From a panel decision Bennie v. Munn (8th Cir. May 11, 2016); shortly after the decision came down, I filed an amicus brief supporting the petition urging the Eighth Circuit to rehear the case before the full court, but I’m sorry to say that the full court just last week denied the petition:

1. Bob Bennie, like millions of Americans, was working for a private business but under government regulation; and what got him into trouble is that the regulators didn’t like Bennie’s political speech:

Until November 2010, Bennie worked for LPL Financial (LPL). LPL is a broker-dealer, meaning it holds accounts and assets and executes financial transactions. It operates through agents like Bennie, who deal with customers. As a broker-dealer, LPL is subject to regulation by the Nebraska Department of Banking and Finance (department). Among other things, the department regularly reviews LPL’s agents’ advertisements and other public statements for compliance with applicable rules. The department can sanction broker-dealers and their agents for violations, including by fining them or barring them from operating in Nebraska.

In late 2009, … [t]he compliance supervisor at the [D]epartment, Rodney Griess, … reviewed a television commercial in which Bennie rode a horse and said he would give customers who did business with him “a hundred dollars towards the purchase of a firearm.” Because Griess thought the offer “unusual,” he suspected Bennie had not gotten the necessary approval from LPL to run the commercial. Eventually, Griess scheduled a conference call to talk to LPL about the issues with Bennie.

A few days before the call, the Lincoln Journal Star ran a story about Bennie’s role in the Tea Party political movement. The article quoted Bennie denouncing the government and politicians, including President Barack Obama. The article also mentioned Bennie’s business and was accompanied by a photograph of Bennie at his desk in his office.

Griess emailed LPL a link to the online version of the article. In the body of his message, Griess quoted Bennie, in the article, calling President Obama “a communist,” “dishonest,” and “an evil man.” The next day, Griess told a colleague his upcoming call with LPL would cover Bennie’s “recent string of activities; i.e., lack of … disclosure, gun slingin ads, and calling Obama a ‘communist’ and an ‘evil’ man issues.”

On the call, department employees and LPL discussed Bennie’s CD and commercial and the article about him, and the department asked for information about LPL’s review, if any, of the commercial and the article. Afterward, in an email exchange with LPL to schedule a follow-up call, Griess wrote that it “would be nice to know” if

LPL anticipate[d] imposing any kind of heightened supervision, more frequent/unannounced exam schedule, specialized advertisement approval process or other sanction(s) that may provide the Department with a little better sense that the firm is “on top of” addressing this type of activity which in turn may be of some comfort to us and really is in the best interest of the public….

LPL sent Griess materials from its review and approval of the commercial and told him it had not reviewed the article. In response to Griess’s question about LPL’s supervision of Bennie, LPL explained that since a recent internal reorganization, Bennie’s proposed advertisements were reviewed by a senior analyst. Department employees asked whether LPL had any guidelines about agents like Bennie publicly communicating their political views. LPL said it did not….

Bennie contacted Nebraska Governor David Heineman and told him the department was targeting Bennie and “harassing [Bennie] because of his political views.” Governor Heineman called Munn to discuss the situation. Afterward, Munn had Griess review a draft memorandum responding to the Governor. Griess observed the draft did not say anything about the newspaper article and explained he “felt compelled to at least mention it” because

While Mr. Bennie did not author the article, and does not appear to be subject to our regulatory purview regarding it, the comments made regarding the President etc., regardless of anyone’s political views do tend to be quite polarizing to say the least, not all that dissimilar to the firearm purchase statement. Anyway, it’s another piece of the puzzle and just saw that it was missing.

Munn agreed to “mention that situation.” …

LPL fired Bennie at the beginning of November 2010. [The district had court concluded that there the firing was not connected to Bennie’s politics, and the court of appeals concluded that this decision wasn’t clearly erroneous.-EV] …

2. Bennie then sued, arguing that the state regulators violated the First Amendment by retaliating against Bennie — through their calls to his employer — based on his political speech. The Eighth Circuit agreed that the regulators’ actions were improper:

[Even though] the state regulators were ostensibly doing their jobs and for the most part did not directly target Bennie’s political speech in their investigations, they were [not] justified in, as the district court found, “looking for reasons to go after” Bennie…. “Conduct that retaliates against the exercise of a constitutionally protected right is actionable, even if the conduct would have been proper if motivated by a different reason.” …

[We reject] the state regulators’ brief that the only thing they did wrong was “to quote Bennie’s statements about the President in [the]… email [to LPL].” Far from it.

The email was certainly problematic, because it implicitly pressured LPL to curtail Bennie’s speech to avoid problems with the department. The email was also evidence of a deeper problem, as the district court found: that the state regulators “were looking for reasons to go after” Bennie and “made regulatory inquiries of LPL that were motivated, to varying degrees, by the content of [Bennie’s] speech.”

For the state regulators to allow their apparent disagreement with or even distaste for what Bennie had to say politically, or how he said it, to influence how the department treated him and his employer was wholly inappropriate — and absolutely inconsistent with the First Amendment. That inappropriate, unconstitutional conduct was wrong, regardless of whether the state regulators revealed their retaliatory motives to LPL or anyone else or whether the consequences of their actions were severe enough to be actionable in this case….

[W]e also note Griess’s testimony that he “would be doing a disservice not only to the people of the State of Nebraska but to [him]self” if he did not investigate Bennie after reading Bennie’s comments about President Obama accompanied by the photograph of Bennie at his desk, as well as Munn’s assertion, in an email to Griess, that Bennie being photographed at his desk making political comments “would be like me standing up in front of the flags and seal in my office and talking about a topic like abortion.” Like the language quoted above, these statements appear to reflect a troubling misunderstanding of the — nonexistent — role that political speech by persons in regulated entities should play in the department’s investigatory and enforcement activities….

3. Nonetheless, the panel majority held against Bennie. Recall that the district court had made the factual finding that LPL’s firing of Bennie wasn’t caused by the regulators’ actions, and the court of appeals concluded that this finding wasn’t clearly erroneous. Because there was no job loss or other such tangible harm caused by the regulators’ actions, the legal question was whether their actions were sufficient to “chill[] an ordinary person’s speech.” The court concluded that this was a close call but deferred to the district court’s finding that an ordinary person would not have been chilled by such actions:

To be sure, “the threat” — not to mention the reality — “of continued and heightened regulatory scrutiny” sometimes can have a chilling effect, regardless of whether it ultimately results in sanctions being imposed. The increased attention from the department undeniably affected LPL — not the least by necessitating the expenditure of time and money responding to questions about Bennie — and thus had the potential to harm Bennie indirectly by turning his employer against him. However, while the record in this case might well have supported a conclusion that an ordinary person’s speech would have been chilled, it did not compel such a finding….

We are not of a definite and firm conviction that a mistake was committed by the district court such that the district court clearly erred by finding the state regulators’ actions against Bennie would not have quieted a person of ordinary firmness. Based on this standard of review, we affirm.

Judge Beam dissented in part:

A correct evaluation of the record, in my view, compels a finding that the department’s actions would deter a person of ordinary firmness from engaging in protected political speech. The court correctly recognizes that the department, which may suspend or revoke Bennie’s registration as an agent, raised the “issue” of Bennie’s political speech, inquired as to whether LPL regulated such speech, and threatened both Bennie and LPL with “whatever administrative action deemed necessary and appropriate under its authority … to insure compliance.”

Because Bennie, and therefore a similarly situated ordinary person, is employed in a profession “heavily regulated and closely overseen” by the department, the chilling effect of the department’s actions must be evaluated in that context. It is apparent that an agency, empowered to deprive those whom it regulates of their livelihood, could easily overcome the firmness of an ordinary, regulated person by “engag[ing] the punitive machinery of government.”

Bennie, as anyone else, needs to earn a living. It is unsurprising that when confronted with even “marginally increased interest” by his registering agency, Bennie’s unusually firm resolve gave way to self-censorship after the mid-2011 records request. It is clear error, I think, not to have concluded in this case that an ordinary person would have done the same.

Additionally, it is clear, as the district court recognized, that employees at the department “were bothered by the plaintiff, in no small part because of the plaintiff’s political views, or at least the manner in which he expressed those views. And that antipathy was manifested in the Department’s regulatory attention to the plaintiff.” I would therefore hold that the department’s actions were motivated in part by retaliation against Bennie’s speech and thus that each of the three elements of a First Amendment retaliation claim were satisfied here….

4. I think that, when it comes to decisions about what would “chill an ordinary person’s speech,” appellate courts should not defer to trial court findings. This sort of question isn’t a pure question of historical fact, as to which such deference is usually proper; rather, it’s a question of application of law to fact, which courts should review de novo, especially when First Amendment issues are at stake. There was some Eighth Circuit precedent suggesting that courts should indeed defer on such questions, which is why I think the Eighth Circuit should have reheard the matter en banc; I quote the amicus brief below.

But for now, whether or not Bennie should have won his case, I think that the Nebraska regulators’ actions were quite wrong, as the Eighth Circuit panel pointed out; and I thought they were worth airing.

Here, for those who are interested in the standards of review question (and who isn’t?), is the text of my amicus brief, filed on behalf of 10 professors (including myself) who “belong to the regrettably small group of law professors who have written on standards of appellate review.” Many thanks to the signatories, Profs. Bryan Adamson (Seattle University School of Law), Jonas Anderson (American University-Washington College of Law), S. Alan Childress (Tulane University Law School), Joshua B. Fischman (Northwestern University Pritzker School of Law), Steven Morrison (University of North Dakota School of Law), Amanda Peters (South Texas College of Law), Max Schanzenbach (Northwestern University Pritzker School of Law), Steven Semeraro (Thomas Jefferson School of Law) and Ned Snow (University of South Carolina School of Law), for participating in this case.

Summary of Argument

This case involves a question that can arise in nearly every First Amendment retaliation appeal: Should de novo review or clear error review apply to decisions about whether a person of “ordinary firmness” would be chilled by the government action? The answer should be de novo review, under both the basic principles of appellate review and the special rules for constitutional cases set forth by the Supreme Court.

1. How a person of “ordinary firmness” would react to government action is a classic “mixed question of law and fact.” It is not a question of historical fact — how the speaker in this case actually reacted — for which clear error review may be proper. Rather, it is a question of how to apply the legal “person of ordinary firmness” standard to the historical facts.

2. This Court has held that, “We review mixed questions of law and fact de novo.” Acuity v. Johnson, 776 F.3d 588, 594 (8th Cir. 2015). Though a panel has held that the “ordinary firmness” inquiry “is usually best left to the judgment of a jury,” Garcia v. City of Trenton, 348 F.3d 726, 729 (8th Cir. 2003), it did not consider whether that inquiry is a mixed question of law and fact, and did not discuss the general rule (already established then) that such mixed questions — whether first decided by juries or by trial judges — should be reviewed de novo on appeal.

3. The Supreme Court has repeatedly held that, in constitutional cases involving such “mixed findings of law and fact,” “an appellate court has an obligation to ‘make an independent examination of the whole record’ in order to make sure that ‘the judgment does not constitute a forbidden intrusion on the field of free expression.’” Bose Corp. v. Consumers Union,/i>, 466 U.S. 485, 499, 501 (1984) (internal citations omitted). The Court has so held as to many mixed questions of law and fact, including questions that have to do with how an ordinary or reasonable person would react under given circumstances — e.g.,

  • whether certain statements are so “likely to provoke the average person to retaliation” as to be fighting words;
  • whether certain facts would warrant a police officer of “reasonable prudence” to believe that there is probable cause for a search;
  • whether certain facts would lead a “reasonable person” to feel not free to leave (the standard for determining who is in “custody” for Miranda v. Arizona purposes).

And there are three good reasons for applying independent review to such questions. First, as Bose noted, such review helps correct erroneous denials of constitutional rights. Appellate judges are as able as trial judges to apply legal standards such as “would chill a person of ordinary firmness” or “would provoke the average person to retaliation.”

Second, such review provides greater consistency and equality of treatment. The “person of ordinary firmness” standard should not differ from one trial judge or jury to another. When two different cases involve highly similar facts, they should come out the same way even in different federal courtrooms. Yet under the clear error standard of review, two trial court decisions reaching opposite results when applying the same legal standard to the same facts would both be affirmed.

Third, the process of independent appellate review of mixed questions of law and fact, whether in First Amendment cases or other cases, helps set precedents that make the legal rule clearer and more precise. Indeed, this is the way that the common law is supposed to work. If, for instance, this Court grants en banc rehearing in this case and then either affirms or reverses the district court decision using de novo review, future judges and government decision makers will know more about where the “person of ordinary firmness would be chilled” line is drawn.

But under the panel decision, which used clear error review, no such precedent is set. All people know is that a particular decision by a particular judge was not clearly erroneous — but a different judge would be free to reach the opposite result on the same facts. That is not helpful for the rational development of the law, or for maintaining viewpoint neutrality in the application of the First Amendment.

Moreover, this case offers a perfect opportunity for this Court to clarify this area of the law, because here, as the panel itself acknowledged, “[t]he standard of review … likely is dispositive,” slip op. at 8. Even if the panel was correct that the decision below was not clearly erroneous, “the record in this case might well have supported a conclusion that an ordinary person’s speech would have been chilled.” Id. at 11.

For these reasons, this Court should grant rehearing en banc, and apply the de novo standard of review to this mixed question of law and fact.


I. The “would a person of ordinary firmness be chilled?” inquiry is a mixed question of law and fact

There are questions of historical fact. There are questions of law. And there are “mixed questions of law and fact” — also called questions of the applications of law to fact — which arise when “‘[t]he historical facts are admitted or established, the rule of law is undisputed, and the issue is … whether the rule of law as applied to the established facts is or is not violated.’” Ornelas v. United States, 517 U.S. 690, 696-67 (1996) (quoting Pullman-Standard v. Swint, 456 U.S. 273, 289 n.19 (1982)).

Thus, for instance, “When does the Fourth Amendment require probable cause for a search?” is a question of law. “What did this police officer know about the defendant’s conduct?” is a question of historical fact. The probable cause question involved in Ornelas, 517 U.S. at 696 — “Are the known facts … sufficient to warrant a man of reasonable prudence in the belief that contraband or evidence of a crime will be found?” — is a mixed question of law and fact.

The panel in this case incorrectly concluded that it was reviewing a “factual finding” that, “on the evidence presented, the state regulators’ actions were ‘insufficiently substantial’ to be actionable,” slip op. at 8. But that is not a question of historical fact; it is a mixed question of law and fact.

The rule of law is undisputed: A First Amendment retaliation claim is established if “government official[s] took adverse action,” motivated at least partly by a speaker’s constitutionally protected activity, “that would chill a person of ordinary firmness from continuing in the activity,” Revels v. Vincenz, 382 F.3d 870, 876 (8th Cir. 2004). The historical facts are established. Applying the rule to the facts requires legal judgment as to whether the state regulators’ actions were “insufficiently substantial.” “[T]he issue is … whether the rule of law as applied to the established facts is or is not violated,’” which is not “a determination of historical facts, but … a mixed question of law and fact,” Ornelas, 517 U.S. at 696 (internal quotation marks omitted).

II. Under this Court’s precedents, mixed questions of law and fact related to constitutional questions are reviewed de novo

This Court has held that “We review mixed questions of law and fact de novo.” Acuity, 776 F.3d at 594. In particular, ““In an ‘appeal from a civil bench trial, we review the trial court’s findings of fact for clear error. Its conclusions of law are subject to de novo review. Mixed questions of law and fact that require the consideration of legal concepts … are also reviewed de novo.’” Meecorp Capital Markets, LLC v. PSC of Two Harbors, LLC, 776 F.3d 557, 562-63 (8th Cir. 2015) (citations omitted). When “a decision is predominantly one of determining whether the established facts fall within the relevant legal definition, albeit a constitutional definition, we apply a de novo standard of review.” Falls v. Nesbitt, 966 F.2d 375, 377 (8th Cir. 1992) (reviewing de novo application of Eighth Amendment “reckless disregard” standard); see also, e.g., Williams v. United States, 452 F.3d 1009, 1012 (8th Cir. 2006) (“‘Ineffective assistance of counsel claims raise mixed questions of law and fact, and we accordingly review those claims de novo.’”) (quoting United States v. Davis, 406 F.3d 505, 508 (8th Cir. 2005)).

III. The Supreme Court has held that mixed questions of law and fact related to constitutional questions — especially First Amendment questions — should be reviewed de novo

Indeed, the Supreme Court has concluded that the Constitution commands de novo review of mixed questions of law and fact that bear on a constitutional test. That is what the Court held as to probable cause in Ornelas, see supra Part I. It is what the Court held as to whether “a reasonable person [would] have felt he or she was not at liberty to terminate the interrogation and leave,” the test for whether a defendant was “in custody” for Miranda v. Arizona purposes. Thompson v. Keohane, 516 U.S. 99, 112-13 (1995). And that is what the Court held as to the First Amendment in Bose Corp. v. Consumers Union, 466 U.S. 485 (1984), Hurley v. Irish-American Gay, Lesbian & Bisexual Group of Boston, 515 U.S. 557 (1995), and many other cases. As the Court held in Hurley,

[T]he “requirement of independent appellate review … is a rule of federal constitutional law,” which does not limit our deference to a trial court on matters of witness credibility, but which generally requires us to “review the finding of facts by a State court … where a conclusion of law as to a Federal right and a finding of fact are so intermingled as to make it necessary, in order to pass upon the Federal question, to analyze the facts.”

515 U.S. at 567; see also Bose, 466 U.S. at 505-09 (applying this to review of federal court findings); Ridley v. Massachusetts Bay Transp. Auth., 390 F.3d 65, 75 (1st Cir. 2004) (“We engage in de novo review of ultimate conclusions of law and mixed questions of law and fact in First Amendment cases.” (citing Bose and Hurley)).

Indeed, the Supreme Court has repeatedly used such independent appellate review when a constitutional standard turns on how a reasonable or ordinary person would react to certain facts. The Court did this as to a reasonable police officer’s evaluation of probable cause in Ornelas. The Court did this as to a reasonable person’s perception of whether he or she was free to leave (and thus not in custody for Miranda purposes) in Thompson.

And in the First Amendment context, the Court has applied independent review with regard to the “fighting words” inquiry, which is “whether particular remarks” are “so inherently inflammatory as to come within that small class of ‘fighting words’ which are ‘likely to provoke the average person to retaliation, and thereby cause a breach of the peace.’” Bose, 466 U.S. at 505 (citation omitted). What actual words were said may be a pure question of historical fact. What constitutes the definition of “fighting words” may be a pure question of law. But whether the words that were said are indeed “likely to provoke the average person to retaliation” is a question of application of law to fact, which the appellate court must review independently, id. Likewise, whether government officials’ actions “would chill a person of ordinary firmness from continuing in the activity” should also be reviewed independently. See also, e.g., Braun v. Soldier of Fortune Magazine, Inc., 968 F.2d 1110, 1117, 1120 (11th Cir. 1992) (applying Bose to independently review the question whether an advertisement “would convey to a reasonably prudent publisher that it created a clearly identifiable unreasonable risk that the advertiser was available to commit serious violent crimes,” and thus would constitute unprotected commercial speech).

Indeed, in First Amendment cases, the Supreme Court has generally required independent review whenever mixed questions of law and fact arise: not just as to fighting words, but as to libel, obscenity, incitement, symbolic expression, and more. Bose, 466 U.S. at 505-08 (citing cases); Hurley, 515 U.S. at 567. (This independent review amounts to de novo review of the particular application of law to fact, Bose, 466 U.S. at 508 n.27, though not necessarily of the judgment as a whole, id. at 514 n.31.) But the fighting words example is especially relevant to this case, because it shows the need for de novo review of constitutional judgments about what a reasonable person or ordinary person would do given particular facts.

IV. The panel erred in concluding that Bose does not apply here

The panel concluded that Bose does not apply to this case, because “the issue presented here is not whether Bennie’s speech was protected by the First Amendment — no one disputes it was — but the chilling effect of the state regulators’ actions, which is not the sort of finding that might trigger” a constitutional duty of independent review. Slip op. at 9 n.3.

Yet the issue here is precisely whether Bennie’s speech was protected by the First Amendment against the retaliation that the government imposed. Though all agree that Bennie’s speech could not lead to criminal punishment, here it did lead to retaliatory targeting by regulators. The question that the district court was answering was whether the speech was constitutionally protected against such retaliatory targeting.

Consider an analogous situation: A government employee engages in speech “protected by the First Amendment” (in the sense that the speech does not fall within any First Amendment exception), but the government claims it has extra latitude to retaliate against the employee because of its special role as an employer (for instance, under the balancing test set forth by Pickering v. Bd. of Ed., 391 U.S. 563 (1968)). There too the question is whether “the chilling effect of the state [employers’] actions” should suffice to forbid such retaliation. And courts do apply Bose independent appellate review in such government employee cases just as they do in other contexts. See, e.g., Rankin v. McPherson, 483 U.S. 378, 385-87 nn.8-9 (1987) (applying independent review to the question whether employee’s speech could constitutionally form grounds for dismissing employee); Swineford v. Snyder County, 15 F.3d 1258, 1265 (3d Cir. 1994) (same); Langford v. Lane, 921 F.2d 677, 680 (6th Cir. 1991) (same); Posey v. Lake Pend Oreille School Dist. No. 84, 546 F.3d 1121, 1128 (9th Cir. 2008) (same); Powell v. Gallentine, 992 F.2d 1088, 1090 (10th Cir. 1993) (same); McMullen v. Carson, 754 F.2d 936, 938 (11th Cir. 1985) (same). When the government is acting as regulator (a context in which the government’s power to restrict speech is even less than when the government is acting as employer), independent review is likewise required.

And the rationale of Bose fully applies to retaliation cases. First, “the constitutional values protected by the [no-retaliation] rule make it imperative” that appellate judges “make sure that it is correctly applied.” Bose, 466 U.S. at 502. (Bose requires independent appellate review by intermediate appellate courts as much as by the Supreme Court. See, e.g., Carver v. Nixon, 72 F.3d 633, 639 (8th Cir. 1995).) The rule here is that speakers are protected against retaliation that would chill a person of ordinary firmness. This rule protects constitutional values just as much as do the rules defining the scope of various First Amendment exceptions.

Second, “[t]he principle of viewpoint neutrality that underlies the First Amendment … also imposes a special responsibility on judges whenever it is claimed that a particular communication is unprotected.” Bose, 466 U.S. at 505. Independent review can help prevent viewpoint discrimination, both by regulators and by factfinders, because it provides greater consistency and equality of treatment.

The “person of ordinary firmness” standard should not vary from one trial judge (or one jury) to another; indeed, such variation may often unconsciously stem from a factfinder’s sympathy or antipathy towards the speech involved in each case. Independent appellate review can help make sure that lower court decisions about what constitutes retaliation are consistent. And the fact that each such decision would set a binding precedent for future cases — cases that would foreseeably involve very different kinds of speech — can help check the natural human tendency to apply vague standards in ways that favor speakers whose views we share.

Third, “the content of the rule” involved here — just as with the rule in Bose — “is not revealed simply by its literal text, but rather is given meaning through the evolutionary process of common-law adjudication; though the source of the rule is found in the Constitution, it is nevertheless largely a judge-made rule of law.” Id. Independent appellate review of the applications of this rule “is the process through which the rule itself evolves and its integrity is maintained.” Id. at 503.

“When the standard governing the decision of a particular case is provided by the Constitution, this Court’s role” — and the role of appellate courts more generally — “in marking out the limits of the standard through the process of case-by-case adjudication is of special importance.” Id. And it is especially important that the “person of ordinary firmness” standard be better elaborated, to offer further guidance both to lower courts and to government regulators.

Finally, though the panel opinion suggested that Bennie may have waived the Bose argument, slip op. at 9 n.3, the opinion proceeded to discuss Bose, and to conclude that Bose was inapplicable, id. This conclusion will likely be treated by the bar, and by judges of this court, as circuit precedent. Only en banc review of this conclusion can correct this misconception.


Whether a government official’s actions suffice to chill a speaker of ordinary firmness is a classic mixed question of fact and law — and a question that is part of the test for what constitutes unconstitutional retaliation. Under this Court’s precedents, and under the Supreme Court’s rulings related to constitutional questions, a district court’s answer to this question should be reviewed de novo.