Heim v. Daniel

81 F.4th 212

Opinion

22-1135-cv Heim v. Daniel

UNITED STATES COURT OF APPEALS FOR THE SECOND CIRCUIT

August Term, 2022

Argued: June 30, 2023 Decided: August 30, 2023 Docket No. 22-1135-cv

JOHN J. HEIM,

Plaintiff-Appellant,

— v. —

BETTY DANIEL, ADRIAN MASTERS, Defendants-Appellees.*

Before:

LYNCH, BIANCO, and PÉREZ, Circuit Judges.

This appeal concerns a First Amendment retaliation claim brought in the

*

The Clerk is respectfully directed to amend the caption to conform to the above.

Northern District of New York by Plaintiff-Appellant John Heim, an adjunct professor of economics at SUNY Albany, who attributes his failure to advance within his department to his colleagues’ unfavorable view of the methodology he employs in his scholarship. The district court (Hurd, J.) granted summary judgment to Defendants, two of Heim’s colleagues who were involved in the hiring decisions at issue. Although we disagree with much of the district court’s reasoning, we nonetheless agree with its ultimate disposition. We hold that Garcetti v. Ceballos, 547 U.S. 410 (2006), does not apply to speech related to academic scholarship or teaching, and that Heim’s speech addressed matters of public concern, but that Heim’s First Amendment claim nonetheless fails because under Pickering v. Bd. of Educ., 391 U.S. 563 (1968), a public university’s interest in deciding for itself what skills, expertise, and academic perspectives it wishes to prioritize in its hiring and staffing decisions outweighs Heim’s asserted interest in competing for academic positions unencumbered by university decision- makers’ assessment of his academic speech. We therefore AFFIRM the judgment of the district court.

PHILLIP G. STECK, Cooper Erving & Savage LLP, Albany, NY, for Plaintiff-Appellant.

SARAH L. ROSENBLUTH, Assistant Solicitor General, Buffalo, NY (Letitia James, Attorney General; Barbara D.

Underwood, Solicitor General; Andrea Oser, Deputy Solicitor General, on the brief), for Defendants-Appellees.

DARPANA M. SHETH, Washington, DC, for Amicus Curiae Foundation for Individual Rights and Expression.

3

GERARD E. LYNCH, Circuit Judge:

Plaintiff-Appellant John Heim, an adjunct professor of economics, appeals from a judgment of the United States District Court for the Northern District of New York (David N. Hurd, J.) granting summary judgment to his colleagues Betty Daniel and Adrian Masters (together, “Defendants”) who, as the relevant decision-makers in the Economics Department at the State University of New York at Albany (“SUNY Albany”1), declined to interview Heim for more desirable positions he believes he was qualified for. Heim’s challenge is premised on the allegation that Defendants rejected his candidacy in substantial part because he is a proponent of traditional Keynesian economics, an approach that Defendants consider to be outdated.

Although we accept that factual premise underlying Heim’s appeal, we disagree with the legal theory it supports: that, under the First Amendment, a public university’s hiring decisions cannot be informed by methodological preference. Rather, applying the employer/employee interest-balancing framework first set forth in Pickering v. Bd. of Educ., 391 U.S. 563 (1968), we hold

1 The institution is also referred to in the record as “The University at Albany” and, simply, “Albany.”

4

that a public university’s interest in deciding for itself what skills, expertise, and academic perspectives it wishes to prioritize in its hiring and staffing decisions outweighs Heim’s asserted interest in competing for academic positions unencumbered by university decision-makers’ assessment of the approach or methodology underlying his academic speech. We therefore AFFIRM the judgment of the district court.

BACKGROUND

I. Factual Background The following facts, which are largely drawn from Heim’s own testimony, are undisputed except where otherwise indicated, and are those that a reasonable factfinder could find, construing all ambiguities in Heim’s favor. See Cugini v. City of New York, 941 F.3d 604, 608 (2d Cir. 2019).

A. The Parties

Heim, an adjunct professor2 at SUNY Albany, initially brought this lawsuit against the entire SUNY system, SUNY Albany, its president Havidan Rodriguez,

2 In his deposition, Heim explained that although his title was technically “visiting professor,” for “payroll purposes” he was “known as an adjunct professor or lecturer.” App’x 142, 151. All parties refer to him as an “adjunct.”

5

and two members of the SUNY Albany Economics Department (the “Department”): Professors Betty Daniel and Adrian Masters. Masters has chaired the Department since 2015, and was a member of its hiring committee at all relevant times. In addition to leading the Department prior to Masters, Daniel also chaired its hiring committee at all relevant times.

B. Heim’s Professional Path Heim was a relative latecomer to academia. After receiving his Ph.D. in Political Economy from SUNY Albany in 1972, he worked in government as an econometrician (and in other similar capacities) for many years, interrupted only by his time in the mid-1980s obtaining a Master’s degree in Public Administration from Harvard. He began his teaching career in 1997, accepting a non-tenured position at Rensselaer Polytechnic Institute (“RPI”) where he eventually achieved the rank of (non-tenured) “full clinical professor.” App’x 126.3 Heim’s duties at RPI “were limited to teaching and administration,” including teaching “Master[’s] and Ph.D.-level Advanced Macroeconomics I & II and Master’s level Econometrics” and supervising both undergraduate- and

3 Because “[t]enure in that position was not available,” RPI never made any “decision on granting or denying [Heim] tenure.” App’x 528.

6

graduate-level research. Id. at 528.

Around 2012, he accepted an adjunct position at SUNY Albany with a lighter teaching load that he felt would allow him to “concentrate more on research.” Id. That research, the methodology it employs, and the difference between that methodology and the approach favored by his colleagues at SUNY Albany are at the root of this dispute.

C. Methodological Perspectives at SUNY Albany Heim describes himself as a traditional Keynesian economist. In Heim’s characterization, Keynesians “stud[y] the operation of the entire economy as a whole (in comparison to microeconomists who study the behavior of firms or sectors of the economy or particular markets),” with a particular focus on demand-driven government intervention. Appellant’s Br. 8. That perspective, he explains, derives from prominent 20th-century economist John Maynard Keynes’s view that “[w]hen consumers spend money in the economy, businesses respond by producing goods and services” to satisfy that robust aggregate demand, stimulating the macroeconomy; conversely, “when aggregate demand is lacking, the economy is depressed.” Id. For example, “[f]or Keynes, the Great Depression was caused by insufficient aggregate demand, that is[,] consumers having

7

insufficient purchasing power to drive the economy forward.” Id. Accordingly, Keynesian economics generally supports countering economic lulls with government intervention targeted at increasing consumers’ purchasing power through policies like tax cuts, heightened government spending, and increases in the money supply. That view represented a significant departure from prevailing early 20th-century perspectives, which effectively trusted to market forces to sort everything out in the long run. Keynes, of course, famously quipped that “[i]n the long run, we are all dead,” App’x 531, and the school of economics he founded enjoyed a run of prominence spanning several post-Great Depression decades.

Rampant stagflation in the 1970s inspired many economists to re-evaluate.

When traditional Keynesian methods were seen as unable to account for the oppressive mix of inflation, unemployment, and stagnant growth of the era, new approaches began to supplant Keynesianism. That reckoning was spurred by the so-called “Lucas Critique,”4 which posited that Keynesian models’ reliance on fixed coefficients to relate consumers’ disposable income to consumer spending did not account for the fact that consumers’ behaviors may vary based on their

4 Named for its creator, University of Chicago economist and Nobel laureate Robert Lucas.

8

expectations for future government interventions – for example, by responding to a short-term tax cut aimed at increasing consumer purchasing power by instead saving newfound disposable income that Keynesian models would expect them (and policymakers would prefer them) to spend. That critique demanded a new way to predict macro, economy-wide, trends based on those kinds of micro, consumer-level, variables. So was born the “micro foundations of macroeconomics” school which, in Heim’s words, attempts to “extrapolate the behavior of the macro economy from the micro economy.” Id.

One prominent technique employed under that banner, and by many post-

Lucas Critique macroeconomists, is central to this lawsuit: “dynamic stochastic general equilibrium” (or “DSGE”) modeling. Heim characterizes DSGE as “a method that uses sophisticated mathematics such as logarithms to derive the behavior of the macro economy from the micro economy.” Id. at 532. DSGE’s proponents consider it a more flexible and robust technique than its forebears. They believe that by permitting macroeconomic indicators like economic output to be linked to coefficients that are sensitive to consumer expectations, researchers are able to model economic trends more dynamically and more realistically than would be possible using traditional Keynesian tools. To Heim,

9

though, none of this is really macroeconomics at all because it does not “deal with the behavior of the economy as a whole.” Id. at 535. In his view, “you cannot simply derive the macro from the micro.” Id.

The merits of that debate5 are not for us to assess; judges are neither qualified nor commissioned to resolve academic debates among scholars in any particular discipline. What matters is that the debate exists at all, and that Heim – who practices traditional Keynesian economics6 – is on one side of it, while his colleagues at SUNY Albany – who do not – are on the other.

D. Heim’s Work at SUNY Albany According to Heim, it did not take long for his colleagues’ “hostil[ity]”

5 The parties, for example, dispute the extent to which Heim’s approach is or is not out-of-step and antiquated, precisely how ubiquitous and/or infallible the DSGE technique is or is not, and whether DSGE has invited a reversion to pre- Keynesian perspectives or is simply a technique used by modern economists of varying perspectives. Compare Appellant’s Br. 9 (“DSGE economists are said to be a modern version of neoclassical economists, the economists who emphasized market clearing before the advent of Keynes.”), with Appellee’s Br. 8 (“Contemporary Keynesian analysis . . . often embraces DSGE techniques – it does not, as plaintiff claims, stand apart from DSGE.”). 6 Heim describes his own work as “demand-driven Keynesianism, as Keynes himself would have recognized it,” employing a “highly detailed statistically based look at how the economy operates.” App’x 535.

10

towards his approach to affect the scope of his work within the Department. App’x 552. In his first semester at SUNY Albany, he taught a class on large-scale macroeconomic modeling. When Daniel took over as Chair, however, “she denied [Heim] the opportunity to continue teaching that class.” Id. at 528. Thereafter, although Heim maintained a full courseload teaching classes like “Principles of Economics” and “Economic Statistics,” he was never again assigned to teach macroeconomics.

Meanwhile, Heim continued to forge ahead with his research, although as an adjunct he was not required to do so. For example, between 2017 and 2021 he published four books studying “every fiscal stimulus program going back to 1960,” id. at 529, and as of February 2022, he had published 23 articles in economic journals. His research was publicized and celebrated by the university, including at a 2018 reception honoring several dozen “Authors & Artists” from across the College of Arts and Sciences, and a 2018 “Celebration of Scholarship” event featuring presentations from 45 SUNY Albany scholars across a range of disciplines. Id. at 573, 575.

Heim remains employed as an adjunct at SUNY Albany. That is not because he always wanted to continue in that capacity, however. To the contrary,

11

at the heart of this lawsuit are three other positions in the Department that he was (or would have been) interested in, but for which he was never interviewed.

1. The 2013 Position

First, in December 2013, Heim learned of a tenure-track macroeconomics opening that the Department was in the process of filling. That was not an everyday occurrence; the Department is small, employing fewer than 20 faculty, only three or four of whom specialize in macroeconomics. Heim quickly reached out to Daniel by email to express his interest:

I just found out last Friday at our holiday party that we were recruiting for a macroeconomics/Money and banking position. No one had mentioned this to me before, and I would have been interested in applying, had I known . . . .

That said, my research interests are principally in large scale econometric modeling of the macroeconomy. . . . I am finalizing this year work on a 30-40 equation model of the macroeconomy, which could become known as the “SUNY Albany Econometric Model” . . . . I think my area of my research is likely to undergo a resurgence of interest in the near future[.]

Of course, the department here at SUNY has already made a substantial investment in developing a micro foundations focus, and may be more interested in adding depth to its capacity in this area. Though I have taught graduate-level micro foundations courses, it is of course, an area substantially different from my own area

12

of interest and expertise. If this is the direction the department sees itself headed in, certainly someone far different than me would a better match for the department and its plans than I would be.

Id. at 588-89. In response, Daniel confirmed that the Department was indeed looking for someone with a micro foundations focus, but that it nonetheless supported Heim’s ongoing, divergent research:

Thanks for your interest in the macro position. You are correct that we are heavily invested in micro foundations of macro as this is the research trend in the top macro and general field journals. Since we expect our faculty to publish in these journals, we do intend to continue with this direction.

I did consult with Adrian [Masters] . . . and we agree with you that your research differs from the course of research we want to pursue. I expect that we will hire a junior person recently trained in these techniques. I do want to encourage you to continue your research. It just does not match with the direction we are taking macro research in the Department.

Id. at 588. The Department ultimately hired Yue Li, a recent Ph.D. with a micro foundations research focus who, in Heim’s estimation, had limited familiarity with traditional Keynesian modeling.

2. The 2016 Position

A few years later, in 2016, Heim learned that the Department had hired for

13

a full-time (non-tenure-track) lecturer position in financial economics – and that despite Heim’s background in that area, he had again never been approached about the opportunity, and in fact did not even learn of it until the job had already been filled by a fellow adjunct; the Department had sought and obtained a waiver from the university to make the offer to Lewis Segal, a former Goldman Sachs employee with a Ph.D. in economics from Northwestern, without going through the normal competitive hiring process. Although he concedes that Segal was qualified for the role, Heim maintains that he himself was “equally qualified, maybe a little bit more.” Id. at 199.

3. The 2017 Position

In August 2017, while having lunch with Daniel and Masters, Heim learned of another open tenure-track macroeconomics position in the Department. In October, he applied. But apart from an initial confirmation email from the university that his application had been received, several weeks passed without any response. Finally, in December, Heim followed up with Masters (now Chair of the Department) who told Heim that he was not being considered because his recommendations had not been received in time. The mix-up was apparently attributable to a technical snafu at the third-party online system

14

responsible for compiling application materials, leaving a frustrated Heim to wonder why no one had ever reached out to him about the missing documents even though it was widely known within the Department that he was applying.

But all of that was a “red herring” anyway. Id. at 542. Eventually, after being pressed for further explanation by Heim, Masters put the “real reason,” id., in an email:

The department respects you and respects the fact that you continue to conduct your research as an adjunct professor. The technical reason we put down for not interviewing you is that we did not have letters and I told you that at the party. At the party I also gave you the “Cliff Notes” on the general feeling of the committee towards your application. Until now I have not given you chapter and verse because I am still happy to have you as a member of the department but you have now forced my hand on this. The fact is the committee did discuss your application and, as with many applicants with a track record, concluded that regardless of what any letters said we would not interview you for the position. Here is why:

We are looking for someone who:

1. Can teach and train students to conduct research in the modern (i.e. post Lucas Critique)

macroeconomics that by your own admission everyone else but you and Ray Fair do[es]. Nothing in your credentials supports that possibility.

2. Has a reasonable expectation of making tenure within the department within the 6 year tenure track

15

window. Here again the fact that you have a record speaks for itself. Nothing a letter writer can do can change that. The journals in which you have published do not achieve the standard that we expect for tenure. (Typically 4 articles in journals at the level of top field e.g. Journal of Monetary Economics or the International Economic Review.[)]

3. Has sufficient synergies with our research agendas that we can learn from them and them from us with the possibility of constructive collaboration. Your work is not consistent with that expectation. Indeed, we rejected a number of applications that met the first 2 criteria above but they do New Keynesian macro that we do not appreciate.

I hope you are not too discouraged by this and continue to teach and do your work within the department as an adjunct professor but the fact is that you will not be hired for this job.

Id. at 597-98. Heim asked Masters to reconsider, “noting that DSGE economics had come under heavy criticism within the profession” including from two prominent economists, and urging that the Department would “benefit from having a recognized alternative, or at least by having a healthy debate on these issues in economics.” Id. at 545. Masters responded by simply dismissing those prominent critics as “old guys!” Id. The Department ultimately hired Ben Griffy, a recent Ph.D. whose research, like Masters’s, focused in part on macroeconomics and the labor market and who Heim believes also had limited familiarity with

16

traditional Keynesian methods.

II. Procedural Background A. Early Proceedings

Heim filed this lawsuit in July 2018, initially asserting three causes of action: (1) a claim for damages pursuant to 42 U.S.C. § 1983 against Daniel and Masters for “violat[ing] plaintiff’s right of freedom of speech while acting under color of” state law; (2) a claim pursuant to § 1983 for injunctive relief against SUNY Albany President Rodriguez in the form of a court order to “prevent ongoing discrimination against Keynesian economists” in violation of the First Amendment; and (3) an age discrimination claim under New York State’s Human Rights Law. App’x 64-67.7 In September 2018, Defendants moved to dismiss the Complaint, and in a ruling from the bench, the district court granted that motion in part, dismissing the second and third causes of action. Along the way, Heim abandoned any claim

7 The Complaint also foreshadowed that “[u]pon issuance of a right to sue letter by [the U.S. Equal Employment Opportunity Commission],” Heim would seek leave to add an age discrimination claim under the Age Discrimination in Employment Act. App’x 67. That never happened, and the district court therefore “decline[d] to consider any claim under the ADEA.” Heim v. Daniel, No. 1:18-CV-836, 2022 WL 1472878, at *1 n.2 (N.D.N.Y. May 10, 2022). Heim does not challenge that decision on appeal.

17

against President Rodriguez, SUNY Albany, and the broader SUNY system, opting to pursue only the First Amendment retaliation claim against Daniel and Masters. Heim does not challenge any of that on appeal.

B. Summary Judgment

His appeal targets what happened next. After the parties exchanged discovery on Heim’s sole remaining claim – First Amendment retaliation – Defendants moved for summary judgment in January 2022.

In May 2022, the district court granted that motion and entered judgment for Defendants, holding in relevant part that the First Amendment did not insulate Heim’s academic speech from adverse action by his public university employer. Heim v. Daniel, No. 1:18-CV-836, 2022 WL 1472878, at *14 (N.D.N.Y. May 10, 2022). It arrived at that conclusion applying two separate First Amendment frameworks in the alternative and reaching the same result under both, reasoning (1) that academic research and writing are part of a public university professor’s “official duties,” dooming Heim’s claims under Garcetti v. Ceballos, 547 U.S. 410, 421 (2006); and (2) that even if Garcetti did not apply, Heim’s “complex statistical modeling intended for consumption by a relatively narrow audience” did not address any matter of public concern, warranting

18

dismissal under Pickering. Heim, 2022 WL 1472878, at *14. The court also offered yet another alternative basis for its ruling: that because Heim did not meet the hiring committee’s stated qualifications for the position, “a reasonable jury would be compelled to conclude that defendants would not have promoted plaintiff . . . regardless of plaintiff’s protected speech.” Id. at *16.8 DISCUSSION

Heim challenges each of those points on appeal. He contends that the district court erred by, among other things, (1) misconstruing the role that Heim’s Keynesian viewpoints played in his colleagues’ hiring decisions; (2) applying Garcetti to academic speech at all; (3) concluding, under Pickering, that Heim’s speech did not address any matter of public concern; and (4) failing to properly “balance the employee’s First Amendment interests against the employer’s

8 The court also (1) concluded “that to the extent Heim’s § 1983 claim is based on the general First Amendment right of ‘academic freedom,’ that claim must be dismissed, Heim, 2022 WL 1472878, at *7; (2) rejected Heim’s argument that this case should be analyzed under a “limited public forum” analysis, id. at *17 (adding that “‘forums’ are places, not people”); and (3) remarked that although its conclusions as to the merits made it “unnecessary” to address qualified immunity, that defense “almost certainly applies” to Defendants’ conduct in this case, id. at *16 n.14. Heim does not challenge any of that on appeal, and because we resolve the sole claim he continues to press on other grounds, we have no occasion to express any view as to the merits of those observations.

19

legitimate interests.” Appellant’s Br. 49.

Heim’s arguments are not altogether without merit. Although we affirm the district court’s ultimate disposition, we disagree with much of its reasoning. In particular, we disagree with its conclusions that Heim’s speech did not substantially motivate Defendants’ rejection of his candidacy, and that Heim’s speech did not address matters of public concern. We also join several other Circuits in recognizing that First Amendment retaliation claims founded upon speech that relates to academic scholarship or teaching are properly evaluated under Pickering’s employer/employee interest-balancing framework irrespective of whether, under Garcetti, that speech was part of an employee’s official duties. Ultimately, though, because Heim’s First Amendment retaliation claim cannot withstand that balancing test, it nonetheless fails.

I. Legal Standards A. Standard of Review

We review a district court’s grant of summary judgment de novo, “resolving all ambiguities and drawing all reasonable inferences” in favor of the non-movant. Woolf v. Strada, 949 F.3d 89, 92-93 (2d Cir. 2020) (internal quotation marks omitted). We affirm “only if there is no genuine issue of material fact and

20

the prevailing party was entitled to judgment as a matter of law.” Harris v. Miller, 818 F.3d 49, 57 (2d Cir. 2016) (internal quotation marks omitted).

B. First Amendment Retaliation To establish a prima facie First Amendment retaliation claim, a plaintiff must show (1) “that the speech or conduct at issue was protected” from the particular retaliatory act alleged; (2) that the retaliatory act qualifies as an “adverse action [taken] against the plaintiff”; and (3) “that there was a causal connection between the protected speech and the adverse action.” Shara v. Maine-Endwell Cent. Sch. Dist., 46 F.4th 77, 82 (2d Cir. 2022) (internal quotation marks omitted). However, “[e]ven if the plaintiff makes out a prima facie retaliation claim, a government defendant may still receive summary judgment if it establishes its entitlement to a relevant defense.” Anemone v. Metro. Transp. Auth., 629 F.3d 97, 114 (2d Cir. 2011).

The district court held, and Defendants do not dispute on appeal, that the Department’s refusal to hire Heim for the 2017 position qualifies as an adverse action.9 We therefore address below only the other, contested, elements of a First

9 Nor, for the most part, could Defendants dispute that point. “In a First Amendment retaliation case, a government employer’s response to speech

21

Amendment retaliation claim.

constitutes an adverse action if it would deter a similarly situated individual of ordinary firmness from exercising his or her constitutional rights.” Connelly v. County of Rockland, 61 F.4th 322, 325 (2d Cir. 2023) (internal quotation marks omitted). This is a less “demanding” (that is, easier to satisfy) standard than the “materially adverse change in the terms and conditions of employment” test that we use to determine an “adverse action” in other employment contexts. E.g., Zelnik v. Fashion Inst. of Tech., 464 F.3d 217, 225 (2d Cir. 2006) (internal quotation marks omitted). For purposes of First Amendment retaliation, an adverse action may include, among other things, “discharging, . . . demoting, reducing the pay, or reprimanding an employee” and, as especially relevant here, both “refusing to hire” and “refusing to promote” an applicant. Montero v. City of Yonkers, 890 F.3d 386, 401 (2d Cir. 2018). Whether the 2017 decision not to interview Heim is better understood as a refusal to promote – a characterization Defendants have questioned, see App’x 872 (Daniel testifying: “The position of adjunct does not have the concept of promotion. An adjunct is a dead-end position. . . . The assistant professor position is a completely different position.”) – or a refusal to hire, it indisputably satisfies the “adverse action” requirement.

The earlier acts of alleged retaliation present closer questions. The district court held that the circumstances surrounding the 2013 and 2016 hires did not amount to “sufficiently adverse actions,” emphasizing that Heim did not actually apply for either position, and brushing aside Heim’s arguments that “Daniel should not have discouraged him from applying” in 2013 and that the Department had “followed an improper internal or administrative process” in 2016. Heim, 2022 WL 1472878, at *14. Because we reject Heim’s claim for other reasons, we need not decide whether those conclusions (which Heim does not challenge on appeal) faithfully reflect our permissive and fact-specific “adversity” standards in this context. See Hoyt v. Andreucci, 433 F.3d 320, 328 (2d Cir. 2006) (noting that “lesser actions may meet the adversity threshold” under what is “a heavily fact-specific, contextual determination”).

22

II. Causation We begin with the causation requirement. To show causation, “a plaintiff must show that the protected speech was a substantial motivating factor in the adverse employment action.” Smith v. County of Suffolk, 776 F.3d 114, 118 (2d Cir. 2015) (internal quotation marks omitted). Nevertheless, “[b]ecause protected speech could not substantially cause an adverse action if the employer would have taken that action in any event, a defendant can rebut a prima facie showing of retaliation by demonstrating by a preponderance of the evidence that it would have taken the same adverse employment action even in the absence of the protected conduct.” Nagle v. Marron, 663 F.3d 100, 111-12 (2d Cir. 2011) (internal quotation marks omitted), citing Mt. Healthy City Sch. Dist. Bd. of Educ. v. Doyle, 429 U.S. 274, 285-86 (1977) (“The constitutional principle at stake is sufficiently vindicated if such an employee is placed in no worse a position than if he had not engaged in the conduct.”).

With those principles in mind, Defendants insist that they “would have declined to interview plaintiff even in the absence of any hostility to his allegedly protected speech because he was not qualified for the position.” Appellees’ Br. 25. The centerpiece of their argument is that Heim lacked any of the qualifications

23

articulated in Masters’s December 2017 email, which they characterize as: (1) “an ability to teach and train students in contemporary research techniques”; (2) “a research agenda that would permit collaboration with the Department’s other faculty members”; and (3) “a publication record suggesting a strong likelihood of receiving tenure.” Id. at 15-16.

But any attempt to construe those three requirements as content-neutral cannot survive even a superficial look at Masters’s actual email. Justifiably or not, the requirements were soaked in hostility to Keynesianism. At the outset, when describing the requisite ability to teach and train students in contemporary research, Masters explicitly defined “modern” as the sort of “post Lucas Critique[] macroeconomics that by your own admission everyone else but you and [one other economist] do[es].” App’x 597. In other words, to the Department, “modern” was antonymous to “Keynesian.” Next, Masters’s explanation for why Heim’s work was “not consistent” with the collaboration-friendly research requirement was that the people in the Department Heim would be expected to collaborate with “do not appreciate” Keynesian research perspectives. Id. at 598. In fact, the Department had already “rejected” other applicants who “met the [other] criteria” but “do New Keynesian macro.” Id. Finally, the third job

24

requirement, setting forth the Department’s publication expectations, is at least on its face more content-neutral. But even if we take it at face value, the record suggests that an applicant’s failure to meet the Department’s preferred publication history standards was not on its own disqualifying. For example, at the time he was hired, Griffy had also never published (except perhaps as a research assistant) in the top journals Masters invoked in his email. Nor had Li. Heim contends, moreover, that Defendants’ “self-reinforcing” assessment of the “top” journals in the field, id. at 547, is also colored by the fact that those particular journals “do not publish non-DSGE Keynesian articles,” id. at 543-44.

Whatever their merits, those assessments – which, as we discuss below, were permissible for other reasons – firmly foreclose any attempt to escape liability on causation grounds. It is not sufficient for Defendants to demonstrate that they considered Heim unqualified because, when we “focus[] precisely” on Defendants’ motivations, we cannot ignore the role their views on Heim’s speech played in shaping why they considered him unqualified. Anemone, 629 F.3d at 114 (internal quotation marks omitted). The undisputed facts compel the conclusion that the content of Heim’s academic scholarship was central to Defendants’ assessment of his qualifications. It is therefore impossible for a reasonable jury to

25

conclude that Defendants “would have taken the same action in the absence of” that speech. Id. at 115 (internal quotation marks omitted).

III. Extent of First Amendment Protection As a result, the outcome here hinges on whether the First Amendment forecloses decision-makers within a public university from making hiring decisions based on such content-driven assessments.

“Regardless of the factual context, we have required a plaintiff alleging retaliation to establish speech protected by the First Amendment” in that relevant context. Weintraub v. Bd. of Educ., 593 F.3d 196, 200 (2d Cir. 2010) (internal quotation marks omitted). That requirement is somewhat elusive in the public employment setting, where employees must generally “‘by necessity . . . accept certain limitations’” on their First Amendment freedoms because, as government insiders, their speech “can ‘contravene governmental policies or impair the proper performance of governmental functions.’” Id. at 201 (alteration in original), quoting Garcetti, 547 U.S. at 418-19; see Connick v. Myers, 461 U.S. 138, 143 (1983) (articulating “the common sense realization” underpinning Pickering “that government offices could not function if every employment decision became a constitutional matter”). Still, a public employee “does not relinquish all First

26

Amendment rights otherwise enjoyed by citizens just by reason of his or her employment.” City of San Diego v. Roe, 543 U.S. 77, 80 (2004).

A. First Amendment Frameworks Over time, guided by those principles, the Supreme Court has assembled a two-pronged test to determine whether the First Amendment insulates a public employee from an alleged retaliatory act by their employer: (1) “whether the employee spoke as a citizen on a matter of public concern” and (2) if so, whether the employer had “an adequate justification for treating the employee differently based on his or her speech from any other member of the general public.” Montero v. City of Yonkers, 890 F.3d 386, 395 (2d Cir. 2018) (internal quotation marks and alteration omitted), citing Garcetti, 547 U.S. at 418 and Pickering, 391 U.S. at 568. We begin this portion of our analysis by determining how those factors apply in the “unique” context of academia. Adams v. Trustees of the Univ. of N.C.-Wilmington, 640 F.3d 550, 564 (4th Cir. 2011).

1. Pickering and Garcetti For its part, the district court applied two different analyses in the alternative. The older framework it applied was the line of authority extending from Pickering, looking only to (1) whether the employee was speaking on a

27

matter of “public concern,” Connick, 461 U.S. at 146 (under Pickering, if the speech at issue “cannot be fairly characterized as constituting speech on a matter of public concern, it is unnecessary for us to scrutinize the reasons for [the adverse action]”); and if so, (2) whether the employer’s reaction to that speech was nonetheless justified, balancing the employee’s interest “in commenting upon matters of public concern” against the public employer’s interest in “promoting the efficiency of the public services it performs through its employees.” Pickering, 391 U.S. at 568; see Matthews v. City of New York, 779 F.3d 167, 172-73 (2d Cir. 2015).

The district court also applied the framework that the Supreme Court, nearly 40 years after Pickering, articulated in Garcetti. The Garcetti Court “narrowed” the scope of First Amendment protection for public employees, inserting an additional rung into that longstanding Pickering framework. Weintraub, 593 F.3d at 201 (internal quotation marks omitted). “[W]hen public employees make statements pursuant to their official duties,” it explained, “the employees are not speaking as citizens for First Amendment purposes, and the Constitution does not insulate their communications from employer discipline.” Garcetti, 547 U.S. at 421. Thus, in effect, under Garcetti, “employee speech” – i.e.,

28

speech pursuant to a public employee’s official duties – is weeded out before even reaching the Pickering framework. Lane v. Franks, 573 U.S. 228, 237 (2014). Only where the speech at issue is so-called “citizen speech” does a court then proceed, under Pickering, to determining whether the speech addressed a matter of public concern and, if so, to balancing the relevant employee and employer interests. Id.

2. Garcetti Does Not Apply Here This Court, however, has yet to decide whether Garcetti even applies in the “special” context of academia. E. Hartford Ed. Ass’n v. Bd. of Ed., 562 F.2d 838, 842- 43 (2d Cir. 1977). The Garcetti Court deliberately left that question unanswered. Writing for the majority – in response to a passage in Justice Souter’s dissent expressing his “hope” that “today’s majority does not mean to imperil First Amendment protection of academic freedom in public colleges and universities,” Garcetti, 547 U.S. at 438 (Souter, J., dissenting) – Justice Kennedy expressly acknowledged that the Court’s holding might not apply to academic speech. “There is some argument,” he allowed, “that expression related to academic scholarship or classroom instruction implicates additional constitutional interests that are not fully accounted for by this Court’s customary employee-speech jurisprudence.” Id. at 425 (majority opinion). That was the extent of the majority’s

29

discussion of that question; the dispute in Garcetti itself did not arise in a university setting, but rather from a deputy district attorney’s claim that he had faced retaliation for voicing (in an official memo that it was his job to write) his critiques of a prosecution brought by his own office. See id. at 421. Consequently, the Garcetti majority “need not, and for that reason d[id] not, decide whether [its analysis] would apply in the same manner to a case involving speech related to scholarship or teaching.” Id. at 425. The Supreme Court has yet to revisit that question.

We, in turn, have recognized that “it is an open question in this Circuit whether Garcetti applies to classroom instruction,” Lee-Walker v. New York City Dep’t of Educ., 712 F. App’x 43, 45 (2d Cir. 2017) (summary order) (internal quotation marks and alteration omitted), and district courts in our Circuit have varied in their application of Garcetti to various other kinds of “academic speech,” Schulz v. Commack Union Free Sch. Dist., No. 21-CV-5646-RPK, 2023 WL 2667050, at *7 (E.D.N.Y. Mar. 28, 2023) (collecting cases). We are not the only Circuit to flag the uncertain implications of Garcetti’s academic carveout. See, e.g., Gorum v. Sessoms, 561 F.3d 179, 186 n.6 (3d Cir. 2009) (“The full implications of the

30

Supreme Court’s statements in Garcetti regarding speech related to scholarship or teaching are not clear.” (internal quotation mark omitted)).

But other Circuits have started to answer that question. The Fourth Circuit has cautioned that “[a]pplying Garcetti to the academic work of a public university faculty member . . . could place beyond the reach of First Amendment protection many forms of public speech or service a professor engaged in during his employment.” Adams, 640 F.3d at 564. And so, relying on the Garcetti Court’s “clear reservation” of the issue, it opted to apply the approach established in its own binding, pre-Garcetti authority (based on Pickering) to a public university professor’s First Amendment retaliation claim predicated on speech that “involve[d] scholarship and teaching.” Id. at 562-63.

The Ninth Circuit soon followed suit, agreeing that “if applied to teaching and academic writing, Garcetti would directly conflict with the important First Amendment values previously articulated by the Supreme Court,” and instead electing to “us[e] the analysis established in Pickering.” Demers v. Austin, 746 F.3d 402, 411-12 (9th Cir. 2014). More recently, the Sixth Circuit construed “the academic-freedom exception to Garcetti” to extend to “all classroom speech related to matters of public concern, whether that speech is germane to the

31

contents of the lecture or not,” enumerating a number of “critical interests” unique to the university setting that demand more “robust speech protections” there than in other public employment contexts. Meriwether v. Hartop, 992 F.3d 492, 507 (6th Cir. 2021); see also Buchanan v. Alexander, 919 F.3d 847, 852-53 (5th Cir. 2019) (applying Pickering and Connick rather than Garcetti – without mentioning the latter – to a university professor’s First Amendment claim centered on classroom speech and conduct, emphasizing that “academic freedom is a special concern of the First Amendment” (internal quotation marks omitted)).10

10 Although it has not expressly split from these other Circuits in any directly analogous case, the Seventh Circuit has been somewhat more hesitant to sidestep Garcetti in adjacent contexts. See Mayer v. Monroe County Cmty. Sch. Corp., 474 F.3d 477, 478-80 (7th Cir. 2007) (applying Garcetti to a First Amendment claim brought by an elementary school teacher who claimed she was fired after she “took a political [anti-Iraq war] stance during a current-events session in her class,” but caveating that “[h]ow much room is left for constitutional protection of scholarly viewpoints in post-secondary education was left open in Garcetti . . . and need not be resolved today”); Renken v. Gregory, 541 F.3d 769, 770, 773 (7th Cir. 2008) (applying Garcetti to a First Amendment claim brought by a university professor who alleged he was retaliated against after he “complained about the University’s use of grant funds”); see also Porter v. Bd. of Trustees of N. Carolina State Univ., 72 F.4th 573, 578, 583 (4th Cir. 2023) (applying Garcetti to a professor’s “sarcastic[]” email to colleagues that voiced his concern that a faculty search committee had “‘cut corners’ in vetting a candidate ‘out of a desire to hire a Black scholar whose work focused on racial issues,’” reasoning that the email

32

We agree with those Circuits’ treatment of Garcetti. It cannot reasonably be disputed that the Supreme Court reserved whether Garcetti’s “official duties” framework applied to “case[s] involving speech related to scholarship or teaching” by public employees. 547 U.S. at 425. Likewise, no one disputes that the speech in this case relates to scholarship or teaching; thus, by its own terms, Garcetti does not bind us in this case. In other words, Garcetti leaves undisturbed our own pre-Garcetti authority in cases involving scholarship and teaching by professors at public universities. Those cases applied Pickering and its progeny. See, e.g., Blum v. Schlegel, 18 F.3d 1005, 1011 (2d Cir. 1994); see also Dube v. State Univ. of N.Y., 900 F.2d 587, 598 (2d Cir. 1990).11 Of course, the Garcetti Court also did not affirmatively announce that its

amounted to no more than “an unprofessional attack” on a colleague that was “plainly . . . unrelated to [the professor’s] teaching or scholarship” and was therefore not subject to the Fourth Circuit’s decision in Adams). 11 Notably, in addition to their weighty discussion of concepts like academic freedom, that is also how the Fourth and Sixth Circuits approached this issue – treating Garcetti as non-binding in this very specific context, and relying instead on their own pre-Garcetti authority. See Adams, 640 F.3d at 564, citing, e.g., Boring v. Buncombe County Bd. of Educ., 136 F.3d 364, 368 (4th Cir. 1998); Meriwether, 992 F.3d at 505, citing, e.g., Hardy v. Jefferson Cmty. Coll., 260 F.3d 671, 678 (6th Cir. 2001).

33

“official duties” framework does not apply to speech related to scholarship or teaching. Nor have we addressed the matter in any post-Garcetti opinion. There is therefore no binding authority that would preclude us from importing that framework into academia if we felt that Garcetti (or other subsequent precedents) had injected new principles into our understanding of the First Amendment that required that result. See Deem v. DiMella-Deem, 941 F.3d 618, 623 (2d Cir. 2019) (“[A] published panel decision is binding on future panels unless and until it is overruled by the Court en banc or by the Supreme Court.” (internal quotation marks omitted)).

But we do not; we in fact believe quite the opposite. Key to the Garcetti Court’s reasoning was its assessment that, because the memo at issue there was authored pursuant to the employee’s official job duties, the employee’s speech “owe[d]” its very “existence” to those job duties. 547 U.S. at 421-22. For that reason, it continued, imposing restrictions on that speech would not “infringe any liberties the employee might have enjoyed as a private citizen” but rather would “simply reflect[] the exercise of employer control over what the employer itself has commissioned or created.” Id. That makes perfect sense in the typical government hierarchy, where the purpose of an employee’s speech is to further

34

the ends of the employer, on a course charted by someone higher up the chain-of- command, and where the wrong word at the wrong time risks “contraven[ing] governmental policies” or “impair[ing] . . . governmental functions.” Id. at 419.

The problem with applying that reasoning here is that professors at public universities are paid – if perhaps not exclusively, then predominantly – to speak, and to speak freely, guided by their own professional expertise, on subjects within their academic disciplines. See Sweezy v. New Hampshire, 354 U.S. 234, 250 (1957) (“Teachers and students must always remain free to inquire, to study and to evaluate, to gain new maturity and understanding; otherwise our civilization will stagnate and die.”). That is their job. They “necessarily speak and write pursuant to [their] official duties.” Garcetti, 547 U.S. at 438 (Souter, J., dissenting) (internal quotation marks and alteration omitted).12 And their university’s

12 For those reasons, applying Garcetti to this case would pose a major obstacle to Heim’s claim. Heim disagrees, emphasizing that he did not undertake his research “pursuant to government policy” nor as “part of his ordinary duties” as an adjunct lecturer. Appellant’s Br. 42-43. But “under the First Amendment, speech can be ‘pursuant to’ a public employee’s official job duties even though it is not required by, or included in, the employee’s job description, or in response to a request by the employer.” Weintraub, 593 F.3d at 203. Rather, courts are instructed to “examine the nature of the plaintiff’s job responsibilities, the nature of the speech, and the relationship between the two, along with other contextual factors such as whether the plaintiff’s speech was also conveyed to the public.”

35

“governmental function[]” is to provide them a forum to do so. Id. at 419 (majority opinion). Garcetti’s bar on First Amendment protection for any “official-duty” speech would thus have the effect of exiling all public-university faculty scholarship and instruction from the shelter of the First Amendment.

That cannot be. It certainly cannot be squared with the Supreme Court’s long-professed, “deep[] commit[ment] to safeguarding academic freedom” as “a special concern of the First Amendment.” Keyishian v. Bd. of Regents of Univ. of State of N.Y., 385 U.S. 589, 603 (1967); see also Sweezy, 354 U.S. at 250; Regents of Univ. of Michigan v. Ewing, 474 U.S. 214, 226 (1985); Kennedy v. Bremerton Sch. Dist., 142 S. Ct. 2407, 2424 (2022). It was that very commitment that inspired the

Shara, 46 F.4th at 83 (internal quotation marks omitted); see Ross v. Breslin, 693 F.3d 300, 306 (2d Cir. 2012) (“The inquiry . . . is not susceptible to a brightline rule.”). The contextual factors here strongly suggest that Heim’s scholarship, though not required, was nonetheless pursuant to his official duties considering (1) that there was at least some overlap between his research focus and the material he was assigned to teach; (2) that he discussed his research ideas with his colleagues in professional settings; (3) that the university conveyed his research to the public by honoring him at events celebrating his publications; (4) that he repeatedly leveraged his research to try to affect his work scope and position within the Department; and most significantly, (5) that research would have been a requirement of at least some of the positions for which he applied, and the Department’s decision not to interview him for those jobs, based in substantial part on his research focus, are the very alleged adverse actions at the center of this lawsuit.

36

Garcetti Court to carve academic scholarship and instruction out from its analysis in the first place. See 547 U.S. at 425; id. at 439 (Souter, J., dissenting). And to the extent the Garcetti Court left that door ajar, we echo the reasons several of our sister Circuits have expressed for closing it. We agree that “[t]he need for the free exchange of ideas in the college classroom is unlike that in other public workplace settings,” we agree that a professor’s academic speech is “anything but speech by an ordinary government employee,” Meriwether, 992 F.3d at 507, and therefore, fundamentally, we agree that “‘given the important purpose of public education and the expansive freedoms of speech and thought associated with the university environment,’ . . . Garcetti does not – indeed, consistent with the First Amendment, cannot – apply to [a public university professor’s] teaching and academic writing,” Demers, 746 F.3d at 411-12, quoting Grutter v. Bollinger, 539 U.S. 306, 329 (2003).13

13 We note that our discussion here focuses specifically on the public scholarship of university professors, and we express no view as to how or whether Garcetti might apply to, say, an elementary school teacher’s speech, or speech unrelated to a professor’s teaching or scholarship, the potentially distinguishable academic contexts in which other Circuits have applied Garcetti. See supra note 9 (citing Mayer, 474 F.3d at 478; Renken v. Gregory, 541 F.3d at 770, 773; Porter, 72 F.4th at 578, 583).

37

Accordingly, we join those Circuits in holding that we must evaluate claims founded on such speech outside of Garcetti’s“official duties” framework.

B. Applying Pickering

But that holding still does not resolve whether any particular burden imposed by a public university on a professor’s speech is permissible. Instead, in Garcetti’s absence, we are left with the line of authority extending from Pickering that instructs us to ask (1) whether the employee is speaking on a matter of “public concern,” and if so (2) “whether the relevant government entity had an adequate justification for treating the employee differently from any other member of the public based on the government’s needs as an employer” – balancing “the interests of the employee, as a citizen, in commenting upon matters of public concern and the interest of the State, as an employer, in promoting the efficiency of the public services it performs through its employees.” Matthews, 779 F.3d at 172-73 (internal quotation marks omitted).

1. Matter of Public Concern As a preliminary matter, we disagree with the district court’s assessment that because “Heim’s academic writings about Keynesian economic concepts . . . concern complex statistical modeling intended for consumption by a relatively

38

narrow audience,“ they “do not qualify as speech on matters of ‘public concern.’” Heim, 2022 WL 1472878, at *14. The fact that Heim’s research is neither targeted towards nor consumed by the general public may well suggest that the public is not especially concerned with him, or with his work. But that has little to do with whether that work addresses matters that concern the public. That inquiry measures neither popularity nor technical complexity; rather, it is a question of law to be determined based on “the content, form, and context of a given statement, as revealed by the whole record,” Connick, 461 U.S. at 147-48 & n.7, looking to, among other things, “whether the speech . . . had a broader public purpose,” Singer v. Ferro, 711 F.3d 334, 339 (2d Cir. 2013) (internal quotation marks omitted), and whether the speech can “be fairly considered as relating to any matter of political, social, or other concern to the community,” Lane, 573 U.S. at 241 (internal quotation marks omitted).

Here, once again, the special academic setting of this dispute guides our analysis. We have recognized that underlying Pickering’s “public concern” requirement is “the principle that debate on public issues should be uninhibited, robust, and wide-open.” Singer, 711 F.3d at 342, quoting New York Times Co. v.

39

Sullivan, 376 U.S. 254, 270 (1964). Nowhere is it more important to safeguard that interest, nor more difficult to confidently declare that it is not implicated, than in academia, where disputes about the best theoretical approaches or most productive avenues of research abound within and across countless disciplines. Some of those debates may strike outsiders as arcane, inconsequential, or even “trivial.” Demers, 746 F.3d at 413. But the entire premise powering academic freedom is that the advancement of the arts and sciences is of long-term value to society, and that the benefits of academic scholarship are no less valuable even though the eventual benefits of particular works may be unexpected, indirect, or diffuse. See Sweezy, 354 U.S. at 250. Scholarly debates in theoretical physics, evolutionary biology, literary studies, or many other fields could produce lawsuits just like this one, and we would be hard-pressed to declare that the speech at issue in any of those cases does not “relat[e] to any matter of political, social, or other concern to the community.” Lane, 573 U.S. at 241 (internal quotation marks omitted). “Recognizing our limitations as judges, we should hesitate before concluding that academic disagreements about what may appear to be esoteric topics are mere squabbles over jobs, turf, or ego.” Demers, 746 F.3d

40

at 413.

But we need not decide today whether all academic scholarship satisfies that inclusive standard, because the scholarship at issue here comfortably does so. We have recognized that, whatever other speech may also qualify, “discussion regarding current government policies and activities is perhaps the paradigmatic matter of public concern.” Harman v. City of New York, 140 F.3d 111, 118 (2d Cir. 1998) (internal quotation marks and alteration omitted). That is precisely what Heim and other macroeconomists do: they discuss sweeping questions of economic policy, analyze macroeconomic conditions, and debate the government’s proper role in shaping those conditions. Their work may perhaps be unlikely to attract a broad audience, but it nonetheless serves a broad “public purpose,” targeting matters of political, social, and public policy salience. Singer, 711 F.3d at 339; see Lane, 573 U.S. at 241. That is more than sufficient to clear, with plenty of room to spare, the “public concern” bar, and we therefore proceed to the final step under Pickering: balancing the employer and employee interests.

2. Balancing the Interests Turning last to that pivotal step, “[b]ecause of the infinite variety of factual

41

circumstances in which such conflicts might arise, the [Supreme] Court has not announced a general standard in this area, but has instead relied upon an identification and weighing of competing interests on a case-by-case basis.” Locurto v. Giuliani, 447 F.3d 159, 173 (2d Cir. 2006) (internal quotation marks omitted). For our part, we have recognized that “[t]he extent of permissible regulation of State employee speech . . . depends both on the nature of the speech and on the nature of the public services performed through a particular employee or class of employees” and, accordingly, that the appropriate balance between employer and employee interests “may differ with respect to different types of State employees.” Blum, 18 F.3d at 1011. To that end, in a case like this, we must undertake a “nuanced consideration of the range of issues that arise in the unique genre of academia.” Adams, 640 F.3d at 564.

In that special context, we balance educators’ interest in speaking on matters of public concern against the interests of their state-school employers, taking care to “protect the First Amendment principles that underlie both [an employee’s] interest in free speech and [a university’s] underlying mission.” Blum, 18 F.3d at 1011 (emphases added); see also E. Hartford Ed. Ass’n, 562 F.2d at

42

843 (exalting academia as “a special profession essential to the public welfare” (internal quotation marks and alteration omitted)).

We have already said much about the employee’s interests. No one disputes the wealth of authority championing individual educators’ interest in academic freedom and establishing, broadly, that the First Amendment “tolerates neither laws nor other means of coercion, persuasion or intimidation ‘that cast a pall of orthodoxy’ over the free exchange of ideas in the classroom.’” Dube, 900 F.2d at 598, quoting Keyishian, 385 U.S. at 603. Typically, though, those cases operate as “a caution to governmental administrators not to discipline a college teacher for expressing controversial, even offensive, views,” or for criticizing their employer, or for speaking in a way that may upset or disturb their students. E.g., Vega v. Miller, 273 F.3d 460, 467 (2d Cir. 2001), citing Dube, 900 F.2d at 589, 597-99 (holding that the First Amendment did not permit a public university to deny a professor tenure “in response to pressure exerted by government officials and community activists outraged by” that professor’s controversial curriculum comparing “Nazism, apartheid, and Zionism”).14

14 See also, e.g., Keyishian, 385 U.S. at 592 (requirement that teachers answer under oath whether they were affiliated with “any . . . group . . . which taught or

43

What sets this case apart from those cases is that, here, the professor’s well-established First Amendment interests are not set only against the usual government employer’s interests in the efficient, effective, disruption-free delivery of public services, see, e.g. Reuland v. Hynes, 460 F.3d 409, 418 (2d Cir. 2006), but also against the countervailing “First Amendment principles” that propel a public university’s own “underlying mission,” Blum, 18 F.3d at 1011. On that side of the ledger, courts have consistently celebrated the need to safeguard universities’ self-determination over the substance of the education they provide

advocated the doctrine that the Government . . . should be overthrown or overturned by force, violence or any unlawful means” was unconstitutional); Perry v. Sindermann, 408 U.S. 593, 598 (1972) (“[A] teacher’s public criticism of his superiors on matters of public concern may be constitutionally protected and may, therefore, be an impermissible basis for termination of his employment.”); Meriwether, 992 F.3d at 510 (Pickering balancing favored professor who had been reprimanded for flouting the university’s mandate to use students’ preferred pronouns, “refus[ing] even to permit [the professor] to comply with its pronoun mandate while expressing his personal convictions in a syllabus disclaimer,” because notwithstanding the university’s “interest in stopping discrimination against transgender students,” the notion that “gender identity is an idea embraced and advocated by increasing numbers of people is all the more reason to protect the First Amendment rights of those who wish to voice a different view” (internal quotation marks omitted)); Porter, 72 F.4th at 596 (Richardson, J., dissenting) (“[T]he freedom for professors to discuss or investigate controversial problems without interference or penalty plays a vital role in our democracy.” (internal quotation marks and alteration omitted)).

44

and the scholarship they cultivate. In one oft-quoted concurrence, Justice Frankfurter proclaimed that “[i]t is the business of a university to provide that atmosphere which is most conducive to speculation, experiment and creation,” and to exercise “the four essential freedoms of a university – to determine for itself on academic grounds who may teach, what may be taught, how it shall be taught, and who may be admitted to study.” Sweezy, 354 U.S. at 263 (Frankfurter, J., concurring) (internal quotation marks omitted). We have likewise recognized that the frequently invoked but “difficult to define” ethos of “academic freedom,” Dow Chem. Co. v. Allen, 672 F.2d 1262, 1275 (7th Cir. 1982), encompasses concepts like “the University’s right to make its own rules concerning academic standards,” Burt v. Gates, 502 F.3d 183, 190-91 (2d Cir. 2007), its “prerogative to determine for itself on academic grounds who may teach,” Lieberman v. Gant, 630 F.2d 60, 67 (2d Cir. 1980) (internal quotation marks omitted), its “right to set its own criteria for promotion and then to evaluate a candidate’s fitness for promotion under them,” Bickerstaff v. Vassar Coll., 196 F.3d 435, 455 (2d Cir. 1999), and so on. See also Piggee v. Carl Sandburg Coll., 464 F.3d 667, 671 (7th Cir. 2006) (“[A] university’s ability to set a curriculum is as much an element of academic

45

freedom as any scholar’s right to express a point of view.” (internal quotation marks omitted)). Those principles loom large here.

Still, we do not appear ever to have faced this precise scenario: a claimed First Amendment violation arising from a public university department’s open and deliberate adherence to a particular intellectual methodology or approach. But although we have never expressly decided the question, the Department’s motivations here recall those that we strongly implied were permissible in Dube. There, a professor alleged that he was denied tenure “in response to pressure exerted by government officials and community activists outraged,” Dube, 900 F.2d at 597, by a course description announcing that one of his classes would explore “three forms of racism . . . for comparative purpose[s]; e.g., Nazism, apartheid, and Zionism.” Id. at 589 (alteration in original). We agreed that there was a genuine issue of fact whether the defendants had impermissibly retaliated against the professor based on the content of his class and the public outrage it inspired. Id. at 598. But we also added an important caveat. The defendants, we instructed, “may defend against that claim on the merits by contending that they denied tenure and promotion to Dube for permissible academic reasons, without

46

regard to the surrounding community pressure.” Id. (emphasis added).

The defendants in Dube proffered two such reasons. Both resonate here.

The first was the university’s professed desire to foster collaboration between faculty in Dube’s Africana Studies department and those across other departments. Id. at 591. As the dean (one of the administrators who formally “recommended that Dube be denied tenure and promotion”) explained:

Africana [S]tudies has not yet fulfilled its potential or accomplished its mission at Stony Brook. It needs greatly to elaborate its relations with other departments.

. . . Professor Dube’s apparent withdrawal from academic psychology is not helpful. Africana Studies also needs to become a department and develop a graduate program in conjunction with other departments; Professor Dube’s scholarly posture would be a hindrance here.

Id. The university, in other words, professed an interest in prioritizing tenure candidates whose research would facilitate collaborative synergies with other scholars. That desire for inter-departmental collaboration closely tracks the desire for intra-departmental collaboration Masters expressed in his email to Heim – that the Department was looking to hire someone who “[h]as sufficient synergies with our research agendas that we can learn from them and them from us with

47

the possibility of constructive collaboration.” App’x 598. In her deposition, Daniel further elaborated on the rationale underlying that preference:

[Y]ou do need some synergy there. Like if I were to take someone on just in a totally, totally different field, even though it’s macro, I’m not terribly helpful to them. And they need someone to help them.

Good departments are strong in particular areas. They are not super spread out and thin. So do you want a department that’s super spread out and thin and no one does anything related to each other or do you want people that have a working relationship . . . ?

Id. at 852.

Heim’s critique of that rationale is, of course, also perfectly reasonable. He maintains that a university benefits from assembling faculty members with divergent approaches, methodologies, and points of view within a given discipline. Some universities may well decide to hire with that goal in mind. But decision-makers within a university might also reasonably decide that concentrating on scholars with similar interests or approaches will enhance collaboration, the university’s reputation, or the quality of its collective research output, and hire with those goals in mind instead.

That touches upon the second asserted justification in Dube: concerns that

48

the candidate’s written scholarship did not meet the “usual standards” for the position. Dube, 900 F.2d at 591. In this case, Heim was told by Daniel that the university was prioritizing the techniques favored by “the top macro and general field journals,” where the Department “expect[s] our faculty to publish,” and that because (by Heim’s own admission) his research did not use those techniques, the Department instead intended to hire someone else who had been “trained in those techniques.” App’x 588. Masters later echoed those sentiments, explaining to Heim that the Department was seeking candidates to publish research featuring the “modern (i.e., post Lucas Critique) macroeconomic[]” techniques that Heim conceded his research did not feature, and that the journals that had been willing to publish Heim’s research did not “achieve the standard that we expect for tenure.” Id. at 597. To be sure, unlike in Dube, the Department’s concerns here (at least those expressed openly) were less about the “quantity and quality” of Heim’s work on its own terms, Dube, 900 F.2d at 591 (internal quotation marks omitted), and more about what those terms were – that is, the methodology and approach they reflected. And so, once again, the Department’s concerns circled back to its overriding preference for DSGE.

49

Therein lies the rub. Because the Department’s desire both for collaborative synergies and for publication clout go hand-in-hand with its methodological preferences, they invite decision-makers to evaluate candidates through exactly the sort of “content-based judgment[s]” that are normally “anathema to the First Amendment.” Demers, 746 F.3d at 413. But academia is not normal; it is “unique.” Adams, 640 F.3d at 564. In this exceptional setting, we implied in Dube that such justifications are “permissible academic reasons” for declining to hire or promote a candidate. 900 F.2d at 598. Today we say so explicitly.15 In the “special niche” that academia occupies, such judgments are “both necessary and appropriate.” Demers, 746 F.3d at 411, 413 (internal quotation marks omitted).

And, crucially, they are best left to scholars. See id. at 413 (“[W]e should hesitate before concluding that we know better than the institution itself the nature and strength of its legitimate interests.”). Evaluations of the quality of academic work necessarily turn on judgments about its contents. Certainly, even

15 And, importantly, although the Dube panel remanded for a trial to determine whether those departmental interests were, in fact, the actual motivations for the university’s decision, 900 F.2d at 598, here Heim disputes only the wisdom (and the constitutionality) of those asserted interests, not their veracity. To the contrary, the central premise of his lawsuit is that the Department’s transparently pro-DSGE preferences were the driving force behind its hiring decisions.

50

evaluators with subject matter expertise may miss the mark. Innovative thinkers may be discounted by proponents of conventional, but ultimately mistaken, views. Conversely, departments in thrall to fashionable ideas may deride as “old- fashioned” scholars who, in time, are proven to have been correct all along. Work that builds upon premises that the evaluator rejects may be dismissed out-of- hand as less valuable regardless of that work’s quality in its own right, and regardless of whether, in the long run, it is the evaluator’s predisposition or the rejected premise that prevails.

But to the extent any of that is, as Heim insists, “wrong,” App’x 546, it is a wrong that we cannot right. If the academic enterprise is to remain an engine of progress, decisions about the value of academic work must be left to academics. Though, in nearly all contexts, government officials are barred from discriminating among speakers based on their own judgments of the quality or content of the speech, in this exceptional context where professors’ advancement necessarily depends on the quality of their work, and where the evaluation of that quality necessarily depends on evaluators’ assessments of the work’s content, the First Amendment operates differently. It must operate differently. Much as importing Garcetti’s rationale from typical public employment settings into this

51

atypical setting – effectively stripping professors at public universities of much of their First Amendment protection in the process – would be inconsistent with the core function of those universities, so too would any interpretation of the First Amendment that bridles the efforts of experts at those universities to foster high- quality scholarship with content-neutral constraints imposed by lay courts.

In that spirit, we have cautioned before, and reiterate today, that courts “should not substitute their judgment for that of the college” because things like “teaching ability, research scholarship, and professional stature are subjective, and unless they can be shown to have been used as the mechanism to obscure discrimination, they must be left for evaluation by the professionals.” Bickerstaff, 196 F.3d at 455 n.7, quoting Kunda v. Muhlenberg Coll., 621 F.2d 532, 548 (3d Cir. 1980). Here, there is no indication that the Department’s affinity for DSGE was really a pretextual veil to “obscure discrimination,” or a cudgel to stamp out controversial or dissenting viewpoints, or some other mechanism to advance the views of non-academic public officials. Id. Heim was never deprived of the opportunity to continue his research, or to share his Keynesian perspectives in his lectures. To the contrary, even as they transparently preferred tenure-track candidates with research focuses that differed from Heim’s, both Daniel and

52

Masters actively encouraged Heim to continue his own research. And he did; even after he missed out on the 2017 position, he was able to publish several books alongside his ongoing teaching responsibilities. That research was promptly celebrated and publicized by the university. No speech – controversial, dissenting, or otherwise – was muzzled.

We do not minimize Heim’s interest in retaining the freedom to perform scholarship as he wishes, or in competing for positions that might better facilitate that scholarship; nor do we suggest that the decision not to hire an applicant based on the applicant’s views on academic debates can never prevail over the employer’s interests under Pickering. But we are tasked here with balancing (1) an employee’s asserted interest not simply in speaking freely through his scholarship, but in being considered, without regard to the content of that scholarship, for advancement from a job where he is already empowered to do that research and entrusted to espouse his views as a teacher; against (2) a university’s interest in deciding for itself what skills, expertise, and academic perspectives it wishes to prioritize in its hiring and staffing decisions.

In this case, considering both “the nature of the speech and . . . the nature of the public services performed” by this particular public employer, Blum, 18

53

F.3d at 1011, the latter interest is the weightier. Defendants have decided to prioritize, for purposes of scarce tenure-track positions, a particular methodology. Heim does not dispute that their decision was the product of a learned and strategic choice, made by experts in their good faith professional judgment, free from any influence from political entities in the state or other governmental or university officials outside the relevant discipline. Compare Dube, 900 F.2d at 597. That is, and must be, permissible. If the Supreme Court’s (and this Court’s) enthusiastic endorsement of the First Amendment principles supporting a university’s academic freedom is to be given any practical bite, decision-makers within a university must be permitted to consider the content of an aspiring faculty member’s academic speech, and to make judgments informed by their own scholarly views, when making academic appointments.

CONCLUSION

We have considered Heim’s other arguments and conclude that they are without merit. Thus, for the foregoing reasons, we AFFIRM the judgment of the district court.