. Nonetheless, it is implicit in a number of decisions involving government-funded speech,6 commercial speech,7 and other areas.8 This Article seeks to fill the lacuna left by courts and scholars by offering an account of the doctrinal and theoretical bases of professional speech and its application to controversial First Amendment questions.

First Amendment protection for professional speech, I argue, rests on distinctive theoretical justifications, and the key to understanding professional speech lies in understanding the character of the so-called “learned” professions. These learned professions, I submit, should be thought of as knowledge communities, that is, communities whose principal raison d’être is the generation and dissemination of knowledge.9 Conceptualizing the professions as knowledge communities not only informs the theoretical justifications for First Amendment protection10 but also the limits of that protection, the permissibility of regulating the professions, and the imposition and extent of tort liability for professional malpractice.11 Imposing professional malpractice liability has never been found to offend the First Amendment. Why that is so, however, merits further investigation. Conceptualizing the learned professions as knowledge communities guides this undertaking.

Professionals speak not only for themselves but also as members of a learned profession: they “assist[] individuals in making personal choices based on the cumulative knowledge of the profession.”12 The professions as knowledge communities thus function in a way akin to what Paul Horwitz calls “First Amendment institutions.”13 First Amendment scholars concerned with professional speech have hinted at the connection between the professions and institutions14 but have yet to provide a full explication. This Article takes on that task.

My analysis abuts and engages the emerging institutionalist First Amendment literature.15 In my account, it is the institutionalization of professional discourse that builds the basis for the knowledge community. The subsequent dissemination of that knowledge within the professional-client relationship ties the individual professional back to the knowledge community. That the individual professionals are bound together by the knowledge community is also the underlying assumption of professional malpractice law, in which the knowledge community’s standard of care determines the benchmark against which the individual professional’s liability is assessed.

This Article proceeds in four parts. Part I provides a definition of professional speech, with particular attention to the role of the learned professions as knowledge communities. It then situates professional speech in the doctrinal context of the First Amendment. Commentators have analyzed professional speech primarily in relation to—and by analogy with—commercial speech,16 which has received increasingly robust First Amendment protection.17 But the underlying comparison, I argue, is tenuous. The speech interests are fundamentally different. The doctrinal fate of professional speech, therefore, ought not to be tied to that of commercial speech.

Part II undertakes a normative defense of First Amendment protection for professional speech. The traditional justifications for speech protection apply in a distinctive fashion to professional speech. Professional speech is unique in the way it implicates the autonomy interests of both the speaker and the listener. I will call “decisional autonomy interests” the interests of the listener who depends on the information provided by a professional to make an informed decision.18 The professional-client relationship is typically characterized by an asymmetry of knowledge. The client seeks the professional’s advice precisely because of this asymmetry. At the same time, the agency of the listener requires that the ultimate decision rest with her. The other autonomy interests are those of the speakers, which I will call “professional autonomy interests.” The qualifier “professional” signals that it is not the autonomy interest to freely express one’s personal opinions that is at stake—as is the case in most free speech theory—but rather to express one’s professional opinion as a member of the knowledge community.

Turning then to marketplace considerations, I argue that the classic notion of a “free trade in ideas”19 has little purchase as between the professional and the client. The professional does not seek to subject her professional opinion to “the competition of the market”20 when speaking within the confines of the professional-client relationship. Yet, there is a dimension to the marketplace idea in the professional speech context that is generally underappreciated and comes into relief when the professions are thought of as knowledge communities. Within the discourse of the knowledge community itself—that is, outside the professional-client relationship—a marketplace of ideas exists, which we might call an epistemic marketplace. Professional standards are generated by testing insights in that marketplace. The current state of the knowledge community’s discourse provides the foundation for the professional’s advice.

Finally, theories of democratic self-government also provide a normative basis for the protection of professional speech. The information that the knowledge community communicates to clients through individual professionals cumulatively enhances the basis upon which public opinion is formed. This is not simply a matter of enabling self-government through ordinary deliberation by adding another opinion to the public discussion. Rather, professionals contribute specialized, technical knowledge to which lay citizens would not otherwise have access. It is precisely in their capacity as members of knowledge communities that professionals enhance the process of self-governance, and so as members of knowledge communities that they should enjoy First Amendment protection.

Part III considers the appropriate limits on professional speech. It interrogates the extent to which the state may regulate the professions’ educational and knowledge standards. It also considers the interplay between the First Amendment and tort liability for professional malpractice. In order to avoid malpractice liability, professionals must exercise their profession according to the degree and skill of a well-qualified professional. For example, the Restatement (Third) of the Law Governing Lawyers states, “[A] lawyer who owes a duty of care must exercise the competence and diligence normally exercised by lawyers in similar circumstances.”21 It is thus the knowledge community that determines the standard of care. This Part engages contemporary tort scholarship that incorporates this insight by focusing on the profession’s distinctive expertise.22 This emerging approach mirrors my concern with granting deference to the knowledge community’s insights.

The extent of tort liability, I argue, should be consistent with the scope of protection of the knowledge community’s discourse under the First Amendment. Only if liability and protection are coextensive can this liability mechanism yield fair results. If liability is properly measured against the standard of care determined by the profession, the knowledge community’s formation of this standard should remain uncorrupted and its application within the professional-client relationship should receive robust First Amendment protection.

Part IV applies this approach to controversial First Amendment disputes, returning to the cases referenced at the outset. In so doing, it considers how the theory of professional speech focused on knowledge communities plays out in litigation terms, a question that traditionally remains underexamined in the First Amendment literature.

State regulation interacts with knowledge communities’ insights in multiple and varied ways. It sometimes reinforces professional knowledge, and it sometimes contradicts such knowledge. The questions raised in cases challenging regulations that contradict professional knowledge play out against the larger jurisprudential backdrop concerning the role of legislative findings of fact. Whose knowledge should state regulation rely on? The knowledge community theory of professional speech provides a conceptual framework to assess this question. This theory of professional speech, informed by the role of knowledge communities, thus allows us to reconceptualize how we think about government involvement in professional speech. Under this view, to borrow loosely from Alexander Meiklejohn, the First Amendment is directed against the “mutilation of the thinking process” of the knowledge community.23

I. situating professional speech

When a lawyer advises a client, she engages in professional speech. Likewise, when a physician advises a patient, she engages in professional speech. Scholarship and case law seem to assume, almost intuitively, that professional speech exists. But the instinct that professional speech is distinctive as a category of speech, and the way in which it is distinctive, is not sufficiently explained in the case law or First Amendment theory.

This Part first explores the character and function of the professions before distinguishing professional speech from professionals’ private speech and from government speech. Throughout this Article, I argue that the professions should be thought of as knowledge communities. The role of knowledge communities defines the type of speech that ought to be protected from outside—particularly, state—interference, and the extent to which state regulation of the professions is permissible.

This Part then situates professional speech in First Amendment doctrine. The concept is implicit in numerous Supreme Court decisions, though it is not identified as a separate category of speech. Courts and scholars sometimes analogize professional speech to commercial speech, which is increasingly receiving First Amendment protection from state interference. While a heightened level of protection is desirable as a doctrinal matter in the professional speech context as well, the underlying analogy is tenuous. In questioning the analogy between commercial speech and professional speech, I suggest that professional speech more than commercial speech should receive robust First Amendment protection.

A. What Is Professional Speech?

The First Amendment fragments speech. We treat different types of speech differently all the time.24 But to the extent we treat professional speech differently for different professions—thus distinguishing between speakers engaged in arguably the same type of speech—differential treatment is problematic. For example, why should the speech of a lawyer be more protected than that of a physician?25 Nonetheless, the level of attention afforded to the regulation of professional speech varies significantly across professions. First Amendment questions surrounding lawyers’ professional speech, for instance, remained largely unexplored until recently.26 Legislative interference with physician speech, conversely, has received comparatively more attention.27

Unlike other analyses focused on specific professions,28 I aim to develop a broad conceptual approach to professional speech. Doing so avoids creating professional speech silos within the First Amendment and the subsequent problem of sorting professional speech into subcategories. A unified approach to professional speech also shields some professions from being “especially vulnerable to excess constriction by judges and juries too concerned with the moral or social undesirability of those . . . carrying the First Amendment claim.”29 Consider, for example, the situation of reproductive health care providers. There may be less desire to protect professional speech concerned with abortion—and more tolerance for government demands to read inaccurate, legislatively drafted scripts, compelled descriptions of mandatory ultrasounds and the like—based on moral disapproval.30 But if professional speech is worthy of protection as such, then the underlying topic of the speech is irrelevant to its protection. A unified approach to professional speech, then, provides protection for all professional speech.

I submit that the kind of professional speech worthy of protection, irrespective of the particular profession involved, includes three core elements: (1) a knowledge community’s insights, (2) communicated by a professional within the professional-client relationship, (3) for the purpose of providing professional advice. The first element concerns the role of knowledge communities; the second and third elements distinguish the context and purpose. I will address these elements in turn.

1. The Role of Knowledge Communities

The connection to a knowledge community circumscribes the type of communication rendered as professional advice. Not all occupations are considered professions. There are certain core professions we intuitively think of, medicine and law traditionally chief among them.31 Psychologists, dentists, pharmacists, and accountants—to only name a few—are likely part of the group as well. The list has expanded historically, and some occupations that were once considered only marginally professionalized have now come to be understood as professions.32 The process of professionalization is contested, and I do not aim to offer my own theory. Rather, I am concerned with the question of what First Amendment protection for professional speech looks like once an occupation has attained professional status. Thus, for the remainder of this Article, whatever the current debates are at the margins, I am primarily concerned with the core professions. And although the clergy is historically considered the third quintessential learned profession next to medicine and law,33 I explicitly exclude that group from my discussion of professions and professional speech.34

Definitions of “the professions” vary,35 but the most relevant defining feature for present purposes—and one generally shared among the numerous definitions—is their knowledge-based character.36 As we have already observed, the professional-client relationship is asymmetric: the professional has knowledge the client does not have, which leads the client to seek out her advice. The reason the professional’s advice is valuable to the client is that she possesses knowledge that the client lacks.37

Because the professions are knowledge-based, I contend that they should be thought of as knowledge communities. Individual professionals “may differ in their individual judgments about particular issues, [but] their role as professionals traditionally implies their subscription to a body of knowledge that is shared among their peers.”38

What are knowledge communities?39 I use the term to describe a network of individuals who share common knowledge and experience as a result of training and practice.40 They are engaged in solving similar problems by drawing on a shared reservoir of knowledge, which, at the same time, they help define and to which they contribute. Their common understandings allow for the generation and exchange of insights within the community. Consequently, members of knowledge communities have shared notions of validity41 and a common way of knowing and reasoning (consider the old adage of “thinking like a lawyer”).42 Additionally, the knowledge community shares certain norms and values: professional norms. This is not to say that knowledge communities are monolithic. But their shared notions of validity limit the range of acceptable opinions found within them.

The connection to a knowledge community is a distinctive feature of the role of professionals. In a recent case, the Fourth Circuit considered the question of professional speech protection for the “spiritual counselor” (fortune teller) known as Psychic Sophie, who assertedly engaged in providing predictive advice just like a lawyer.43 The Fourth Circuit, relying on Justice White’s concurrence in Lowe v. SEC, stated: “Professional speech analysis applies . . . where a speaker ‘takes the affairs of a client personally in hand and purports to exercise judgment on behalf of the client in the light of the client’s individual needs and circumstances’ . . . .”44 This definition of professional speech allowed the court to conclude that Psychic Sophie’s “activities fit comfortably within the confines of professional speech analysis.”45 Whether or not this assessment of her profession is accurate,46 it importantly lacks the connection to a knowledge community.

First Amendment scholars concerned with professional speech have hinted at the connection between the professions and institutions.47 This emerging institutionalist First Amendment literature is concerned with colleges and universities, libraries, and the press.48 But knowledge communities, while related to these institutions, are in a sense less “institutionalized.”

Their most institutionalized incarnations are professional associations. The Fourth Circuit, for example, invoked the presence or absence of “accrediting institution[s] like a board of law examiners or medical practitioners” in the Psychic Sophie case.49 Likewise, Justice Breyer in dissent once noted that when speech “is subject to independent regulation by canons of the profession[s] . . . the government’s own interest in forbidding that speech is diminished.”50 My account of the role of knowledge communities in the professional speech context makes sense of these intuitions. But professional norms are generated outside of these associations as well. Conferences and the professional literature, for example, are sites of professional knowledge formation, even though they are not necessarily embodied in specific institutions or professional associations.51

Of course, professional associations have held, at one point or another, positions they now consider erroneous or outdated. For instance, the American Medical Association was at the forefront of the campaign to criminalize abortion in the nineteenth century,52 and the American Psychological Association (APA) did not declassify homosexuality as a mental disorder until 1973.53 But the professions themselves can and do revise their positions on the basis of their ongoing intellectual development, as these examples attest.54 In adopting, changing, or updating these positions, the knowledge communities use their own professional standards, elaborated by and through their own community.55

Knowledge communities have specialized expertise and are closest to those affected; they must have the freedom to work things out for themselves. The professions as knowledge communities have a fundamental interest in not having the state (or anyone else, for that matter) corrupt or distort what amounts to the state of the art in their respective fields.56 This is the key feature of professional discourse and the limiting principle of professional speech. The resulting benefit is the generation of insights within the knowledge community that would not otherwise occur. As knowledge communities, then, the professions should be granted deference.57 But where knowledge communities—and, by extension, individual professionals—do not possess such specialized knowledge or competence, such deference is not required as a matter of professional speech. No amount of specialized training, for instance, by itself makes a professional more competent to render value judgments.

The individual professional is linked to the knowledge community in multiple ways. She “is understood to be acting under a commitment to the ethical and intellectual principles governing the profession and is not thought of as free to challenge the mode of discourse or the norms of the profession while remaining within the parameters of the professional discussion.”58 The individual professional thus serves as the conduit between the knowledge community and the client. Malpractice liability likewise assumes this connection in imposing the profession’s standard of care on the individual professional.59

I will return to the role of professional associations and state involvement in regulating the professions in Part III. For now, conceptualizing the professions as knowledge communities allows us to focus our discussion on professional speech as distinct from other forms of speech.

2. Distinguishing Private Speech

Turning to the second and third constitutive elements of professional speech—(2) that it is communicated by a professional within the professional-client relationship, (3) for the purpose of providing professional advice—it is fundamentally important to recognize that professional speech is not private speech. Daniel Halberstam and Robert Post define professional speech as “‘speech . . . uttered in the course of professional practice,’ as distinct from ‘speech . . . uttered by a professional.’”60 This definition crucially distinguishes professional speech from private speech.61

The line between the professional’s private speech and professional speech, then, can be drawn by considering the presence or absence of a professional-client relationship. “Where the personal nexus between professional and client does not exist, and a speaker does not purport to be exercising judgment on behalf of any particular individual with whose circumstances he is directly acquainted,”62 the speaker is not engaged in professional speech. When the professional’s advice is distributed generally or to the public at large, outside of the professional-client relationship, it is most likely not professional speech.63 Investment advice distributed to the general public, for example, does not constitute professional speech;64 nor do books on how to avoid probate,65 diet plans,66 or mushroom guides,67 even though inaccurate information so disseminated may be harmful. When professionals speak in such a manner, they act as ordinary citizens participating in public discourse and accordingly enjoy ordinary First Amendment protection.

The third element of professional speech—that it is for the purpose of providing professional advice—constrains what the professional may say in the context of the professional-client relationship, and so helps distinguish professional speech from other kinds of speech a professional might engage in, whether in public or private. It bears emphasis that First Amendment protection for speech that is not professional advice is unrelated to the speaker’s membership in a knowledge community.68 Although the speaker’s professional training may inform the content of such speech, she is not disseminating the knowledge community’s insights within a professional-client relationship for the purpose of providing professional advice. In fact, in many instances, the speaker may be articulating disagreement with the knowledge community’s consensus, which the professional is not free to do when providing professional advice.69

Post, for instance, recounts the “controversy over the safety of dental amalgams.”70 There, a dentist questioned the professional consensus that dental fillings containing certain substances were safe. Although the dentist no doubt was informed by his professional background, the expression of his opinion was entirely private speech.71 “Within public discourse,” Post explains, “traditional First Amendment doctrine systematically transmutes claims of expert knowledge into assertions of opinion.”72 Any non-dentist’s speech questioning the safety of such fillings would enjoy the same First Amendment protection, though the public would probably ascribe less persuasive force to a non-professional’s assessment of the matter.73

The same reasoning makes political statements like “vote for Obama,” even if uttered within the context of a professional-client relationship, not professional speech but the professional’s private speech.74 It is not communicated for the purpose of providing professional advice, and it is likely not connected to the insights of the knowledge community—even if the knowledge community may have reached a consensus that one candidate for public office will better serve their interests than another.75

3. Distinguishing Government Speech

Another important distinction is between professional speech and government speech. Professional speech must be communicated by a professional, and professionals can operate in different institutional settings with varying degrees of government involvement. The professional may be a government employee, or a government program may fund the professional’s service. Alternatively, the government sometimes seeks to have private individuals disseminate its own message. Under the government speech doctrine, “[t]he government alone may determine its message to the exclusion of all others.”76 Just as the state can be anti-smoking or anti-obesity, it may express a preference on abortion.77 Thus, when the state tries to enlist a private speaker, a key concern is whether the message is attributed to the state or the professional.78

As I have argued elsewhere, effective control over speech should determine responsibility for the message.79 This, in turn, can distinguish government speech from professional speech, and so mark the boundary up to which the state can prescribe speech. When, for example, the state demands that physicians communicate certain claims to their patients in materials of the physicians’ own design, the state effectively tries to obscure authorship even though it is the state that retains effective control over the message communicated.80 Such speech, then, should be understood as an attempt by the government to co-opt or dictate professional speech. According to the theory developed here, such prescriptions would constitute inappropriate regulation of professional speech. But the situation is different where state regulation permits professionals to disavow the state’s message. In Planned Parenthood of Southeastern Pennsylvania v. Casey, the state demanded that a certain message be communicated,81 but the Court’s decision suggested that disclaimers were permissible. The next section more closely examines the doctrinal status of professional speech in light of Casey and other cases that concern government speech.

B. Professional Speech in First Amendment Doctrine

Whether a “professional speech doctrine” currently exists is subject to debate. The Fourth Circuit recently asserted that “[t]he Supreme Court has recognized the regulation of occupational speech under the ‘professional speech’ doctrine at least since Justice Jackson’s concurrence in Thomas v. Collins,” a 1945 case.82 Similarly, some commentators point to Justice White’s concurrence in Lowe v. SEC as declaring the existence of the professional speech doctrine.83 Others are more skeptical.84

Although the Supreme Court has never identified a category of “professional speech” for First Amendment purposes, its existence is implicit in a number of cases.85 The Court most directly addressed the question of First Amendment protection for professional speech in the joint opinion in Casey.86 But the concept is embedded in other decisions as well.87

With Casey—arguably the most on-point treatment—as a starting point, the doctrinal basis of professional speech appears indeterminate at best. But a wide-angle view reveals that, despite the initial lack of clarity in Casey, the Court seems to have at least a hunch that speech communicated by professionals in a professional-client relationship for the purpose of providing professional advice is somehow distinctive.

In Casey, the joint opinion addressed the First Amendment in a somewhat cryptic paragraph:

All that is left of petitioners’ argument is an asserted First Amendment right of a physician not to provide information about the risks of abortion, and childbirth, in a manner mandated by the State. To be sure, the physician’s First Amendment rights not to speak are implicated . . . but only as part of the practice of medicine, subject to reasonable licensing and regulation by the State . . . . We see no constitutional infirmity in the requirement that the physician provide the information mandated by the State here.88

Scholars have been struggling to make sense of this.89 Some appellate courts have arguably taken this obscure statement as license to espouse an exceedingly narrow view of professional speech.90 There is now marked and explicit disagreement among the circuits regarding its proper interpretation.91

But beyond this puzzling paragraph, Casey hints at the doctrinal status of professional speech. The joint opinion directly addressed government speech, compelled speech, and the right to receive information (or not). The government, as the joint opinion and Justice Stevens’s opinion agreed, may communicate its own preference with respect to abortion.92 Regarding compelled speech, the joint opinion found that the government may demand, as part of obtaining the woman’s informed consent, that physicians distribute state-drafted materials and make certain statements to their patients that are “truthful and not misleading.”93 However, the state neither required that the providers communicate this information as their own—which could have made it more difficult for patients to attribute the message to the state—nor prohibited the providers from expressing their disagreement with the state’s policy. Moreover, there was a provision for physicians to refrain from providing certain information if they deemed it harmful to their patients.94 Finally, with respect to the right to receive information or not, women could decline to view the materials.95

As a matter of existing First Amendment doctrine, then, Casey may be read as suggesting that while the government is free to express its own opinion, it may not enlist (potentially unwilling) professionals as mouthpieces to disseminate its message.96

Also in the abortion context, and pre-dating the Casey decision by a year, Rust v. Sullivan further illuminates the doctrinal status of professional speech.97 Although Chief Justice Rehnquist framed the issue as concerning “abortion-related activities,”98 thus apparently avoiding the specific question of professional speech,99 that is in fact what the case concerned.100 The Court noted that “[it] could be argued . . . that traditional relationships such as that between doctor and patient should enjoy protection under the First Amendment from Government regulation, even when subsidized by the Government.”101 But it did not resolve the question, suggesting that the regulations in question “[did] not significantly impinge upon the doctor-patient relationship.”102 The Chief Justice gave the following reasons: first, the doctor was not compelled “to represent as his own any opinion that he does not in fact hold;” second, the professional relationship was not “sufficiently all encompassing” because it “does not provide post-conception medical care” and consequently, “the doctor’s silence with regard to abortion cannot reasonably be thought to mislead a client into thinking that the doctor does not consider abortion an appropriate option for her.”103 Finally, “[t]he doctor [was] always free to make clear that advice regarding abortion is simply beyond the scope of the program.”104

Justice Blackmun’s dissent rejected the “direct regulation of dialogue between a pregnant woman and her physician.”105 In Justice Blackmun’s view, “the regulations impose[d] viewpoint-based restrictions upon protected speech . . . .”106 Importantly for this discussion, Justice Blackmun framed the problem of limiting the scope of advice in terms of both the patient’s expectations as well as professional demands.107 Full, comprehensive advice, in other words, was not only what a pregnant woman expected of her physician—government-funded or not—but also what the medical profession expected of its members.

Rust anticipated the points made in Casey with respect to attribution of speech within government speech doctrine.108 Whether or not the Chief Justice appropriately characterized the extent of the doctor-patient relationship, it is noteworthy that the Rust Court did acknowledge the possibility of First Amendment protection in this professional context. Moreover, it is striking that the Chief Justice suggested drawing an analogy between the doctor-patient relationship and the treatment of universities under the First Amendment.109 This particular institutional analogy likely supports conceptualizing the professions as knowledge communities.

Several decisions concerning legal advice give further doctrinal guidance on professional speech. In Legal Services Corp. v. Velazquez, the Court held unconstitutional a restriction on providing legal advice that “prohibit[ed] legal representation funded by recipients of [Legal Services Corporation (LSC)] moneys if the representation involve[d] an effort to amend or otherwise challenge existing welfare law.”110 As Justice Kennedy explained, “the LSC program was designed to facilitate private speech, not to promote a governmental message. Congress funded LSC grantees to provide attorneys to represent the interests of indigent clients.”111 This makes the legal advice different from government speech—according to Justice Kennedy, “[t]he lawyer is not the government’s speaker”112 and the legal advice is a form of private speech.113 Yet the Velazquez Court did recognize that there is a professional dimension to this speech. The legal system depends on the traditional role of the attorney,114 which includes “complete analysis of the case, full advice to the client, and proper presentation to the court.”115 Limiting the range of permissible speech “prohibits speech and expression upon which courts must depend for the proper exercise of the judicial power.”116 In light of these statements, it is evident that the Court understands professional speech to be distinct from government speech, even when funded by the government. And, although the Court did not make the point explicitly, it appeared to recognize the special import of professional speech—at least that of a lawyer—as distinct from ordinary private speech.

Finally, the two attorney speech cases from 2010, Milavetz v. United States117 and Holder v. Humanitarian Law Project,118 have implications “for those desiring advice about any other area of law where Congress may decide to legislate away the attorney’s ability to advise her client and the client’s right to receive that advice.”119 Milavetz concerned limitations on attorney speech imposed by the Bankruptcy Abuse Prevention and Consumer Protection Act (BAPCPA).120 The Act prohibits “debt relief agencies”—which the Court held attorneys to be—from advising clients “to incur more debt in contemplation of such person filing” for bankruptcy under the applicable provisions.121 In Milavetz, the Court disagreed with the Eighth Circuit’s characterization of the “statute as a broad, content-based restriction on attorney-client communications that is not adequately tailored to constrain only speech the Government has a substantial interest in restricting.”122 As Justice Sotomayor explained, the phrase “in contemplation of bankruptcy” indicates abusive conduct.123 So understood, “advice to incur more debt because of bankruptcy . . . will generally consist of advice to ‘load up’ on debt with the expectation of obtaining its discharge—i.e., conduct that is abusive per se.”124 Importantly, Justice Sotomayor cited Rule 1.2(d) of the ABA Model Rules of Professional Conduct in rejecting the claim that the BAPCPA provisions prohibit frank discussion between lawyer and client.125 Under the crime-fraud provision, lawyers are not prohibited from discussing fraudulent or criminal conduct, but they may not advise their clients to engage in it. In other words, the Court looked to professional standards to provide guidance on the scope of the Act’s prohibition as it concerned “attorney speech.”126 The Court here demonstrated not only an appreciation for the type of professional speech that occurs within the lawyer-client relationship, but also for the role of the professional rules of conduct in defining the scope of this relationship.

The speech at issue in Holder v. Humanitarian Law Project,127 by contrast, occurred outside the boundaries of the lawyer-client relationship; the statute prohibited the formation of that relationship in the first place.128 Various domestic groups and individuals sought to provide information and training to groups designated as “foreign terrorist organizations” on how to assert their own legal claims.129 The Court upheld the “material support” provision of the Antiterrorism and Effective Death Penalty Act against a First Amendment challenge.130 In dictum, Chief Justice Roberts acknowledged that the decision does not suggest “that any future applications of the material-support [for terrorism] statute to speech or advocacy will survive First Amendment scrutiny.”131 Nonetheless, some commentators assert that the decision “is likely to have a chilling effect on attorney advice.”132 The Court in this case allegedly underappreciated the role of attorneys who provide “speech that constitutes legal ‘expert advice or assistance.’”133

In sum, all of these decisions hint at the Court’s incipient conception of professional speech. While professional speech is conceptualized as somehow distinctive, however, the Court lacks the theoretical foundation to properly evaluate First Amendment protection of such speech.

C. The Commercial Speech Analogy

Courts134 and scholars135 have analogized professional speech to commercial speech. But, I argue, the analogy is tenuous; the underlying speech interests are fundamentally different. The content of professional speech, distinctively, is defined by the professional’s connection to the knowledge community.

Most prominently perhaps, Halberstam and Post each propose and defend models that serve as a basis for the analogy. In doing so, however, Halberstam reconceptualizes commercial speech doctrine itself; Post cautions against its wholesale adoption. Halberstam advances the “bounded speech institutions” model, and Post advances a professional speech variation of the democratic self-government model. Both focus on the structure of the communication.

The doctrinal starting point for assessing commercial speech remains the canonical, though increasingly criticized,136 Central Hudson test.137 The Court has ostensibly relied on this doctrinal basis in its expansion of First Amendment protection for commercial speech.138 Writing at the turn of the twenty-first century, Halberstam observed that the classic position of minimal protection of commercial speech was beginning to appear in flux.139 Since then, there has indeed been a considerable expansion of First Amendment protection for commercial speech. The Court now affords what comes close to strict scrutiny review in commercial speech cases.140

But the extent of protection should not be the primary reason to analogize the two types of speech unless doctrine is tethered to theory. This requires “a deeper kinship between the two forms of communication.”141 For Halberstam, this deeper kinship is rooted in the “paradigm of bounded speech institutions.”142 Both professional and commercial speech in this model can be seen as “relational” or “bounded speech institutions,” though Halberstam acknowledges that “the relationship between physician and patient and the duties attendant to that relationship are substantially deeper than those between vendor and purchaser.”143 He makes an (ostensibly descriptive) institutional or structural argument, suggesting that

the Court may be seen as implementing a constitutional theory of bounded speech institutions, based on its perception of various socially defined relationships between interlocutors and, accordingly, rendering contextual judgments about the extent of government intervention that is both necessary for and compatible with the preservation of the particular institution.144

With respect to both professional and commercial speech “[t]he boundaries of the discourse . . . may be policed, but, conversely, as long as the speaker remains within the boundary of the institution, the speaker would be engaged in protected speech.”145 In other words, state regulation serves a definitional purpose—mapping the boundaries of discourse. While speakers remain within those bounds, interference with their speech is impermissible.146 The so-bounded communicative relationships are subject to “contextual First Amendment review that is specifically centered around the social relation, as opposed to an abstract review such as that traditionally applied to the street-corner speaker.”147 Under this model, in both the professional and the commercial speech contexts, “[t]he government may neither suppress the speech entirely nor remodel the institution to its liking.”148

Conceptually, it seems plausible to view both commercial and professional speech in this way. But, while I agree with the differentiation between speech within and outside of a bounded discourse and with awarding First Amendment protection accordingly, I do not embrace the suggested parallel between commercial and professional speech. The “bounded speech institutions” model assumes the equal position of professional and commercial speech in contrast to political or private speech, which is traditionally unbounded.149 However, it does not sufficiently account for the differences between professional and commercial speech. In order to do so, such a structural view is not enough.150 The bounded discourse approach encompasses the individual professional-client relationship, but, in doing so, undervalues the role of the professional’s connection to the knowledge community. In terms of content, the individual professional serves as a conduit for the knowledge community’s insights.

The content of the communication and its relation to the body of knowledge possessed by a knowledge community is distinctive in the professional speech context. So is the imposition of professional malpractice liability and its relation to the professional standard of care. This unique relationship with the knowledge community demands a thicker account of the communication. Thus, the analogy falls short if it is based solely on the structural “bounded speech institutions” model. It explains why the state may impose liability as a structural boundary, but it does not define the content of the boundedness. This makes Halberstam’s model conceptually useful, but ultimately incomplete. To establish a theoretical basis for evaluating professional speech, this model should be supplemented with the theory of knowledge communities.

Post, in setting up the commercial speech-professional speech analogy, focuses on three distinctive features of commercial speech: first, the concern about the flow of information to the public; second, the value attached only to truthful, non-misleading information (and, consequently, the application of content- and viewpoint-based regulations); and third, the permissibility of disclosure requirements based on the emphasis on the public’s right to receive truthful and non-misleading information.151 These three features, in Post’s assessment, closely track the concerns in the professional speech context. In contrast to speech as part of public discourse, the focus of commercial speech, like that of professional speech, is its informational value.152 The knowledge-enhancing character of both types of speech provides the link to the democratic self-government values underlying the First Amendment.153

However, Post offers two distinctions between commercial speech and professional speech, which complicates the analogy.154 The first concerns dissemination of commercial information to the public at large as opposed to the dissemination of professional information only to the client. In an age of sophisticated, highly personalized advertising, however, this characterization of the dissemination of commercial speech may no longer be descriptively accurate.155 The second distinction lies in the presupposed equality of the speaker and the listener in commercial speech and their relative inequality in professional speech. Of course, extensive psychological research on the part of advertisers makes the speaker and the listener unequal in the commercial speech context as well. Product placement, subconscious messaging, and the like give a distinct advantage to commercial speakers over their audiences. The Court may have originally had it right in assuming the vulnerability of consumers, though not because the consumer “lacks sophistication,”156 but because the advertiser has an overabundance of it.157 Thus, Post rightly cautions against pushing the analogy.158

The commercial speech analogy, then, while initially appealing, falls short. It lacks descriptive accuracy and analytical force on numerous counts. A preferable approach, therefore, considers the theoretical justifications for protecting professional speech on its own merits.

II. theorizing professional speech

Conceptualizing the learned professions as knowledge communities allows us to rethink professional speech in light of the traditional theoretical justifications for First Amendment protection. Professional speech as a distinctive form of speech is worthy of First Amendment protection. Situating professional speech within the standard theoretical accounts illustrates the unique ways in which this type of speech intersects with the underlying interests. While some scholars have emphasized the democratic self-government justification for protecting professional speech,159 this Part suggests that other First Amendment theories, based on autonomy interests and the marketplace of ideas, also justify—in a way distinct from other speech contexts—First Amendment protection for professional speech. Without taking a position on which of these traditional theories best justifies First Amendment protection,160 and without ascribing any particular ranking to them,161 I suggest that professional speech interests sound in all standard theories.

With respect to autonomy interests, the role of the professions as knowledge communities reframes the importance of professional autonomy. Although the emphasis is traditionally on the listener when the informational value of the communication is at issue, the speaker’s autonomy interests are implicated as well. Likewise, the knowledge community idea reframes the application of the marketplace theory. The individual professional, under this view, is closely connected to the marketplace of ideas that may be found within the discourse of the profession. Finally, with respect to democratic self-government, the knowledge community concept influences the application of that theory of First Amendment protection for speech. Its effect can be seen in two directions. First, it explains how the individual client can benefit from professional advice directly and how the knowledge basis of the entire community can be enhanced by the individual professional’s communication of the knowledge community’s insights to one client. Second, by providing a close link between the individual professional and the knowledge community, it brings together the individual focus of those who favor a participatory perspective of democratic self-government with those who would focus on the role of the collective.162

A. Autonomy Interests

The autonomy interests implicated by professional speech are somewhat distinct from other speech contexts.163 I will call “decisional autonomy interests” the interests of the listener who needs the information to make an informed decision.164 Decisional autonomy in the professional speech context is very different from the commercial speech context. While commercial speech targets the autonomy of the listener to make commercial choices—thereby contributing to the ability to make independent decisions—the target of professional speech is much more closely connected to the self, at times concerning the physical or psychological integrity of the listener’s own person. Moreover, the speaker pays for the speech in the commercial speech context (though, of course, the goal of commercial speech is often to persuade the consumer to buy a product or service) whereas it is the listener who pays for the speech in the professional speech context, indicating that the economic interests do not align. In professional speech, by contrast with commercial speech, payment for services is secondary to the knowledge-based nature of the service provided.165

The other autonomy interests are those of the speakers, which I will call “professional autonomy interests.” The qualifier “professional” signals that it is not so much the autonomy interest to freely express one’s personal opinions—as is the case in free speech theory concerning public discourse—but rather to communicate insights of the knowledge community as a member of the profession.

1. Decisional Autonomy Interests

The professional relationship is typically characterized by an asymmetry of knowledge. Clients seek professionals’ advice precisely because of this asymmetry. “Clients are presumed to be dependent upon professional judgment and unable themselves independently to evaluate its quality.”166 This is not unique to the learned professions. As Kathleen Sullivan has pointed out, “Lawyers know far more about law than their clients, but information asymmetry creates moral hazards (such as the incentive to lie about the gravity of a problem) for auto mechanics as well.”167 These hazards are exacerbated when the client’s personal health or freedom or significant financial interests are at stake. Thus, “the government may properly try to shield the client from the professional’s incompetence or abuse of trust.”168

The listener’s interests are only served if the professional communicates information that is accurate (under the knowledge community’s current assessment), reliable, and personally tailored to the specific situation of the listener. The client’s agency requires that the ultimate decision rest with her. The nature of the professional-client relationship gives rise to fiduciary duties.169 To bridge the knowledge gap, and to ensure the protection of the client’s decisional autonomy interests, the professional has to communicate all information necessary to make an informed decision to the client.

Thus, the interest in full disclosure is linked to the autonomy interests of those seeking the advice of professionals. To the extent that this is facilitated by an informed consent requirement, as in the medical context, the potential for corruption of the information by outside interference is particularly troublesome. As Justice Stevens pointed out in his opinion in Casey, “Decisional autonomy must limit the State’s power to inject into a woman’s most personal deliberations its own views of what is best.”170 But while this concern is perhaps most obvious in cases involving bodily integrity,171 other forms of professional advice should be equally uncorrupted for the same reason. Concerns regarding the agency of the listener obtain in all professional speech contexts.

2. Professional Autonomy Interests

To the extent autonomy interests matter in professional speech, the focus tends to be on the listener’s interests.172 But the speaker’s autonomy interests are also at stake. Some commentators fall back solely on the professional’s personal autonomy interests.173 Professionals as individuals of course have a First Amendment right to speak their own mind in public discourse, perhaps even challenging the knowledge community’s insights.174 But this is not a primary concern in the professional speech context. Quite to the contrary, there is an expectation within the professional-client relationship that the professional does not challenge the knowledge community’s insights in dispensing professional advice.175

The professional not only speaks for herself, but also as a member of a learned profession—that is, the knowledge community. And that community has an interest of its own. Only if the community remains autonomous can it develop and refine the specialized knowledge that is its essence and the source of its social value. The professional speaker has a unique autonomy interest in communicating her message according to the standards of the profession to which she belongs, precisely in order to uphold the integrity of its knowledge community. Physicians, for instance, should not be compelled to speak in a way that undermines their profession’s scientific insights.

This goes beyond the structural interest in protecting the “bounded speech institutions.”176 It also concerns the content of the communication. While some commentators assert that the professional’s autonomy interests guard against compelled speech “on matters of religion, politics, and values,”177 the professional autonomy interests reach much further. Corrupting the content of a communication to a client within the professional-client relationship fundamentally concerns the professional autonomy interests of the professional. This is an interest that goes to the identity of the professional as a member of a profession,178 rather than the professional’s individual autonomy interest, which is entirely unrelated to her professional role. Conceptualizing the professional as a member of a knowledge community brings the autonomy interest in articulating the uncorrupted insights of the knowledge community into focus.

B. Marketplace Interests

In the realm of professional speech, the classic Holmesian notion of a “free trade in ideas”179 would seem to have little purchase.180 While “the best test of truth is the power of the thought to get itself accepted in the competition of the market,”181 the professional does not seek to subject her professional opinion to this test when speaking within the confines of the professional-client relationship.182 The pragmatic dimension of the market metaphor does not apply: experience and truth in the current assessment of the knowledge community are quite obviously located with the professional, making it inapposite “to capture the idea that truth must be experimentally determined from the properties of experience itself.”183 Indeed, the state may ensure that clients seeking professional advice are not harmed by “false” ideas by way of imposing professional malpractice liability.184 Thus, the classic marketplace paradigm is inapplicable to professional speech within the professional-client relationship.

Nonetheless, there is another facet to the idea of the marketplace theory as applied to professional speech.185 Although scholars have observed that professional speech is distinct from other speech, “which generally treats the truth as just ‘another opinion,’”186 the details remain underexplored. As Paul Horwitz has put it, in the professional speech context, “expertise based on a body of specialized knowledge is the very basis of the value and legitimacy of the speech.”187 It is here that the considerations underlying professional speech intersect with those underlying scientific and academic speech.188

There exists a marketplace of ideas internal to each profession. The issue here is the formation of professional knowledge (rather than, as we saw with the autonomy justification, its dissemination). Within the discourse of the profession, the acceptance of professional insights will depend on the rules established by the profession. Scientific insights, for example, will be subjected to peer review and hypotheses will be subjected to the test of falsification.189 These internal processes serve a purpose akin to that of the Holmesian marketplace of ideas. But, to the extent that such a marketplace of ideas exists as what we might call an epistemic marketplace, and that professional standards are generated by testing insights on that marketplace, nonprofessionals do not participate in it. The current state of the art provides the foundation of the professional’s advice (though current debates within the field may influence what counts as a defensible professional position).190 As knowledge communities, then, the professions should be awarded deference.

As Post notes, the marketplace theory “requires the protection only of speech that communicates ideas and that is embedded in the kinds of social practices that produce truth.”191 It is the professional’s connection to the knowledge community that makes the marketplace theory relevant. If the account offered here is an accurate portrayal of the formation of professional knowledge within the knowledge community, the step from the community to the individual professional follows straightforwardly. In reciprocal fashion, the individual professional’s interest lies in preserving the integrity of the knowledge community’s insights, just as the knowledge community’s interest lies in having the individual professional communicate its insights correctly. While this complements the professional autonomy interests, as just described,192 the focus of this theory is on preserving the integrity of the search for truth—that is, the formation of professional knowledge—within the discourse of the knowledge community.

C. Democratic Self-Government Interests

Focusing on the informational value of professional speech, the democratic self-government theory would find such speech worthy of First Amendment protection because it “cognitively empowers public opinion” and thus “serves the value of democratic competence.”193 (This idea is also reflected in the commercial speech analogy, as discussed earlier.)194 But the democratic self-government value of professional speech might be greater still. Professionals supply information to clients that not only concerns the clients’ own lives but may also “require collective action to change rights and responsibilities in society.”195 For example, courts196 and scholars197 have emphasized the role of lawyers in democratic self-government. Other professionals, too, may contribute to expanding the knowledge base upon which citizens can make informed decisions.

Yet the democratic self-government theory builds on some debatable assumptions. It may seem questionable whether a client or patient would, in fact, be primarily concerned with the policy implications of the professional advice she receives. Is the lawyer’s client really thinking about broad questions of access to justice? Is the physician’s patient really thinking about health policy? Or are both primarily concerned with having their individual problems solved? While these questions are sometimes acknowledged in the literature, the abstract possibility of taking political action based on the individualized professional advice received appears sufficient to justify applying the theory to professional speech.198

Within the theory of democratic self-government, two distinct strands arguably stand in opposition to each other: one emphasizes the “safeguarding of collective processes”; the other emphasizes individual rights.199 Conceptualizing individual professionals as part of the larger knowledge community—as conduits communicating the knowledge community’s insights, and thus as trustees for the speech of others—reconciles the two democratic self-government approaches in the professional speech context. The close connection between the individual professional and the knowledge community links the individual role of the professional and the collective role of the knowledge community to which the professional belongs.


As this Part has demonstrated, the traditional theoretical justifications for First Amendment protection apply to professional speech in a unique way. All standard theories suggest that professional speech deserves robust First Amendment protection.

III. limiting professional speech

This Part considers the appropriate limits on professional speech. The state may regulate the professions, but “[b]eing a member of a regulated profession does not . . . result in a surrender of First Amendment rights.”200 And as Eugene Volokh has noted, “it’s far from clear that the government should be completely free to regulate professionals’ speech to their clients.”201 Therefore, it is worth unpacking what state regulation of the professions means and determining when such regulation directly and impermissibly affects professional speech.

Section III.A briefly considers the history of regulating the learned professions. Initially self-regulating, the professions developed a set of norms that solidified over time. State involvement in professional regulation followed. Turning to three typical kinds of regulations—namely concerning advertising, access to the profession, and unauthorized practice—I will demonstrate that professional speech concerns do not ordinarily arise in these contexts. These types of regulations do not generally concern the body of professional knowledge that forms the repository for individual professionals’ advice to clients and its subsequent communication. Thus, while these types of regulations may have far-reaching consequences, they do not implicate professional speech interests as defined here.202 This makes the importance of distinguishing between regulation of the profession and regulation of professional speech palpable.

Section III.B then turns to the interplay between the First Amendment and tort liability for professional malpractice. The tort regime in this context functions as a form of regulation.203 The imposition of malpractice liability has never been found to offend the First Amendment. But the conventional answer as to why that is so is unsatisfactory. Stated in an oversimplified way, the argument is that the state may regulate the professions, and the permissibility of regulation is incompatible with the First Amendment.204

There is an expansive body of literature on professional malpractice law—its effects on professionals and clients, larger policy implications, and possible need for reform. All of this is well beyond the scope of this discussion. My point here is relatively narrow and conceptual. Professionals may be held liable for “unprofessional” speech—that is, speech within the professional-client relationship, for the purpose of providing professional advice, that fails accurately to communicate the knowledge community’s insights.

The liability scheme thus draws on the same body of professional knowledge that I have argued deserves First Amendment protection. If liability is appropriately allocated against this benchmark, the liability scheme normatively supports—rather than undermines—protection of professional speech. In order to achieve fair results under this scheme, professionals may be held liable only under a standard that is exclusively determined by the profession. It follows that the knowledge community’s insights and their communication to the client by the individual professional must remain uncorrupted.

A. Regulation of the Professions

State regulation of the professions is not incompatible with protecting professional speech. Maintaining a focus on the role of knowledge communities, this section outlines the extent of permissible regulation of the professions in light of its history. The historical perspective illuminates the nexus between licensing, state power, and regulation of professions and professionals. There is a long history of self-regulation of knowledge communities.205 Traditionally, certain professions themselves created barriers to entry into the profession, policed membership, and established a distinct professional “culture.” This culture then solidified into a set of professional norms, enforced by professional bodies overseeing the standards of entry and membership. The state assumed some of these functions over time, either taking on the role of regulator directly or through its interaction with professional associations.206 Licensing requirements for law and medicine in the United States likely date back to the founding period,207 although there was a noticeable retreat from licensing in the Jacksonian era.208 The relationship between the regulated professions and the regulating state generally remained one of collaboration; in the case of licensing, for instance, state involvement was overwhelmingly welcomed—even “eagerly sought”209—by the professions.210

There are now numerous ways in which the state regulates the professions. For example, “[t]he medical and legal professions . . . have long been subject to licensing and supervision by the State ‘for the protection of society,’ and the Court has indicated that such regulations would be upheld if they ‘have a rational connection with the applicant’s fitness or capacity to practice’ the profession.”211 I will consider briefly three prototypical areas of state regulation of the professions: advertising, access to the profession, and unauthorized practice. None of them, as the following discussion demonstrates, directly address the types of professional speech issues with which I am concerned. Therefore, they do not constitute “limits on professional speech” in the strict sense of the term. The takeaway is simple, but important: protecting professional speech does not make state regulation of the professions impossible.

One prominent context in which professional regulation as a matter of free speech has been litigated in the past has been advertising.212 In a series of cases, the Supreme Court has dealt with questions of advertising and solicitation regulations for professional services, such as legal services,213 accounting services,214 and dental215 or medical services.216 The gist of these decisions is that professional advertising is largely—though not uniformly217—protected as a matter of commercial speech.218 Advertising for professional services is commercial speech, and “[c]onstitutional protection for attorney advertising, and for commercial speech generally, is of recent vintage.”219

Historically, professional ethics prohibited advertising, and courts consistently deferred to professional ethics in upholding advertising restrictions. As Walter Gellhorn noted in the mid-1970s, “[t]he unethicality of advertising has long been an article of faith among professionals, and the courts have generally shared this faith.”220 This deference to professional norms was long-standing. Chief Justice Hughes, in a 1935 case involving dentists’ advertising, stated: “What is generally called the ‘ethics’ of the profession is but the consensus of expert opinion as to the necessity of such standards.”221

But, as Kathleen Sullivan observed, “[t]he decisions upholding professional ethics regulations against First Amendment challenges are difficult to square with the Court’s other advertising decisions.”222 And Chief Justice Hughes’s statement—that professional ethics are part of the profession’s expert opinion—goes too far. No specialized knowledge is needed for the question of whether advertising for professional services is appropriate; it is a purely economic question. As a matter of institutional competence, courts can rely on their own expertise in economic matters.223

This helps us understand why courts have turned away from their earlier deference to professional norms prohibiting advertising and why, in embracing commercial speech protection for advertising against professionals’ wishes, they have nevertheless begun to regulate professional speech. On matters of regulation that do not directly concern the specialized knowledge of knowledge communities that constitutes the basis for professional advice, professional speech protection should not require broad deference to the profession. The professional advertising her services is not speaking as part of the knowledge community to transmit advice to a client. She speaks only as a private commercial actor. Professional advertising, like commercial advertising, thus is properly reviewed as a matter of commercial speech.224

Beyond advertising, the state may determine educational and other fitness standards for the profession. Imposing limits on access to a profession by establishing educational standards or licensing and certification requirements does not affect professional speech directly. To be sure, there is a long-recognized tension between restricting access to ensure competent advice and restricting access in order to limit competition.225 And there certainly is potential for abuse.226 “On the one side is the need to preserve the integrity of professional knowledge; on the other side is the fact that professional knowledge sometimes reflects sociological prerogatives of class and power that should be disciplined by democratic political purposes.”227 Indeed, some have pointed out that “there is a large body of historical, economic, and sociological literature that suggests that the primary motivation for professional licensing laws is economic self-interest.”228 Without taking a position on the extent of self-interest in professional licensing, it seems relatively unproblematic from a First Amendment perspective to permit some form of access control.229

Sometimes, First Amendment problems can arise if access to the profession is denied because of the content of an applicant’s speech. One prominent example is the case of George Anastaplo, whose bar application was denied by the Illinois Bar due to his refusal to answer questions regarding his views on the Communist party.230 (He famously argued his own case before the Supreme Court, lost in a 5-4 decision, and became a law professor instead.)231 But the types of First Amendment problems arising here are different from those in the professional speech context. Here, it is not the knowledge community’s specialized knowledge that the state interferes with but rather the individual professional’s opinion.232 Thus, an appropriate shield against such restrictions may be found in the professional’s individual First Amendment rights.

Finally, unauthorized practice regulations raise issues similar to regulations concerning access to the profession. First Amendment challenges to unauthorized practice rules—complicated by definitional opacity233—have mainly centered on the question of whether individuals may disseminate certain “information” (as distinct from professional “advice”). Here, unlike in the professional speech context, however, regulation polices the formation of a professional-client relationship rather than the communication of professional advice within such a relationship.

The state regulations just discussed establish the boundaries of professional-client discourse without directly affecting its content. Structurally, they define the speakers’ “social roles” within the “specific communicative relationship.”234 In this respect, Halberstam correctly observes that “government regulation is not invariably destructive of communicative interests, but may indeed foster the communicative relationship and assist in institutionalizing the bounded discourse.”235 In other words, “content-based government regulation may enhance, rather than compromise, the speech practice.”236 Yet, as already discussed, this structural understanding does not go far enough in determining the substance of the bounded discourse—the knowledge community’s insights provide this dimension.

In sum, then, state regulation may limit access to the professions or what professionals may do in certain circumstances. The wishes of the professions in these respects may be laudable or not. But as long as state regulation remains disconnected from the knowledge that forms the basis of the professionals’ advice, it does not pose the type of First Amendment professional speech problems I am concerned with here. State regulation of the professions is far from unproblematic, but the problems that arise are not of the same kind as those directly concerning professional speech—that is, the communication of the knowledge community’s insights, within the professional-client relationship, for the purpose of providing professional advice. The mere fact that the state may regulate the professions therefore has little bearing on the question of First Amendment protection for professional speech.

B. Tort Liability

The tort regime directly addresses harms caused by “unprofessional” speech, that is, bad professional advice. Conventionally, the relationship between the First Amendment and professional malpractice liability—in this case, medical malpractice—is framed as follows:

Medical activity that consists primarily of speech does not automatically deserve First Amendment protection. There are instances when speech essentially amounts to the practice of medicine and could be considered a regulated activity. For example, physician advice regarding the necessity or wisdom of a particular surgical procedure could give rise to malpractice liability, which many would agree has few First Amendment implications even though the advice is itself speech.237

But this common framing is not entirely accurate in light of the role the knowledge community plays.

Juxtaposing professional speech protection and professional malpractice liability leads to conceptual inaccuracy. It is an exaggeration to assert that professional speech is not—and ought not be—protected because the professional is subject to tort liability for “unprofessional” speech. The contrast between permissible regulation and protection is not as stark as it is commonly portrayed—and the two are certainly not irreconcilable.238 In fact, as already indicated, they are complementary. Protection and liability are best conceptualized as two sides of the same coin, and the substantive content of both is determined by the insights of the knowledge community.

1. Professional Malpractice

It is correctly understood that “[m]alpractice law protects the vulnerability of clients by requiring professionals to maintain strict standards of expert knowledge.”239 But the imposition of liability for professional malpractice is not actually the same as regulation of the profession, or even a limit on professional speech in the strict sense of that term. Malpractice liability ensures that the professional’s speech accurately communicates the knowledge community’s insights within the professional-client relationship. On the flip side, “unprofessional” speech is unprotected.240

Post explains the connection between malpractice liability and professional knowledge as follows:

[M]alpractice law outside of public discourse rigorously polices the authority of disciplinary knowledge. It underwrites the competence of experts. Doctors, dentists, lawyers, or architects who offer what authoritative professional standards would regard as incompetent advice to their clients face strict legal regulation. In such contexts, law stands as a surety for the disciplinary truth of expert pronouncements. By guaranteeing that clients can plan to rely on expert professional judgment, law endows such communication with the status of knowledge.241

Post’s presentation is compelling. But it has some unstated premises. In particular, for his gloss to be correct, the knowledge community must decide for itself what “disciplinary truth” is, and any outside interference with their determination ought to be met with great skepticism.

This is already implicit in the way malpractice liability works. The standard of care against which a given professional is judged to determine malpractice liability is whether she has exercised the profession according to the degree and skill of a well-qualified professional. A lawyer “must exercise the competence and diligence normally exercised by lawyers in similar circumstances.”242 Likewise, “a doctor commits malpractice when he treats a patient in a way that deviates from the norms established by the medical profession.”243 It is thus the knowledge community that determines the standard of care. Moreover, only the knowledge community’s specific insights matter. Deference is thus awarded to the core knowledge, not to peripheral interests.244 This mirrors conceptually the First Amendment interests of the knowledge community and its members.

There may be variations as to who constitutes the appropriate reference group (i.e. whether a national standard or a local standard is applied as the baseline).245 But the technical approach is generally the same: a professional standard is juxtaposed against the individual professional’s activities.246 The imposition of liability does not encompass which specific advice may be given. It only asks whether the advice rendered is appropriate as a matter of professional care.247 As one commentator points out in the medical malpractice context, “the medical profession single-handedly determines the entries into treatment-related liability for malpractice.”248

The extent of liability under the common law should be congruent with the scope of protection of the knowledge community’s discourse under the First Amendment. Only if liability and protection are coextensive can this liability mechanism yield fair results. If liability is properly measured against the standard of care determined by the profession, the knowledge community’s formation of this standard should remain uncorrupted and its application within the professional-client relationship should receive robust First Amendment protection. Post hinted at this mechanism in asserting that “we should expect to see First Amendment coverage triggered whenever government seeks . . . to disrupt the communication of accurate expert knowledge.”249

2. Informed Consent

Independent of the professional malpractice claim, a separate cause of action exists in the medical context based on the physician’s duty to inform the patient of relevant information relating to the treatment.250 There is a troubling history of paternalism in the medical profession that limited the amount of information shared with patients.251 But the last century has seen the recognition of patients’ autonomy interests and, as a result, significant changes in the doctor-patient relationship.252 “Autonomy soon became the driving principle used to resolve issues within medicine,”253 and, with it, “informed consent doctrine . . . driven in large part by a desire to combat the paternalism of medicine.”254

The doctrinal origins of informed consent are often traced to a 1914 New York Court of Appeals decision authored by then-Judge Cardozo in which he stated: “Every human being of adult years and sound mind has a right to determine what shall be done with his own body . . . .”255 The real turn toward information, however, occurred in decisions from the 1970s. In Canterbury v. Spence, emblematic of the trend, the D.C. Circuit emphasized the need for information in self-determination.256 Accordingly, this shift was accompanied by a shift in the treatment of informed consent from sounding in battery to negligence.257

There is continued debate over whether the current tort paradigm appropriately accounts for patients’ interests, or whether it continues to be too physician-centric.258 Courts have adopted a negligence approach to informed consent with “the principle of self-determination as the bedrock of modern informed consent doctrine.”259 But the variations that persist tend to value either the physician’s role or the patient’s autonomy more heavily.260 The two standards are the reasonable patient standard and the reasonable physician standard.261

With respect to the First Amendment, then, “[a]ny physician who has been held liable for failure to obtain the informed consent of his patient could argue that the law impairs his autonomy because it requires him to speak in ways that he would prefer not to.”262 But here, too, the knowledge community’s standards limit the extent to which a physician could reasonably assert such a thing. This is because “the scope of disclosure is bound only by what is material to medical, as opposed to non-medical, interests. Cabining the information that physicians must disclose to that which is material to patients’ medical decisions avoids holding physicians accountable for matters that go beyond their expertise.”263 It is again the knowledge community’s professional knowledge that circumscribes the relevant information.264 And it is therefore necessary to keep the knowledge community’s information-formation process free from outside interference. Thus, imposing an informed consent requirement does not technically restrict the professional’s First Amendment rights if appropriate disclosure is considered a part of medically necessary information flow within the doctor-patient relationship. It is “unprofessional” speech—or “unprofessional” silence—that is punished.

IV. when professions speak

When state regulation directly targets “unprofessional” speech as a matter of tort liability, as discussed in the previous Part, it ensures that information consistent with the knowledge community’s insights is conveyed. As long as state regulation reinforces the knowledge community’s insights—which it does when the knowledge community’s standard is applied as the liability benchmark—no significant problems arise. State regulation delineates the professional-client relationship. And state regulation appropriately tracks concerns related to safeguarding the flow of accurate information from the knowledge community through the conduit of the individual professional.265 As is well understood in the literature, “[g]overnment regulation and licensing of the profession as well as the legal enforcement of professional norms thus may assist in establishing the trust that patients can place in their physicians.”266 Indeed, “content-based government regulation may enhance, rather than compromise, the speech practice.”267 But this is only true as long as the regulation mirrors, and does not contradict, professional norms.

When the state overreaches, significant problems arise. This is the fundamental problem with new types of state regulation we are seeing now. This Part demonstrates how the knowledge community-focused theory of professional speech works when applied to controversial First Amendment questions, returning to the cases referenced at the outset.268 Some of these regulations directly target and attempt to alter the core of the knowledge community’s insights and their communication from professional to client. The following three sections illustrate a spectrum of regulations that defer to the professional standard, (partially) codify the professional standard, or compel professionals to speak in a manner that contradicts the professional standards of the knowledge community (or prohibits the professional from communicating the knowledge community’s insights). These forms of regulatory interaction between legislatures and knowledge communities suggest that state regulation of the professions can sometimes be supportive of professional speech rights and sometimes be in tension with them.

The types of facts relevant in professional speech cases—as in a variety of other constitutional cases that turn on questions of fact—“are not of the ‘whodunit’ variety of what happened between the parties. They are instead more generalized facts about the world: Is a partial-birth abortion ever medically necessary?”269 Or, in the professional speech context, is legal advice to load up on debt in anticipation of bankruptcy always fraudulent? Is SOCE therapy harmful? Does terminating a pregnancy result in an increased risk of suicide? The crux lies in determining whose knowledge we should rely on to provide answers.

The following discussion is embedded in a larger jurisprudential context. A long-standing typology distinguishes between legislative and adjudicative facts.270 Legislative facts are not only the facts found by legislatures in enacting legislation but also the facts that adjudicative bodies find to apply beyond the confines of a particular case.271 The distinction has important implications for the questions of fact review that come into sharp relief when findings of fact deviate from the knowledge community’s insights. The following discussion considers how First Amendment theory plays out in litigation, a problem that has not traditionally received much attention from First Amendment theorists. In doing so, it takes into account important aspects of procedure surrounding the litigation of First Amendment claims.272

A. Deference to the Professional Standard

In Milavetz, the Court upheld the Bankruptcy Abuse Prevention and Consumer Protection Act (BAPCPA) relying in part on the grounds that it aligned with the profession’s own definition of permissible communication within the lawyer-client relationship.273 Interpreting the restriction on attorney speech from the perspective of the knowledge community ensured that professional speech concerns did not arise.

From the First Amendment perspective, this approach constitutionalizes the professional standard. This happens in other doctrinal areas as well. In Sixth Amendment doctrine, for instance, the right to effective counsel to a certain degree constitutionalizes professional standards.274 Thus, in Padilla v. Kentucky, the Court noted that “[t]he weight of prevailing professional norms supports the view that counsel must advise her client regarding the risk of deportation.”275 Beyond applying professional standards in effective counsel cases,276 this conceptual approach aligns speech regulations in a manner consistent with First Amendment protection of professional speech. This “constitutionaliz[ation] of individuals’ professional roles”277 goes beyond delineating the professional-client relationship. It gives the relationship substantive content by deferring to the knowledge community’s insights.

It also has procedural implications.278 Here, it is important to note as a threshold matter that the Supreme Court “never set forth a general test to determine when a procedural safeguard is required by the First Amendment.”279 Yet “[t]he institutional characteristics of the American judicial system are . . . of central importance in realizing the constitutional guarantees.”280 Reconceptualizing the role of the professions as knowledge communities, and advancing a theory of professional speech as I propose, has significant implications for the allocation of authority in the judicial process.

The integrity of professional advice is protected by the First Amendment, as well as by ordinary tort law, which subjects “unprofessional” advice to malpractice liability. But whereas in an ordinary tort law case the jury verdict is conclusive, First Amendment protection of the professional standard gives the professional potentially valuable legal protection. At a procedural level, constitutionalizing the professional standard hands important questions to the judge. On review, these questions are subject to independent assessment of the facts by the court. The resulting procedural allocation of fact review takes account of the interest in maintaining the integrity of professional speech.281 Ultimately, the knowledge community-focused theory of professional speech results in a significant shift of decision-making and review authority to the judge. This gives procedural protections to the professional who speaks in accordance with the knowledge community’s insights, but does not protect the professional who fails to do so.

The justification for contracting the jury’s role flows directly from the First Amendment interests underlying professional speech, discussed in Part II. The fundamental interest lies in accurately communicating the knowledge community’s insights to a client seeking professional advice. Whether speech is protected as professional speech rests on whether it accurately conveys the knowledge community’s insights.

B. Codification of the Professional Standard

California’s SOCE ban282 and similar legislation modeled after it283 arguably “tread[] on ill-defined areas of First Amendment law.”284 Following the Supreme Court’s denial of certiorari in the California cases upholding the ban against First Amendment challenges, the ban will go into effect,285 and legislatures elsewhere may be emboldened to enact similar legislation.286

The Ninth Circuit held the SOCE ban to regulate conduct rather than speech.287 Following the Ninth Circuit, a federal district court in New Jersey likewise concluded that that state’s SOCE ban does not regulate speech but conduct.288 However, “the ‘conduct-speech’ distinction is likely to be more misleading than helpful here. When the government restricts professionals from speaking to their clients, it’s restricting speech, not conduct.”289 Creating a circuit split on the issue, the Third Circuit disagreed with the Ninth Circuit in holding that conversion therapy is speech.290 I contend that the Ninth Circuit and the Third Circuit rightly upheld the respective SOCE bans, though for the wrong reasons.291 Under my account, the activity regulated by the SOCE legislation—“talk therapy”—is speech.292 But as professional speech, it is a specific kind of speech. It is the speech that communicates a knowledge community’s insights within a professional-client relationship for the purpose of providing professional advice.

The California and New Jersey legislatures enacted their findings by referring to various professional organizations’ statements on SOCE.293 Nonetheless, the codification approach is not entirely unproblematic. For one, there is the problem of legislative findings.294 The bill passed by the California legislature entangles the factual and normative elements typical for legislative findings: “Being lesbian, gay, or bisexual is not a disease, disorder, illness, deficiency, or shortcoming. The major professional associations of mental health practitioners and researchers in the United States have recognized this fact for nearly 40 years.”295 Instead of deferring entirely to the knowledge community, the legislature adopts a factual assertion as the premise underlying the legislation. In this instance, the premise is shared by the knowledge community, but it is conceivable that a legislature may enact as a legislative finding a position that has not yet reached majority status or consensus within the knowledge community. In the most egregious instances, as discussed in the next section, the legislative findings may be diametrically opposed to the knowledge community’s insights.

Some suggest that there is no consensus within the “psychological establishment” regarding the harms of talk-therapy SOCE.296 Thus, “[a]ccounting only for clinical evidence of SOCE’s harmfulness could, at least at this point, rationalize only a ban on physical interventions like aversion therapy . . . .”297 But it is difficult for both legislatures and courts to evaluate the scientific literature and determine whether a consensus exists.Here, the more workable approach is to defer to the knowledge community. Indeed, the APA follows a broad definition of harm caused by SOCE therapy.298 The legislature may rightly defer to that professional standard. As a corollary, we would also expect tort liability for licensed professionals who engage in conversion therapy.299 Yet the codification approach may prove inefficient. In order to accurately reflect the knowledge community’s insights, the statute has to be flexible over time, since the knowledge community’s insights might change.300

Consider here also the ban’s limited scope. In addition to the legislature having to choose among scientific opinions that may not be entirely clear within the profession, the legislation’s limited scope might raise concerns. If the knowledge community deems conversion therapy harmful for everyone, limiting the ban to minors may not properly reflect the knowledge community’s insights.301 On the one hand, the underinclusiveness resulting from the law’s limited reach might be seen as First Amendment protective: less speech is restricted. On the other hand, under the knowledge communities-centered theory of professional speech I offer, it raises the problem of selective enactment. Under my account of coextensive liability and protection, consider an adult patient who receives conversion therapy, which is not prohibited by the legislation. The adult later suffers adverse effects and sues the mental health provider for malpractice. Given the statute’s limited reach, the mental health provider might invoke the First Amendment as a defense. But if the First Amendment is properly understood as protecting the knowledge community’s insights and their subsequent communication and if malpractice liability properly mirrors that understanding by sanctioning “unprofessional” speech, the limited scope of the statute should be of no help to the mental health provider.

How would the theory of professional speech offered here play out in practice? Consider first the example in which a licensed mental health provider (a) wants to engage in conversion therapy—attempting to use the First Amendment as a sword (as in Pickup)—or (b) engages in conversion therapy and, under the ban, faces revocation of her license and attempts to use the First Amendment as a shield. Consider then a second example in which a licensed psychologist engages in conversion therapy and is sued for malpractice by a patient.302

In the two scenarios set out in the first example, the procedural story would play out as follows: In (a), the licensed mental health provider would argue that SOCE is protected under the First Amendment. The question of First Amendment coverage is one for the judge. If professional speech coverage is determined by deference to the knowledge community, the judge will not find that SOCE is protected under the First Amendment as a matter of professional speech. In scenario (b), the licensed mental health provider would argue that revocation of the license is impermissible because the SOCE ban infringes on her First Amendment rights, and the subsequent events would unfold as in scenario (a). The shift to the judge is mirrored in the malpractice example. Only “unprofessional” speech is subject to malpractice liability. Professional speech—that is, communication of the knowledge community’s insights within the professional-client relationship for the purpose of providing professional advice—however, is not.

C. Compelled Speech Contradicting the Professional Standard

The most problematic—and, under this theory of professional speech, most likely impermissible—type of regulation is one in which the state either demands that the professional communicate information that is incompatible with the knowledge community’s insights or prohibits the professional from communicating the knowledge community’s insights.303 In addition to offending the individual professional’s interest in communicating accurate and relevant professional information, these types of regulation also offend the knowledge community’s interests in having its insights disseminated accurately by members of the profession.304 An example of compelling the professional to convey inaccurate information is the informed consent requirement at issue in the Planned Parenthood Minnesota, North Dakota, South Dakota v. Rounds litigation, in which physicians have to inform patients of an “[i]ncreased risk of . . . suicide.”305 An example of the state prohibiting the professional from communicating accurate information to the client is on display in the medical marijuana cases. Similar problems arise when the state determines what constitutes relevant information, such as in the mandatory ultrasound cases, or attempts to proscribe some information as irrelevant, a constellation that recently arose in Florida, where doctors are prohibited from inquiring about gun use or ownership. I address these examples in turn.

The suicide advisory at issue in the Rounds litigation represents a recent instance of direct state interference with the knowledge community’s insights. A South Dakota statute requires “the disclosure to patients seeking abortions of an increased risk of suicide ideation and suicide.”306 The district court and a panel of the Eighth Circuit held that the suicide advisory infringed doctors’ First Amendment rights.307 The South Dakota statute required doctors to disclose “all known medical risks of abortion.”308 The Eighth Circuit panel emphasized the importance of the word “known.” It crucially noted: “Legislatures have ‘wide discretion to pass legislation in areas where there is medical and scientific uncertainty,’ but the suicide advisory asserts certainty on the issue of medical and scientific knowledge where none exists.”309 What is “known” as a matter of professional knowledge is for the knowledge community to decide, not the state legislature.

On partial rehearing en banc, limited to the issue of the suicide advisory, however, the Eighth Circuit reversed, holding that the required disclosure of increased risk of suicide ideation and suicide was truthful, non-misleading, and relevant.310 The en banc plurality stressed the state’s ability to regulate in the face of “medical and scientific uncertainty,”311 relying on Gonzales v. Carhart,312 and demand that physicians provide the suicide advisory.313 But two separate concurrences interpreted the plurality’s opinion to “require only a disclosure as to relative risk that the physician can adapt to fit his or her professional opinion of the conflicting medical research on this contentious subject”314 and that “the physician [is] free to augment that description [of the relative risks as reflected in the peer-reviewed literature] based on his or her professional judgment.”315 The concurrences thus give somewhat more weight to professional knowledge and deference to the individual professional.

One critic of Rounds II suggests that “the Eighth Circuit should have performed a more robust First Amendment inquiry, calibrated toward ensuring clinically and professionally appropriate speech within the doctor-patient relationship.”316 Doing so would have required the court to anchor its inquiry in a theory of professional speech. My theory would allow it to do so. Under the theory I propose, the knowledge community’s insights are the first element of professional speech. In deciding whether the speech is protected by the First Amendment, the judge would have to determine whether the knowledge community’s insights are being communicated.

The suicide advisory controversy also illustrates the problem of using terminology in legislative fact-finding that may be inconsistent with the knowledge community’s usage. The South Dakota statute “used ‘risk factor’ in a manner inconsistent with its medical meaning, leaving doctors ‘to guess as to the meaning the legislature intended to give to the phrase.’”317 The district court noted that “the legislative drafters ‘may not have fully understood the meaning of this phrase as used in the medical profession.’”318 Deference to the profession avoids confusion as to the meaning of terms of art within the discourse of the knowledge community.

The contemporaneous reproductive rights controversy over mandatory ultrasounds, while compelling doctors to speak in a state-mandated manner, is slightly different in that it does not require the disclosure of false information.319 Rather, it demands the communication of irrelevant information toward an arguably nonscientific ideological end (dissuading women from obtaining an otherwise legal professional service).320 As compelled ideological speech, it suggests proper First Amendment analysis should be based on the principles set forth in Wooley v. Maynard321 and West Virginia State Board of Education v. Barnette.322 But the Fifth Circuit upheld a Texas mandatory ultrasound and sonogram statute as “the epitome of truthful, non-misleading information” that can be required by the state in the course of regulating medical practice.323 The Fourth Circuit, by contrast, struck down a similar piece of North Carolina legislation.324 Judge Wilkinson did note that “[t]his compelled speech . . . is a regulation of the medical profession.”325 Nonetheless, it “extend[s] well beyond” the measures the state has ordinarily employed to ensure informed consent.326 In the end, the Fourth Circuit rejected the regulation as compelled speech violating the First Amendment. In so doing, the court “borrow[ed] a heightened intermediate scrutiny standard used in certain commercial speech cases.”327 Yet, as discussed in Section I.C above, that analogy is unsatisfactory. Thus, while the Fourth Circuit reaches the right outcome in the case, it does so on feeble theoretical footing. The Texas and North Carolina mandatory ultrasound regulations represent precisely the new type of aggressive state regulation directly targeting professional-client communications. Under the knowledge community-focused theory of professional speech, the professional is to decide what is relevant professional information. The knowledge community’s insights not only determine what accurate information is, but also what is relevant in any given situation according to the specific circumstances of the client.

The flip side of compelling professionals to make statements that do not correspond to the knowledge community’s insights is prohibiting them from giving accurate advice. One prominent example involves the threat to “prosecute physicians, revoke their prescription licenses, and deny them participation in Medicare and Medicaid for recommending medical marijuana.”328 Prohibiting this type of professional communication raised the question of the extent to which regulation of professional speech is permissible under the First Amendment.329 The district court held that “the First Amendment protects physician-patient communication up until the point that it becomes criminal . . . .”330 Therefore, “[t]he First Amendment allows physicians to discuss and advocate medical marijuana, even though use of marijuana itself is illegal.”331 The Ninth Circuit affirmed.332

Under the theory of professional speech advanced here, communication about the medical benefits of marijuana use would be protected as a matter of professional speech. Even if insights regarding the benefits of marijuana were not uniformly shared within the knowledge community,333 communicating them within the physician-patient relationship does not offend the knowledge community’s insights in the way communicating erroneous statements does. This highlights the difference between unclear (or emerging and as yet untested) insights and false (tested and rejected) assertions. It is for the knowledge community to decide the content of its insights rather than for the state to determine them. The legislatively enshrined Rounds suicide advisory thus patently offends the professional knowledge formation and dissemination process. So does the classification of marijuana as a drug listed in Schedule I of the Controlled Substances Act, according to which it has “no currently accepted medical use.”334

Just as the state may not decide for professionals what constitutes relevant information and compel them to communicate it (as in the mandatory ultrasound example), the state may not decide in their stead what constitutes irrelevant information and prohibit professionals from communicating it. The State of Florida, for instance, prohibits doctors from asking questions about guns as a matter of course.335 The Eleventh Circuit held this restriction on a professional’s speech to be constitutional as “a legitimate regulation of professional conduct.”336 Just as the state may impose malpractice liability “for all manner of activity that the state deems bad medicine,”337 it may decide “that good medical care does not require inquiry . . . regarding firearms when unnecessary to a patient’s care.”338 Under the court’s view, it is thus up to the state to determine what constitutes appropriate care.

But it is misleading to assert, as the Eleventh Circuit did, that the state imposes liability for activities that the state deems bad medicine. Rather, the state’s imposition of liability should track what the knowledge community deems bad medicine.339 Applying the knowledge community-focused theory of professional speech proposed here, the state legislature impermissibly deemed all routine inquiries concerning firearms to be irrelevant. Under this theory, it is for the professional to decide—based on the knowledge community’s insights—what constitutes relevant information within the professional-client relationship.

These examples illustrate how the exchange of information between a client and a professional suffers in the face of regulatory overreach. A focus on the role of the knowledge community’s body of knowledge brings the attendant distortions into sharp relief. As demonstrated above, the fundamental defect in these types of regulation is the direct state interference with the content of the body of professional knowledge itself.


As noted at the outset, some professionals speak a lot: “Most of what many lawyers, investment advisors, accountants, psychotherapists, and even doctors do is speech.”340 It is therefore all the more troubling that there has not yet been a comprehensive theory of professional speech advanced in the courts and the legal literature. Understanding the nature of the professions as knowledge communities allows us to reconceptualize this type of speech.

State regulation interacts with knowledge communities’ insights in multiple and varied ways. Sometimes it aligns with professional insights; sometimes it contradicts them. If state regulation aims to interfere with and alter professional knowledge, the First Amendment should protect the client’s as well as the professional’s interest in accurate communication of the knowledge community’s insights when a professional speaks.