Mitchell v. Univ. of N.C. Bd. of Governors

Opinions

Opinion

IN THE SUPREME COURT OF NORTH CAROLINA No. 121A23

Filed 17 October 2025

ALVIN MITCHELL

v.

THE UNIVERSITY OF NORTH CAROLINA BOARD OF GOVERNORS

Appeal pursuant to N.C.G.S. § 7A-30(2) (2023) from the decision of a divided panel of the Court of Appeals, 288 N.C. App. 232 (2023), affirming an order entered on 26 July 2021 by Judge Martin B. McGee in Superior Court, Forsyth County. On 22 March 2024, the Supreme Court allowed in part petitioner’s petition for discretionary review as to additional issues. Heard in the Supreme Court on 18 February 2025.

Fox Rothschild LLP, by Nathan Wilson, Matthew N. Leerberg, and Kip D.

Nelson; and Beechler Tomberlin PLLC, by Allison Tomberlin, for petitionerappellant .

Jeff Jackson, Attorney General, by Lindsay Vance Smith, Special Deputy Attorney General, Ryan Y. Park, Solicitor General, and James W. Doggett, Deputy Solicitor General, for respondent-appellee.

Jacob P. Warner for Alliance Defending Freedom, amicus curiae.

Jonathan D. Guze for John Locke Foundation, amicus curiae.

Phillip Jacob Parker Jr., Stephen A. Woodson, Meghan N. Cook, and Stacy Revels Sereno for North Carolina Farm Bureau Federation, Inc.; and Raymond A. Starling for North Carolina Chamber Legal Institute, amici curiae.

DIETZ, Justice.

Opinion of the Court

This administrative case arose after Winston-Salem State University fired Dr.

Alvin Mitchell, a university professor, for neglecting various job duties such as grading and teaching, and for using offensive racial slurs. Among many other claims, Dr. Mitchell argued that the university failed to follow its own rules and regulations.

When reviewing this procedural claim, the Court of Appeals held that “an agency’s construction of its own regulations is entitled to substantial deference” and that the Court of Appeals must “defer to the agency’s interpretation of its regulations unless it is plainly erroneous.” Mitchell v. Univ. of N.C. Bd. of Governors, 288 N.C. App. 232, 238 (2023) (cleaned up). That holding was based on a decision of this Court applying federal law to a federal agency’s interpretation of its own federal regulations. Id.; see Morrell v. Flaherty, 338 N.C. 230, 237–43 (1994).

With respect to state law, this Court consistently has held that when “the issue on appeal is whether a state agency erred in interpreting a regulatory term, an appellate court may freely substitute its judgment for that of the agency and employ de novo review.” Britt v. N.C. Sheriffs’ Educ. & Training Standards Comm’n, 348 N.C. 573, 576 (1998).

We allowed discretionary review in this case to clarify the appropriate standard of review when construing the meaning of state rules and regulations. As we held in Britt, a state agency’s interpretation of its own rules or regulations can inform a court’s judgment and aid in ascertaining the meaning of the law. Id. But the agency’s interpretation is never binding. Id. We expressly disavow any interpretive

-2-

rule requiring courts to defer to a state agency’s interpretation of state rules and regulations, overrule any previous Court of Appeals case law to the contrary, and instruct all lower courts to apply traditional de novo review to the interpretation of state rules and regulations.

The only remaining issue in this case is a narrow one stemming from the partial dissent at the Court of Appeals. As explained below, our review is constrained by the reasoning provided by the dissent, which asserted that the case must be remanded to the trial court for further proceedings. We reject that argument and affirm the judgment of the Court of Appeals subject to our modifications.

Facts and Procedural History In 2017, Winston-Salem State University fired Dr. Alvin Mitchell, a tenured professor, after he failed to respond to repeated inquiries from a student and faculty members concerning incomplete grading, failed to teach one of his assigned classes, and wrote a letter to a fellow faculty member that used offensive racial slurs.

Dr. Mitchell challenged his firing through the administrative appeal process.

The case worked its way from the Winston-Salem State University Board of Trustees to the University of North Carolina Board of Governors, then to judicial review with the Superior Court, Forsyth County, and then to the Court of Appeals. Mitchell, 288 N.C. App. at 236–37. At every step of the process, Dr. Mitchell lost. Id. He also filed a federal civil rights lawsuit against Winston-Salem State University and various university administrators, which the federal court dismissed. See Mitchell v. Winston-

-3-

Salem State Univ., No. 1:19CV130, 2020 WL 1516537, at *1 (M.D.N.C. Mar. 30, 2020).

After the Court of Appeals rejected his arguments, with one judge concurring in part and dissenting in part, Dr. Mitchell filed two notices of appeal with this Court, one based on the dissent and one based on a substantial constitutional question, and filed a petition for discretionary review as to additional issues.

We allowed a motion to dismiss the constitutional appeal for “lack of substantial constitutional question.” Mitchell v. Univ. of N.C. Bd. of Governors, 898 S.E.2d 295 (N.C. 2024) (order). We also denied discretionary review of most of Dr. Mitchell’s questions presented. We allowed review only on the following issue: “Under North Carolina law, when, if ever, should a court defer to an agency’s interpretation of the rules and regulations that the agency has promulgated?”

Analysis

I. University regulations and the appropriate standard of review We begin with the question presented in Dr. Mitchell’s petition for discretionary review. At the Court of Appeals, Dr. Mitchell argued that the University failed to follow its own rules and regulations governing the dismissal of a tenured professor. Mitchell, 288 N.C. App. at 239. The Court of Appeals rejected this argument after holding that “an agency’s construction of its own regulations is entitled to substantial deference” and that the Court of Appeals must therefore “defer to the agency’s interpretation of its regulations unless it is plainly erroneous.” Id. at 238 (cleaned up). Applying this interpretive doctrine, the Court of Appeals rejected

-4-

Dr. Mitchell’s argument. Id. at 239–41. We allowed discretionary review to examine the appropriate legal standard when reviewing a state agency’s interpretation of state rules and regulations.

As an initial matter, the Court of Appeals holding is at odds with this Court’s precedent. This Court has long held that when “the issue on appeal is whether a state agency erred in interpreting a regulatory term, an appellate court may freely substitute its judgment for that of the agency and employ de novo review.” Britt, 348 N.C. at 576. “De novo” is a Latin phrase meaning “fresh” or “anew.” Taylor v. Bank of Am., N.A., 382 N.C. 677, 679 (2022). Thus, when reviewing a matter de novo, the reviewing court considers the matter anew and is not bound by the interpretation of any lower court or agency. See Brooks v. McWhirter Grading Co., 303 N.C. 573, 580– 81 (1981).

This “de novo” standard of review is compelled by the judiciary’s constitutional role. See Sound Rivers, Inc. v. N.C. Dep’t of Env’t Quality, 385 N.C. 1, 8 n.6 (2023). Although it is “the exclusive right of the legislative department to enact laws and the duty of the executive to enforce them, it is the exclusive right of the judiciary to construe them.” White v. Ayer, 126 N.C. 570, 582 (1900). Thus, North Carolina courts have a constitutional duty to interpret the law and cannot defer in that role to the other branches of government. Id. at 582–84.

Still, this does not mean courts will never afford “weight” or “deference” to an agency’s interpretation of a regulation that it administers. The “interpretation of a

-5-

regulation by an agency created to administer that regulation is traditionally accorded some deference by appellate courts.” Britt, 348 N.C. at 576. This approach reflects our longstanding view that legal interpretations “by persons charged with the administration of the law have always been regarded as legitimate and valuable aids in ascertaining the meaning” of those laws. Gill v. Bd. of Comm’rs, 160 N.C. 176, 188 (1912). In other words, agency interpretations of the law always can inform our judgments, but “those interpretations are not binding.” Brooks, 303 N.C. at 581 (cleaned up). Ultimately, this Court must “employ de novo review” to the interpretation of an agency regulation. See Britt, 348 N.C. at 576.

The confusion at the Court of Appeals appears to stem from our discussion of agency deference in Morrell v. Flaherty, 338 N.C. 230 (1994), a case decided several years before Britt. Morrell involved the meaning of federal regulations. Id. at 237. The federal agency that promulgated those regulations had published an interpretation of them. Id. In describing the “deference” we afforded in Morrell, we first explained “that an agency’s construction of its own regulations is entitled to substantial deference.” Id. (cleaned up). This statement is fully consistent with Gill, Britt, and more recent cases acknowledging that an agency’s interpretation can inform our legal judgment. See, e.g., Gill, 160 N.C. at 188.

But we went on to state that our “task is not to decide which among several competing interpretations best serves the regulatory purpose.” Morrell, 338 N.C. at 238 (cleaned up). Instead, we explained, “the agency’s interpretation must be given

-6-

controlling weight unless it is plainly erroneous or inconsistent with the regulation.” Id. (cleaned up).

We did not distill these latter statements from our own precedent. They are part of a lengthy block quotation from a decision of the Supreme Court of the United States, Thomas Jefferson Univ. v. Shalala, 512 U.S. 504 (1994), which itself cited a long line of federal decisions. Id. at 512. Those decisions employed what is now known as “Auer deference,” named after Auer v. Robbins, 519 U.S. 452 (1997). We used this federal interpretive doctrine in Morrell because we were examining a federal agency’s interpretation of federal regulations. 338 N.C. at 237–43.

In the years since this Court referenced Auer deference in Morrell, the Supreme Court has tightened the reins on that doctrine considerably. See Kisor v. Wilkie, 139 S. Ct. 2400 (2019). The language we quoted in Morrell is no longer accurate because Auer deference is now subject to a number of additional restrictions and qualifications under federal law. Id. at 2415–18.

Moreover, Auer deference has always been a distinctly federal doctrine. It is founded on a presumption that Congress intended for federal agencies to “play the primary role in resolving regulatory ambiguities.” Id. at 2412. Thus, in Kisor, the Supreme Court rejected an argument that the federal Administrative Procedure Act required courts to employ de novo review. Id. at 2419–20. Instead, the Court held that language governing the standard of review in these federal administrative cases, found in 5 U.S.C. § 706, read in light of historical practice and presumptions about

-7-

congressional intent, required courts to “determine the meaning of a rule precisely by deferring to the agency’s reasonable reading.” Id. at 2419 (cleaned up).

Our state’s Administrative Procedure Act does not contain the language that the Supreme Court evaluated to support Auer deference in federal law. To the contrary, it instructs courts to review this sort of legal question “using the de novo standard of review.” Compare N.C.G.S. § 150B-51(c) (2023) (“Scope and standard of review”), with 5 U.S.C. § 706 (“Scope of Review”). Moreover, under the state law precedent discussed above concerning the proper role of the judicial branch, it is unlikely that the General Assembly would even have the constitutional authority to compel state courts to defer to an executive-branch agency’s view of a legal question. See White, 126 N.C. 570. Put another way, North Carolina courts’ de novo review of legal questions is not a choice, it is a constitutional command. See N.C. Const. art. I, § 6.

In sum, the deference discussion in Morrell, which used a now-outdated federal standard for reviewing a federal agency interpretation of federal law, does not apply to state agencies and state regulations. Instead, as this Court emphasized a few years after Morrell, when “interpreting a regulatory term, an appellate court may freely substitute its judgment for that of the agency and employ de novo review.” Britt, 348 N.C. at 576.

Because a number of Court of Appeals decisions have mistakenly read Morrell to create a form of Auer deference under North Carolina law, we take this opportunity

-8-

to expressly hold that courts interpreting state administrative regulations must freely substitute their judgment for that of the agency and employ de novo review. In cases involving a complex or highly technical regulatory program, courts should continue to give due consideration to the views of the agency, as those views may aid “in ascertaining the meaning” of the regulation. Gill, 160 N.C. at 188. But the views of the agency are never binding. Brooks, 303 N.C. at 581. We disavow any state law precedent suggesting otherwise and instruct all lower courts to apply traditional de novo review to the interpretation of state administrative regulations.

II. Court of Appeals dissent on the First Amendment issue We next address a narrow legal issue raised by the Court of Appeals dissent, which is the only remaining issue properly before this Court for review. As with the agency deference issue, this remaining issue concerns a misunderstanding about de novo review.

At the outset, we note that, in an attempt to broaden the scope of our review, Dr. Mitchell filed a notice of appeal based on a substantial constitutional question and a petition for discretionary review, in addition to the notice of appeal based on the dissent. Collectively, these filings would have brought to this Court all of the arguments that Dr. Mitchell asserted to the Court of Appeals. We did not accept review of all these issues. We allowed the University’s motion to dismiss the notice of appeal based on a substantial constitutional question and denied discretionary review of the issues in that notice. Thus, as explained below, our review is limited to

-9-

grounds “specifically set out in the dissenting opinion as the basis for that dissent.” Cryan v. Nat’l Council of YMCAs, 384 N.C. 569, 574 (2023) (cleaned up).

At the Court of Appeals, the majority and the dissent disagreed about whether the offensive racial slurs in Dr. Mitchell’s letter were speech on a “matter of public concern” for purposes of the First Amendment. Compare Mitchell, 288 N.C. App. at 242–43, with id. at 245–52 (Murphy, J., concurring in part and dissenting in part). That distinction matters because, if those racial slurs were speech on a matter of public concern, the applicable First Amendment test then requires the reviewing court to balance Dr. Mitchell’s interest in commenting on those matters with the state’s interest “in promoting the efficiency of the public services it performs through its employees.” Connick v. Myers, 461 U.S. 138, 140 (1983) (cleaned up).

Importantly, this First Amendment balancing test is a question “of law, not fact.” Id. at 148 n.7; see also id. at 150 n.10. Indeed, both the majority and the dissent in this case correctly acknowledged that the “balancing of interests is a question of law.” Mitchell, 288 N.C. App. at 242 (majority opinion) (cleaned up); id. at 244 (Murphy, J., concurring in part and dissenting in part).

As we discussed above, courts review questions of law de novo. See State ex rel.

Stein v. E.I. DuPont de Nemours & Co., 382 N.C. 549, 556 (2022). Thus, if a lower court examining this same First Amendment question fails to engage in the proper legal analysis, there is no need to remand the case for that court to do so; because review is de novo, if the appellate court has jurisdiction over that legal question, it

-10-

can simply assess it using the correct analysis. See Corum v. Univ. of N.C., 330 N.C. 761, 776–77 (1992) (analyzing this same First Amendment “matter of public concern” balancing test on discretionary review even though the Court of Appeals had not done so).

The dissent below did not understand this principle. The dissent first explains why it believes the offensive racial slurs were speech involving a matter of public concern that triggered the need for the balancing test. Mitchell, 288 N.C. App. at 244– 51 (Murphy, J., concurring in part and dissenting in part). Then, the dissent states that it would “remand the case” for the trial court to engage in that balancing test because “that issue has not yet been raised and passed upon in the trial court.” Id. at 252 (cleaned up).

This wording is odd because the issue unquestionably was raised and passed upon in the trial court. The parties argued this issue to the trial court and, in its conclusions of law, the court determined that “the decision to discharge the Petitioner for the reasons stated above was not in violation of any constitutional provisions” and “not in violation of his First Amendment rights.”

The dissent might be referring to the perfunctory nature of the trial court’s conclusions of law, which do not contain lengthy reasoning and analysis. But trial courts often do not articulate their full reasoning on questions of law. This does not impact appellate review. Because legal questions are reviewed de novo, the reviewing court is only concerned with the “disposition of the trial court” on that legal question,

-11-

not “an assessment or review of the trial court’s reasoning.” See Taylor, 382 N.C. at 679. In Taylor, for example, the Court of Appeals reversed a trial court’s order granting a Rule 12(b)(6) motion to dismiss because it “could not determine the reason behind the grant.” Id. at 678 (cleaned up). Thus, the Court of Appeals remanded the case for findings of fact and conclusions of law. Id. We reversed, holding that “there is no legal basis or practical reason for the Court of Appeals to remand the case to the trial court” and that the Court of Appeals “erred by not conducting a de novo review” of the trial court’s ruling. Id. at 680.

Of course, there are times when it is appropriate for an appellate court to remand a matter to a lower court because there are questions of fact that must be resolved. See, e.g., Whitacre P’ship v. Biosignia, Inc., 358 N.C. 1, 39 (2004). This is because appellate courts “cannot find facts.” State v. Jordan, 385 N.C. 753, 757 (2024). But importantly, the dissent does not suggest that there are unresolved fact disputes in the record, and we see none in our own review.

In addition, this Court occasionally remands questions of law to the Court of Appeals after determining that those issues are outside the scope of our intended review. See, e.g., Va. Elec. & Power Co. v. Tillett, 316 N.C. 73, 76 (1986). But this is because our review in those cases is discretionary. See N.C.G.S. § 7A-31. So, for example, to afford “proper deference to the Court of Appeals,” when we reverse on an issue through discretionary review, we often “decline to address the remaining issues raised by the parties but not addressed by that court” and “remand the case to the

-12-

Court of Appeals so that it may address those issues initially on appeal.” Va. Elec. & Power, 316 N.C. at 76.

The Court of Appeals has no similar discretion when it hears an appeal of right.

See generally N.C.G.S. § 7A‑27(b). If a question of law is properly preserved for appellate review, the Court of Appeals must review that legal question de novo. Taylor, 382 N.C. at 679. The Court of Appeals cannot ignore a properly preserved question of law because it wants the trial court to provide further legal analysis; de novo review “does not involve an assessment or review of the trial court’s reasoning.” Id. Thus, the Court of Appeals dissent should not have ignored the First Amendment balancing test that, according to the dissent, was applicable in this case and was a question of law; it should have engaged in that test and, if appropriate, explained why that reasoning would impact the outcome of the appeal.

This error by the dissent constrains our jurisdiction. As we recently reaffirmed, when a litigant pursues a notice of appeal based on a dissent, our review is limited to grounds “specifically set out in the dissenting opinion as the basis for that dissent.” Cryan, 384 N.C. at 574 (cleaned up). So, for example, in Cryan we held that we lacked jurisdiction to review issues that logically flowed from the dissent’s analysis but that were not actually part of the dissent’s stated reasoning. Id. at 574–75.

Here, too, our review is limited to whether the case must be remanded to the trial court to engage in the applicable First Amendment balancing test. As explained above, that reasoning is erroneous and we reject it. Thus, even if we agreed with the

-13-

dissent that the balancing test was triggered, we lack jurisdiction to go beyond the dissent’s reasoning and engage in that constitutional analysis ourselves because the dissent declined to do so. Cryan, 384 N.C at 574–75.

As we noted above, in light of the dissent’s limited reasoning, Dr. Mitchell also filed a notice of appeal based on a substantial constitutional question and a petition for discretionary review as to additional issues, seeking to confer jurisdiction on this Court to conduct the full First Amendment analysis. Had we retained that second notice of appeal or allowed discretionary review on that issue, we could fully address these additional issues. But we did not do so. We dismissed the notice of appeal and denied the corresponding request for discretionary review. As a result, on the limited issue over which we possess appellate jurisdiction, we reject the reasoning of the dissent and affirm the judgment of the Court of Appeals as modified by our discussion above.

We close by noting that the dissent does not agree with this jurisdictional analysis, arguing that it is a “strained use of Cryan.” This is unsurprising. Our dissenting colleague has consistently disagreed with us—and dissented—in every case in which this Court has applied this rule from Cryan. See, e.g., Morris v. Rodeberg, 385 N.C. 405, 417–18 (2023) (Earls, J., concurring in part and dissenting in part); Bottoms Towing & Recovery, LLC v. Circle of Seven, LLC, 386 N.C. 359, 372 (2024) (Earls, J., dissenting). We could hardly expect the dissent to change views today. Still, though, precedent is precedent. We endeavor to address the merits of

-14-

every issue the parties present to us. But, in these appeals based on a dissent, there will be times when principles of appellate jurisdiction constrain our review. Cryan, 384 N.C at 574–75; Morris, 385 N.C. at 415; Bottoms Towing, 386 N.C. at 362. This is one of those cases.

Conclusion

We modify and affirm the judgment of the Court of Appeals.

MODIFIED AND AFFIRMED.

-15-

Justice Earls, concurring in part

Justice EARLS concurring in the result only in part and dissenting in part.

Winston-Salem State University and the University of North Carolina Board of Governors terminated Alvin Mitchell from his tenured professorship on three grounds: failing to open an online course, failing to issue a student a final grade, and sending a colleague a racially charged letter about racism and racial bias in academia. The issues presented are first, whether the public university complied with its Handbook procedures when it terminated Professor Mitchell and second, whether Professor Mitchell’s letter is protected First Amendment speech and thus cannot be the basis for terminating him from his position.

As to the first issue, I concur in the result reached by the majority only. The majority takes this opportunity to broadly proclaim that no Auer/Kisor deference exists in state administrative law and that only “due consideration to the views of the agency” applies for “complex or highly technical regulatory program[s].” See majority supra Analysis, Part I. But such sweeping language goes far beyond our precedent and the narrow issue presented by this appeal. Under our precedent, courts only potentially defer to an agency’s interpretation of its own regulation if the regulation itself is ambiguous. Britt v. N.C. Sheriffs’ Educ. & Training Standards Comm’n, 348 N.C. 573, 576 (1998). That is not the case here. Indeed, no party below argued that the agency’s interpretation was entitled to deference. The Handbook’s termination procedures are clear, the University followed them, and Professor Mitchell’s contrary

-16-

interpretation is not plausible—all of which the Court of Appeals appropriately recognized when it held for the Board of Governors on this issue. In my view, this Court should affirm the Court of Appeals’ judgment––and stick to resolving legal issues actually presented by the facts of a given case.

Second, I dissent from the majority’s decision to sidestep an issue that is presented for our resolution: whether Professor Mitchell’s letter to his colleague and academic department chair, Professor Denise Nation, was protected First Amendment speech that cannot be the basis for his termination. At the Court of Appeals, the majority and the dissent disputed whether the letter addressed a matter of public concern, a threshold requirement for further First Amendment protection in this setting. The majority held it did not, while the dissent would have held it did. Compare Mitchell v. Univ. of N.C. Bd. of Governors, 288 N.C. App. 232, 243 (2023), with id. at 250 (Murphy, J., concurring in part and dissenting in part). The partial dissent’s detailed explanation of why Professor Mitchell’s letter addressed issues of public concern, particularly in an academic setting, was more than enough to present this issue for our mandatory review. I disagree with the majority’s strained use of jurisdiction to skirt resolution of this important legal issue.

Because I would hold that this issue is presented for our review, I would also reach the merits of Professor Mitchell’s First Amendment challenge. The freedom of speech is a paramount American value and cornerstone constitutional right that protects individuals from being punished by the government for expressing their

-17-

views. The First Amendment’s protections extend to professors at public universities engaged in scholarship, teaching, and other academic exercises that require freedom of thought and exchange. I agree with Professor Mitchell that his letter to Professor Nation was protected First Amendment speech and that the public university here violated his rights when it terminated him partly on that basis.

I. Whether the University Complied with Its Procedures In its rush to abolish any judicial deference to administrative agencies, the majority neglects to discuss essential facts about this case. This case comes to us because Professor Mitchell challenges that he was unlawfully terminated when the University failed to comply with its Handbook procedures when it dismissed him. Namely, he argues that after a faculty committee determined that the provost had not made out a prima facie case for his dismissal, the chancellor had no grounds to terminate him anyway. The Court of Appeals rejected this argument: “The text of the UNC Code aligns with the interpretation [of the Handbook] followed by [the University],” that the Chancellor retains final authority over faculty termination after receiving a recommendation and a factual record from the faculty hearing committee. Mitchell, 288 N.C. App. at 240. I agree with the Court of Appeals that the University’s conduct comported with the plain language of the regulations at issue. Professor Mitchell’s challenge on this basis should be rejected and the Court of Appeals’ judgment on this issue affirmed.

The Winston-Salem State University Handbook establishes a procedure for a

-18-

provost to initiate proceedings to terminate a tenured professor. It provides that after the provost notifies a professor of her intent to terminate him, including for neglect of duty, misconduct, or professional ethical violations, the professor has the right to appeal that decision to a faculty committee. At the hearing before that committee, the provost first has the burden to prove a prima facie case for termination on the grounds in the provost’s notice to the professor. After the provost presents her case in chief, the committee withdraws into closed session to decide by majority vote whether she met her burden.

If the committee determines that the provost met her burden, then “it will resume the hearing” and allow the faculty member to present her case. If the committee determines that the provost did not meet her burden, then the committee suspends the hearing and makes a recommendation against discharge to the chancellor. This provision protects the faculty member from having to present his or her own evidence if the faculty committee and the chancellor agree that the provost failed to make a prima facie case. “If,” however, “the [c]hancellor disagrees with the committee’s determination, [the chancellor] will send it back for a full hearing.” The full hearing includes the faculty member’s opportunity to present evidence and call witnesses and the administration’s opportunity to rebut that evidence.

After the conclusion of the full hearing, the committee decides by majority vote whether, in light of all the evidence presented, the provost has proven her case. It then notifies both the administration and the faculty member of its findings and

-19-

recommendation and transfers the evidence it relied on to the general counsel. “[W]ithin 30 Days after receiving the hearing documents including the transcript of the hearing,” “[t]he chancellor shall issue a final written opinion . . . based on the recommendations and evidence received from the hearing committee.” (Emphases added.) “[F]rom the chancellor,” the faculty member may appeal “to the Board of Trustees,” and that board shall make a final decision subject to review by the Board of Governors.

The University followed these procedures in Professor Mitchell’s case. On 31 August 2017, the University provost notified Professor Mitchell of her “intent to dismiss” him from his tenured position. Professor Mitchell then exercised his right to challenge that decision at a hearing before a faculty committee. At that hearing, the provost had the burden to establish a prima facie case for dismissal. She presented her case-in-chief at a day-long hearing with various witnesses and statements from advocates. The faculty committee then withdrew to deliberate and determine, by majority vote, whether the provost met her burden.

After deliberations, the committee issued a unanimous letter to the chancellor explaining its view that the provost had not met a prima facie case to support discharging Professor Mitchell. The committee transmitted its recommendation against dismissal to Chancellor Elwood Robinson.

Chancellor Robinson responded on 30 January 2018 that he “disagree[d] with the [c]ommittee’s determination that a prima facie case was not proven,” and he noted

-20-

that the committee should conclude the hearing consistent with the Handbook. On “remand” of sorts from the chancellor, Professor Mitchell informed the committee that he had “no further evidence to present.” Thus the faculty committee accepted his waiver of his rights and again sent the case file and its recommendation against dismissal to Chancellor Robinson.

On 7 March 2018, Chancellor Robinson “decided to uphold the Provost’s decision to discharge [Professor Mitchell].” He concluded that failing to open an online class for eight days jeopardized the financial aid and graduation timeline of students in the class, that failing to issue the student a grade was inappropriate behavior for a faculty member, and that sending “a racially charged communication to [his] Department Chair” was “inappropriate and unacceptable” and not protected by notions of academic freedom and free speech. Professor Mitchell appealed that decision again to the Board of Trustees, and again to the University of North Carolina Board of Governors, before seeking judicial review.

Professor Mitchell argues that this procedure did not comport with the Handbook because the Handbook rests ultimate decision-making authority with the faculty committee, not the chancellor. In his view, the Handbook provides that if the faculty committee recommends against discharge and the chancellor disagrees and sends the matter back for a full hearing, then the faculty committee can just decide it stands by its initial decision and decline to resume the hearing. But this reading of the Handbook is not plausible. For one, the Handbook repeatedly refers to the faculty

-21-

committee’s mere “recommendation” in contrast to the chancellor’s “final written opinion” and the chancellor’s right to “decline[ ] to accept” a favorable “recommendation” from the faculty committee. It would contradict the plain language of the policy if instead, the chancellor’s opinion were merely a “recommendation” and the faculty committee retained final say. See Britt, 348 N.C. at 576 (“[U]nambiguous . . . regulations [must be given] their plain meaning.”). For another, the Handbook expressly instructs, “If the Chancellor disagrees with the committee’s determination, he/she will send it back for a full hearing.” Professor Mitchell’s reading would render this sentence superfluous, contradicting basic tenants of textual interpretation. See State v. Morgan, 372 N.C. 609, 614 (2019) (applying this canon to statutory construction).

Even more, as the Court of Appeals noted, the University of North Carolina Code, which applies to all seventeen public institutions including Winston-Salem State University, reiterates that faculty committees make “recommendation[s]” which chancellors can “decline[ ] to accept.” Thus the Code confirms that the chancellor, not the faculty committee, has final say over termination of tenured professors.

In summary, because the regulation at issue is not “susceptible of multiple interpretations,” see In re J.E.B., 376 N.C. 629, 634 (2021) (cleaned up), it is not ambiguous and must be given its plain meaning, see Britt, 348 N.C. at 576. How the agency interprets its regulations is not analytically relevant under these

-22-

circumstances. See id. The University’s and the Board of Governors’ conduct comported with the plain language of its policies, and thus Professor Mitchell’s rights were not violated on this basis. I would affirm the judgment of the Court of Appeals.

I must stress my view that the majority’s decision to use this case to opine generally on the scope and extent of agency deference is improper. For one, the record shows that neither party argued agency deference below. Nor is it clear to me that the Court of Appeals’ reasoning really relied on deference––the court apparently held for the Board of Governors because the “text” of the relevant regulations “align[ed] with the interpretation” followed by the University. Mitchell, 288 N.C. App. at 240. Even though we allowed discretionary review of an agency deference issue, the proper disposition where we allow review of an issue only first raised in a petition for discretionary review is to dismiss review as improvidently allowed. See State v. Sturkie, 325 N.C. 225, 226–27 (1989).

Further, because agency deference is not actually at issue in this case, the majority essentially opines on an “abstract proposition[ ] of law.” In re Peoples, 296 N.C. 109, 147 (1978). It is not useful, appropriate, or wise for courts to do this. See id. Decisions of this Court must always be understood and applied “in the light of the facts of the case under discussion.” Home Indem. Co. v. W. Trade Motors, Inc., 258 N.C. 647, 650 (1963) (quoting Armour & Co. v. Wantock, 323 U.S. 126, 133 (1944)). Future cases presenting materially dissimilar facts may invoke different legal principles and considerations that support a different outcome. Boswell v. Boswell,

-23-

241 N.C. 515, 518 (1955) (citing Nantahala Power & Light Co. v. Moss, 220 N.C. 200, 208 (1941)). Thus our Court’s job is to “declare[ ] the law as it relates to the facts of the particular case under consideration.” Id. at 518. Only then do we issue decisions that “may be considered authority.” Id. These fundamental principles are also part of the “proper role of the judicial branch.” See majority supra Analysis, Part I.

Respect for our appropriate judicial role applies with special purchase to the general legal issue the majority discusses here. Many state agencies are tasked by the General Assembly and the people of North Carolina with making rules and regulations that have the binding force of law. As the United States Supreme Court has explained, it makes sense that a legislative body that empowers an agency to issue rules also meant for that agency to primarily resolve the inevitable ambiguities arising from such rules. Kisor v. Wilkie, 588 U.S. 558, 569 (2019). This makes common sense––if you want to know what a text means, asking its author is a good place to start. See id. It also coheres with “the comparative advantages of agencies over courts in making [the] policy judgments” that must be made in the face of regulatory ambiguities. Id. at 571. Agencies are generally subject to greater political accountability for their actions than are courts who review those actions––and it is appropriate to vest policy decision-making authority in bodies who are most accountable to the public. See id.

Where agencies carry out responsibilities vested in them by the legislature, the executive, and ultimately the public, courts act consistent with judicial review

-24-

principles when we defer to an agency’s reasonable interpretation of a “genuinely ambiguous” regulation that “reflect[s] its authoritative, expertise-based, and fair and considered judgment,” taking account of “reliance interests and avoid[ing] unfair surprise.” Id. at 591 (Roberts, C.J., concurring in part). Put another way, we may still employ “de novo review” while affording “some deference” to an agency’s “interpretation of a regulation” that it is “created to administer.” Britt, 348 N.C. at 576. We should address the contours of such deference in a case where such contours are actually at issue.

II. Jurisdiction Over the First Amendment Issue As to the second issue, I dissent from the majority’s conclusion that this Court lacks jurisdiction over the First Amendment issue. I would hold that we do have jurisdiction over this dispute.

There used to be an appeal as of right to this Court when an argument advanced at the Court of Appeals did not prevail but did persuade at least one judge on the panel. N.C.G.S. § 7A-30(2) (2023) (repealed 2023).1 Winning one vote earned the losing party the right to appeal and present the issue for final resolution by this Court. After all, a dissent signaled that reasonable judges disagreed about the proper

1 The General Assembly repealed subsection 7A-30(2) through the Current Operations

Appropriations Act of 2023, S.L. 2023-134, § 16.21(d)–(e), 2023 N.C. Sess. Laws 760, 1171. We still have mandatory review over this case because the “appeal was filed and docketed at the Court of Appeals before the effective date of that act.” Bottoms Towing & Recovery, LLC v. Circle of Seven, LLC, 386 N.C. 359, 361 n.1 (2024).

-25-

resolution of a matter and strongly enough to explain publicly why they thought their colleagues erred.

Accordingly, our mandatory review under this mechanism was “limited to a consideration of those issues that are . . . specifically set out in the dissenting opinion as the basis for that dissent.” N.C. R. App. P. 16(b), 369 N.C. 763, 813 (2016) (repealed 2025). In Cryan v. National Council of YMCAs, 384 N.C. 569 (2023), we clarified that a dissenting judge did have to “set out in detail the reasons why” he disagreed with his colleagues to confer this right of appeal. Id. at 574. But when the dissent did so, particularly “[i]n several pages of thorough analysis,” that reasoned dispute conferred mandatory jurisdiction to this Court to hear the appeal. Id.

Below, the Court of Appeals majority concluded that Professor Mitchell’s speech was not on a matter of public concern and thus was not even eligible for the First Amendment protections of public employees. Mitchell, 288 N.C. App. at 242–43. The court observed that “[t]here is no evidence in this Record . . . that Dr. Nation’s decision to deny funding to [Professor Mitchell]’s students . . . was racially motivated or a product of racial bias in academia,” nor was there “evidence that [Professor Mitchell] intended his letter to be an effort to combat racism in academia or to advocate on the part of his students for funding to attend his preferred conference on that basis.” Id. at 242. Because it concluded that the letter addressed only matters of private concern, its application of First Amendment doctrine ended there, and it held

-26-

that the Board of Governors did not err by discharging him in part based on his letter to Professor Nation. Id. at 243.

The dissent would have held that Professor Mitchell’s letter did address matters of public concern and even implicated special academic freedom concerns. Id. at 246 (Murphy, J., concurring in part and dissenting in part) (reviewing examples of public concern with how race issues are addressed in academic settings). To him the letter “reads, simultaneously and inseparably, as a defense of the academic legitimacy of a conference, an expression of dissatisfaction on the state of racial diversity in academia, and a statement of frustration with Dr. Nation, both personally and with any potential unconscious biases.” Id. at 249. The letter’s racial epithets, in context, were not “racial abuse” or “simple derogation,” but they were Professor Mitchell’s expression of how he thought other academics perceived Professor Nation, motivated by a student’s concern that Professor Nation’s decision to decline to fund the students’ attendance at a conference was in part an expression of Professor Nation’s own bias. Id. at 250–52.

Simply put, the Court of Appeals disagreed about the First Amendment; it did not spend a dozen appellate reporter pages arguing over “whether the case must be remanded to the trial court to engage in the applicable First Amendment balancing test.” See majority supra Analysis, Part II.

Contrary to the majority, then, the First Amendment issue is properly presented for our resolution, and Professor Mitchell has the right to appeal to this

-27-

Court. See N.C.G.S. § 7A-30(2) (2023). He did so. Professor Mitchell’s notice of appeal—an appeal which this Court retained—indicated that he would present the following issue for appellate review based on this dissent: “Did the UNC Board of Governors violate Professor Mitchell’s free speech rights under the federal or state constitutions?” We have mandatory jurisdiction to resolve precisely that.

Underscoring this strained use of Cryan, the majority’s jurisdictional analysis fails on its own terms. If this Court concludes that the dissent below misapplied the law—because it should have proceeded to analyze the second prong of Pickering de novo, rather than remanding for the trial court to pass first on that legal question, see Pickering v. Bd. of Educ., 391 U.S. 563 (1968)—it does not follow that that legal error is jurisdictional. It seems to me that, faced with such circumstances, this Court should just correctly apply the law. After all, we swear an oath to “support and maintain the Constitution and laws of the United States” and enforce the federal constitutional rights of the parties before us. N.C. const. art. VI, § 7. The dissent’s purported misapplication of First Amendment doctrine does not strip this Court of our obligation to resolve the dispute consistent with First Amendment rights.2

2 As to the majority’s final paragraph in Part II, which does not respond to these

critiques but rather asserts that I have dissented “in every case in which this Court has applied this rule from Cryan,” I kindly direct the Court’s attention to Jones v. J. Kim Hatcher Insurance Agencies, Inc., 387 N.C. 489, 497 (2025) (Earls, J.) (“For appeals based on a dissent in the Court of Appeals, our review is limited to the scope of the issues specifically set out in the dissent and argued for by the parties.” (first citing Cryan v. Nat’l Council of YMCA’s of U.S., 384 N.C. 569, 574 (2023); and then citing N.C. R. App. P. 16(b) (amended 2025))), and McKinney v. Goins, 387 N.C. 35, 80 (2025) (Earls, J., concurring in result only) (pointing out that “the majority sidestep[ped]” the issue of how “to mesh the vested-rights doctrine with

-28-

In summary, I do not agree with the majority’s application of Cryan to this case, and I would hold that we do have jurisdiction over the First Amendment issue based on the dissent at the Court of Appeals. I would resolve the merits of that dispute.

III. The First Amendment and Professor Mitchell’s Termination Academic freedom is “a special concern of the First Amendment.” Keyishian v.

Bd. of Regents, 385 U.S. 589, 603 (1967). It is a bulwark of our democracy on which other constitutional rights and freedoms depend. See id. Schools and universities promote knowledge and truth through the “robust exchange of ideas” over “authoritative selection” of the same. Id. (quoting United States v. Associated Press, 52 F. Supp. 362, 372 (1943)). Teachers in particular, from “the primary grades to the university,” are the “priests of our democracy” because they are entrusted with instilling in students such essential American values as “open-mindedness and critical inquiry.” Wieman v. Updegraff, 344 U.S. 183, 196 (1952) (Frankfurter, J., concurring). “Teachers and students [in such settings] must always remain free to inquire, to study and to evaluate, to gain new maturity and understanding . . . .” Sweezy v. New Hampshire, 354 U.S. 234, 250 (1957). Thus professors may not be convicted for contempt for refusing to answer questions about their political views,

the fundamental-rights doctrine” notwithstanding that the issue was specifically disputed by the majority and dissent below and thus preserved for appellate review under Cryan (cleaned up)).

-29-

id. at 254–55, nor may public school teachers be barred from employment based on vague, ideological litmus tests, Keyishian, 385 U.S. at 599–600.

Even so, public university professors are still public employees. And public universities must be able to maintain “regular operation[s]” and perform public functions. Corum v. Univ. of N.C., 330 N.C. 761, 775–76 (1992). Among other things, “working relationships” among academic colleagues are a necessary part of regular university operations. See id. at 776.

Thus when a public university professor engages in speech on a matter of public concern and related to scholarship and teaching, that speech is protected if the professor’s interest in speaking outweighs the University’s interest in promoting efficient public services. See Pickering, 391 U.S. at 568; Garcetti v. Ceballos, 547 U.S. 410, 425 (2006) (excluding speech “related to scholarship or teaching” from its enunciation of a public-employee official-duties test); accord Adams v. Trs. of the Univ. of N.C.-Wilmington, 640 F.3d 550, 563–64 (4th Cir. 2011) (applying Pickering to claims that a public university professor was retaliated against for his off-campus comments as well as his books and other publications).

Balancing such interests requires careful attention to “the manner, time, and place of the employee’s expression” as well as “the context in which the speech occurred.” Corum, 330 N.C. at 776. Public and private speech on matters of public concern may be protected. Givhan v. W. Line Consol. Sch. Dist., 439 U.S. 410, 413 (1979). Issues of public concern capture those that are broadly of social, political, or

-30-

community interest. Ridpath v. Bd. of Governors Marshall Univ., 447 F.3d 292, 316 (4th Cir. 2006). In the university setting in particular, such issues comprise all but the “narrow spectrum” of speech that is of purely “personal concern” like “private personnel grievance,” Piver v. Pender Cnty. Bd. of Educ., 835 F.2d 1076, 1079 (4th Cir. 1987) (quoting Berger v. Battaglia, 779 F.2d 992, 998–99 (4th Cir. 1985)), such as where an “aggrieved employee” speaks out about “his own employment situation when that employment situation holds little or no interest for the public at large,” id. (citing Berger, 779 F.2d at 998–99). “[I]nternal grievance[s]” or personal attacks that do not “seek to communicate to the public or advance a political or social point of view beyond the employment context” are not so protected. Borough of Duryea, Pa. v. Guarnieri, 564 U.S. 379, 398 (2011). Where a public university professor’s “speech more substantially involved matters of public concern,” or implicated academic freedom concerns, the university must show that the speech more than minorly impaired interfaculty relationships. See Connick v. Myers, 461 U.S. 138, 151–52 (1983).

To apply these principles to Professor Mitchell’s speech, some context is in order. Professor Mitchell was a tenured professor of justice studies in the Department of History, Politics, and Social Justice at Winston-Salem State University, a historically Black institution. Professor Nation was also a professor in that department and the justice studies program coordinator. She became department co- chair in July of 2015. Department co-chairs have additional powers and

-31-

responsibilities, including some role in supervising other professors and some say in authorizing academic conference travel.

At some point in the 2016–2017 academic year, two students in Professor Mitchell’s undergraduate research methods class conducted original research and wrote a paper on juvenile offenders and the effectiveness of juvenile rehabilitation centers. The students hoped to present their findings at a conference in New Orleans, the Race, Gender and Class conference. The students approached Professor Nation about obtaining necessary funding to attend that conference. According to one student’s account, Professor Nation declined to fund them to attend that conference, and instead she directed the students to a different conference up north, the American Society of Criminology conference. The student believed, based on this conversation, that Professor Nation thought the predominately white American Society of Criminology conference was more prestigious than the more racially diverse Race, Gender and Class conference. The students reported that this meeting with Professor Nation “was the last [the students] spoke about [their] research paper,” because Professor Nation would not approve them to attend the Race, Gender and Class conference.

Upon hearing the students’ account of this exchange, Professor Mitchell took two actions: he e-mailed both department co-chairs about his concerns and he wrote a private letter to Professor Nation. His e-mail to both co-chairs included various

-32-

concerns about how the chairs handled other administrative issues at a recent department meeting. The relevant part of the e-mail read as follows:

Hi Cynthia [Villagomez, the other department co-chair], it was brought to my attention that Dr. Nation told a student that the conference I and two of my students are presenting at has no substance or standards, meaning that it is useless and unaccredited, and anyone can present. In addition, she told the student that she should try to present at the ASC [American Society of Criminology conference] held in November because it is a better conference and has a lot of substance. Dr. Nation is entitled to her opinion. However, Dr. Nation should not be telling the students things like that, especially with no proof. The Race, Gender, & Class conference is locally, regionally, and internationally known and have [sic] scholars from around the world presenting.

In addition, this conference has been in existence for over 20 years. Thirdly, this conference does not take anyone. A presenter has to be accepted through their process. It would have been better if Dr. nation [sic] told the student that the [sic] she or the school was not going to fund her instead of telling the student falsehoods about the RGC conference and asking the student to present in our department on scholarship day. This is not appropriate behavior as chair. You know the importance of undergraduates presenting at conferences in reference to them being selected into graduate school. Students should be encouraging [sic] to continue their quest to present at conferences; not belittle them [sic] based on her opinions and feelings toward me or an organization. Alvin

Professor Mitchell then wrote a private letter to Professor Nation. He did not circulate it publicly, print it on school letter-head, or send it via his school e-mail. He copied over the first paragraph, rephrased it to address Professor Nation directly, and added further remarks. He then slid it under Professor Nation’s door. It read as follows:

Hi Denise, it was brought to my attention that you told a student that the conference I and two of my students are

-33-

presenting at has no substance or standards, meaning that it is useless and unaccredited, and anyone can present. In addition, you told the student she should try to present at the ASC held in November because it is a better conference and has a lot of substance. You are entitled to your opinion.

However, you should not be telling the student things like that, especially with no proof. The Race, Gender, & Class conference is locally, regionally, and internationally known and have [sic] scholars from around the world presenting.

In addition, the conference has been in existence for over 20 years. Thirdly, this conference does not take anyone.

You have to be accepted through their process. It is amazing how you always try to debunk what I do. Yet you complain that I tell students negative things about you. It would have been better to tell the student that you did not want to help fund her instead of telling her falsehoods about the RGC conference and asking her to present on scholarship day. That is not appropriate behavior as a chair.

After all these years, it is amazing that you still think that anything white is better. I looked up the ASC and nothing but a bunch of white men (some white women) are running it. Keep promoting and praising those white folks who are associated with the ASC. As I told you before, you can graduate from and praise their schools, come up with a great theory, hangout with them, praise Latessa and other European professors (you need to ask them about their civil rights record), wear their European style weaves, walk with their bounce, hire them, present at their conferences, and even publish in their journals. In their eyes you will never be equal to them. They still look at you as a wanna be white, an international nigger, an international coon, and an international sambo (lol) because you display that kind of behavior. You will never get it. Wake up.

Professor Nation found the letter “disrespectful” and “demeaning.” She forwarded it to the University provost and the dean but did not file a formal complaint. She thought this correspondence was one “escalating” instance among the other of her

-34-

“disturbing . . . interactions with Dr. Mitchell over the years.” Indeed, at this point, others at the University had assumed at least some supervisory role over Professor Mitchell instead of her.

Nearly six months later, the provost cited Professor Mitchell’s letter to Professor Nation as one of the grounds for his termination. As the matter was appealed through the University’s process, administrators maintained that Professor Mitchell’s letter was not protected speech because Professor Mitchell’s comments were “about Dr. Nation’s job advising students about conferences” and not made “in the context of scholarship, teaching, or academic writing.”

Academic freedom and speech on issues of public concern in the public university setting are not so narrowly protected under the First Amendment. Understood in context and with attention to the “content[ and] form” of the statement, Professor Mitchell’s speech addressed matters of public concern and related to teaching and scholarship. See Connick, 461 U.S. at 147–48; Garcetti, 547 U.S. at 425. The first paragraph expresses Professor Mitchell’s strong disagreement with the guidance Professor Nation gave one of his students about the merits of one academic conference over the other. His concerns related to teaching because it was his student in his research class who needed funding to present her academic research, and because Professor Nation’s response carried implications for that student’s future— undergraduate opportunities for conference presentations shape future opportunities for postgraduate studies, as Professor Mitchell’s e-mail to both co-chairs noted.

-35-

The second part of the letter builds on those concerns and asserts Professor Mitchell’s perception of Professor Nation’s own racial biases (“anything white is better”). Although some of Professor Mitchell’s references to “you” appear to address Professor Nation directly (e.g., “[a]s I told you before”), others appear to take a generic “you” form (e.g., “you can graduate from . . . their schools”). The generic “you” statements are where Professor Mitchell expresses his opinion of how others may view Professor Nation and other academics who make the choices he is criticizing (e.g., “present at their conferences[ ] and even publish in their journals”).

Professor Mitchell’s use of racial epithets did not directly assail Professor Nation, assert that he holds these racist views, or demean her based on her race. This is not an internal grievance devoid of social and political significance. See Guarnieri, 564 U.S. at 398. Rather, his letter expressed his views of how Black faculty may be perceived by a predominately white academy like the American Society of Criminology (“In their eyes you will never be equal to them. They still look at you as a wanna be white . . . .”). That expression was directly related to his earlier concern about scarce opportunities to present academic scholarship. In broader terms, Professor Mitchell’s second paragraph offers his opinion of how Black academics should conduct themselves in predominately white academic spaces.

This speech on racial bias and anti-Black racism in academia addresses matters of public concern. See Ridpath, 447 F.3d at 316. “[R]acial discrimination” and racial bias are “matter[s] inherently of public concern,” and an individual’s right to

-36-

protest and debate those issues “is not forfeited by her choice of a private forum.” Connick, 461 U.S. at 148 n.8. This speech also implicates academic freedom: It seems especially important that faculty members in a “History, Politics, and Social Justice” department can freely express concerns to each other about the ongoing role of racial bias at their university and in their academic fields. Professor Mitchell’s choice to speak his views privately to his colleague does not strip the letter of its public concern character. See Givhan, 439 U.S. at 413.

The second Pickering prong asks whether Professor Mitchell’s interest in addressing matters of public concern outweighs the University’s interests. Pickering, 391 U.S. at 568. In these circumstances, the Board of Governors must meet an unusually high showing to prevail: “The government employer must make a stronger showing of the potential for inefficiency or disruption when the employee’s speech involves a ‘more substantial[ ]’ matter of public concern.” Love-Lane v. Martin, 355 F.3d 766, 778 (4th Cir. 2004) (alteration in original) (quoting Connick, 461 U.S. at 152).

The University did present evidence that Professor Mitchell’s comments offended Professor Nation. She took his letter to personally demean her, found it “disrespectful,” and believed that it contributed to a “hostile workplace.” But there is little evidence of disruption to university operations based on this letter specifically. There is no testimony that the letter in particular caused Professor Nation or anyone

-37-

else to feel afraid or impaired their ability to do their jobs.3 Indeed, the record suggests that Professor Mitchell and Professor Nation’s relationship was already substantially deteriorated from events preceding the letter and that he was being supervised in part by other administrators at the time. Rather than demonstrating how this letter impaired efficient university operations, the record reveals that the interpersonal hostility between Professor Mitchell and Professor Nation had sprung from other sources, like Professor Mitchell’s refusal to “change how he taught, how he set up his syllabus, how he taught the online class,” his unresponsiveness to e- mails, and his failure to attend department meetings, among other things.

On the other side of the scale is Professor Mitchell’s strong interest in speaking out about the merits of one academic conference over another, academic opportunities the University would make available to his students, racism and racial bias in academia, and how Black academics can best navigate predominately white spaces. His speech is within the “academic freedom” of paramount value not only to the University community, but to all of us in North Carolina and the nation. See Keyishian, 385 U.S. at 603. Professor Mitchell could have chosen less inflammatory and racially charged language, but he did not. Under these circumstances, though,

3 Professor Nation never testified that the letter made her feel afraid of Professor

Mitchell. She testified that she thought other people were afraid of him, that she was afraid of escalating tensions with him, and that he was “hostile” to her over the months of “January to August.” She seemed to wish that the dean would “reprimand him.” But the record does not seem to support that the letter caused Professor Nation to become afraid of Professor Mitchell or that the letter specifically substantially impaired their relationship.

-38-

the free exchange of ideas in academic settings “may not be shut off in the name alone of conventions of decency.” Papish v. Bd. of Curators, 410 U.S. 667, 670 (1973) (cleaned up).

Thus Professor Mitchell’s letter to Professor Nation was protected First Amendment speech and cannot be the basis for his termination from public employment. See McVey v. Stacy, 157 F.3d 271, 277 (4th Cir. 1998). I would reverse the Court of Appeals’ judgment on that basis and remand this matter to the University to consider whether the remaining grounds offered for his termination are sufficient to support that decision. I respectfully dissent from the majority’s failure to reach the merits of this issue.

Justice RIGGS joins in this concurring in the result only in part, dissenting in part opinion.

-39-