Peggy WALTON, Plaintiff–Appellee, v. Ray POWELL, Defendant–Appellant, New Mexico State Land Office; Donald Britt; Delma Bearden, Defendants.
Sometimes we face questions of procedure; sometimes we face questions of substance. This appeal is heavy, very heavy, on procedure. The parties begin by debating the rules we must follow when analyzing qualified immunity appeals under Johnson v. Jones, 515 U.S. 304, 115 S.Ct. 2151, 132 L.Ed.2d 238 (1995). Next they spar over whether the eponymous burden-shifting framework of McDonnell Douglas Corp. v. Green, 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668 (1973), should or shouldn't be transplanted from the Title VII context to First Amendment retaliation disputes like theirs. Only after exhausting these procedural disputes do they at last reach the substantive question whether the plaintiff in this case has amassed sufficient evidence of a First Amendment violation to warrant trial. And once there it soon seems clear to us, as it did to the district court, that a triable claim exists worthy of a jury's time.
Our case arises from Peggy Walton's time in the New Mexico State Land Office. She started there as a political appointee of an elected Republican Land Commissioner, Patrick Lyons. By her own telling, she was a long-time member of the Republican Party, active in Republican politics, and closely associated with Mr. Lyons and his administration. But Mr. Lyons's decision not to seek reelection for a third term put Ms. Walton's job at risk: as a political appointee, after all, the next administration could easily dismiss her. Apparently seeking to avoid just this eventuality, Mr. Lyons decided to appoint Ms. Walton to a senior civil service job where she'd be protected by state law against removal for political reasons. And at first the move must have seemed prescient for Ray Powell, the Democratic candidate, won the election on a platform highly critical of Mr. Lyons and his administration.
But Ms. Walton's promotion did not go unnoticed and soon it unraveled. A local television reporter ran a report titled “[c]ronies move up as officials move out.” As its lead suggested, the report was highly critical of Mr. Lyons and Ms. Walton and charged the pair with trying to frustrate Mr. Powell's incoming administration. Another reporter introducing the story aired his view that Ms. Walton was “distinctly unqualified” for her new job and claimed the hiring was “rigged.” In a meeting just a few days after taking office, Mr. Powell suggested that the Lyons administration had mismanaged state lands and that “men in suits with guns” were going to come to the office and arrest anyone involved in the wrongdoing. Later, two of Mr. Powell's political appointees confronted Ms. Walton and accused her personally of administering an illegal land sale. Comments like these persisted until Mr. Powell decided—protected status or not—Ms. Walton had to go. Eight days after making the decision to dismiss her but before announcing it publicly, Mr. Powell held a meeting with the land office's advisory board; glared across the conference table at Ms. Walton; spoke of the television news report denouncing her appointment; and, referring to her in all but name, said he “was concerned about ․ ‘protected employees' ” who “for some reason didn't have to meet the leadership criteria” for their appointments.
That dismissal led to this lawsuit. As a protected civil service employee under state law, Ms. Walton alleged that Mr. Powell unlawfully retaliated against her for exercising her right to free political association in violation of the First Amendment and 42 U.S.C. § 1983. In reply and at summary judgment Mr. Powell claimed qualified immunity. But the district court denied that requested relief and set the case for trial—and it's this ruling we're now asked to review.
Before getting to the substance of that decision, we first have to work our way through the parties' procedural puzzles. And here we begin with Ms. Walton's challenge to our power to hear this appeal under Johnson. It's long since settled that government officials usually may take interlocutory appeals (like this one) challenging district court orders denying qualified immunity at the summary judgment stage. After all, qualified immunity is supposed to be a defense from suit and not just liability, a defense that's effectively lost if its vindication on appeal must await final judgment. Mitchell v. Forsyth, 472 U.S. 511, 525–27, 105 S.Ct. 2806, 86 L.Ed.2d 411 (1985). But to this longstanding rule the Supreme Court has added an exception of more recent vintage. In Johnson the Court indicated that, when reviewing an order denying qualified immunity at the summary judgment stage, a court of appeals should usually take as true the facts the district court has determined a reasonable jury could find at trial. 515 U.S. at 313. Now before us, Ms. Walton contends that Johnson 's exception to Mitchell 's rule effectively precludes us from reviewing most (or maybe all) of this appeal. In particular, she suggests we are powerless to assess the district court's holding that a reasonable jury could find her dismissal was the result of (caused by) her political affiliation.
We can see how Ms. Walton might read Johnson as standing for so much, but in our opinion it is too much. In an effort to ensure the “wise use of appellate resources,” Johnson did tell us to take as given the district court's assessment of what facts a reasonable jury could accept at trial and focus our attention instead on “abstract” questions of law. Id. at 317. But what was supposed to be a labor-saving exception has now invited new kinds of labor all its own. Often enough, a party will argue that the district court failed to identify what facts a jury might reasonably find—an assertion that requires us, first, to decide if the district court did or didn't determine the facts a jury could find and, second, to determine the facts for ourselves if the district court didn't. See, e.g., id. at 319; Behrens v. Pelletier, 516 U.S. 299, 312–13, 116 S.Ct. 834, 133 L.Ed.2d 773 (1996); Lewis v. Tripp, 604 F.3d 1221, 1225 (10th Cir.2010). Then there are the cases where the district court's assessment of the facts is “blatantly contradicted” by the record—or someone alleges it is—and we must again sort out the dispute by asking whether there is a “blatant” contradiction and, if so, what a reasonable jury could find given the record at hand. Scott v. Harris, 550 U.S. 372, 380, 127 S.Ct. 1769, 167 L.Ed.2d 686 (2007); Lewis, 604 F.3d at 1225–26. Indeed, without special modifications like these the Johnson exception would leave appellate courts often unable to adjudicate appeals from interlocutory rulings denying qualified immunity (rendering Mitchell 's promise a dead letter) or bound to accept a clearly mistaken factual account and so left to decide less a case or controversy than a hypothetical question. See id. at 1226 n. 2; Mark R. Brown, The Fall and Rise of Qualified Immunity: From Hope to Harris, 9 Nev. L.J. 185, 217–24 (2008). Neither, we must always remember, is the Johnson exception applicable outside the summary judgment context. When we review denials of qualified immunity at the motion to dismiss stage or after trial the Supreme Court has told us not to apply Johnson and instead determine for ourselves de novo which facts are and are not sufficiently well-pleaded or proven that a reasonable jury could adopt them before proceeding to determine whether those facts suffice to state a claim or support the verdict. See Ashcroft v. Iqbal, 556 U.S. 662, 674–75, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009); Johnson, 515 U.S. at 316–17; Lewis, 604 F.3d at 1226; Marshall v. Columbia Lea Reg'l Hosp., 474 F.3d 733, 738–40 (10th Cir.2007). Much as we do when reviewing decisions granting summary judgment outside the qualified immunity context. Lewis, 604 F.3d at 1225. Indeed, in most every situation but those Johnson carves out appellate courts traditionally and routinely do assess de novo what facts a jury might accept and that task is not considered incompatible with normal principles of appellate review or unduly inefficient.
But however far Johnson 's exception extends and whatever its consistency with general practice or capacity to fulfill its promised efficiencies, it doesn't extend so far as to bar consideration of Mr. Powell's appeal or any part of it. To be sure, Johnson requires us to accept as true the facts the district court expressly held a reasonable jury could accept. And in our recitation above and analysis below we do just that, treating as true all the facts the district court held a reasonable jury could find even as we are quite confident Mr. Powell would dispute nearly all of them. But Johnson does not also require this court to accept the district court's assessment that those facts suffice to create a triable question on any legal element essential to liability. That latter sort of question is precisely the sort of question Johnson preserves for our review.
To be fair, we can understand why Ms. Walton might think otherwise. Indeed, we have struggled ourselves to fix the exact parameters of the Johnson innovation. But the Supreme Court has recently provided some helpful and clarifying guidance in Plumhoff v. Rickard, ––– U.S. ––––, 134 S.Ct. 2012, 188 L.Ed.2d 1056 (2014). There the Court explained that Johnson only forecloses courts of appeals from reconsidering a district court's assessment of “evidence sufficiency, i.e., which facts a party may, or may not, be able to prove at trial.” Id. at 2019. By way of illustration, the Court pointed to Johnson itself, where the parties disputed whether certain officers were or “were not present at the time of [an] alleged beating” and the district court held a reasonable jury could find the officers were indeed present. Id. It is that sort of “evidence sufficiency” question Johnson (usually) precludes us from reconsidering. Meanwhile, Johnson does not forbid a court of appeals from deciding whether the facts as determined by the district court are sufficient as a matter of law to state a triable question under each legal element essential to liability. Deciding “evidence sufficiency” questions of this sort is, instead, “a core responsibility of appellate courts, and requiring appellate courts to decide such issues is not an undue burden.” See id. Since Plumhoff, our own decisions have made this distinction clear and abided it emphatically. See Leatherwood v. Welker, 757 F.3d 1115, 1117–19 (10th Cir.2014); Felders ex rel. Smedley v. Malcolm, 755 F.3d 870, 878 (10th Cir.2014); see also Penn v. Escorsio, 764 F.3d 102, 109–10 (1st Cir.2014) (Baldock, J.). Indeed, if the rule were otherwise and we could not consider the sufficiency of the (given) facts to sustain a lawful verdict, a great many (most?) qualified immunity summary judgment appeals would be foreclosed and Mitchell 's promise of assuring a meaningful interlocutory opportunity to vindicate what is supposed to be an immunity from trial would be “irretrievably lost.” Plumhoff, 134 S.Ct. at 2019.
Neither can we discern a reason why questions of causation would be immune from this arrangement. Ms. Walton suggests that, even if we have authority to hear certain portions of this appeal, we may not hear that portion of it. In her view, the district court's assessment that a reasonable jury could find causation—that her political affiliation was a substantial or motivating factor in her dismissal—is particularly inappropriate for our review. But Mr. Powell asks us to decide whether the facts the district court held a reasonable jury could find suffice as a matter of law to permit a favorable judgment for the plaintiff on the element of causation. And that seems to us precisely the sort of question Plumhoff preserves for appellate review. Courts of appeals regularly decide whether the facts as presented (in a complaint, at summary judgment, or after trial) are enough to permit a reasonable jury to render a favorable judgment on causation (just like any other legal element essential to liability). See, e.g., W. Page Keeton et al., Prosser and Keeton on Torts § 41, at 264 (5th ed.1984) (“A mere possibility of ․ causation is not enough; and when the matter remains one of pure speculation or conjecture, or the probabilities are at best evenly balanced, it becomes the duty of the court to direct a verdict for the defendant.”); Clark Cty. Sch. Dist. v. Breeden, 532 U.S. 268, 273, 121 S.Ct. 1508, 149 L.Ed.2d 509 (2001) (per curiam) (holding as a matter of law that facts were insufficient to prove causation in a retaliation case). And Plumhoff makes clear we may do the same thing in the qualified immunity context while respecting the district court's special role in ascertaining the relevant facts for our analysis. Under Johnson, it is for the district court to tell us what facts a reasonable jury might accept as true. But under Plumhoff, it is for this court to say whether those facts, together with all reasonable inferences they permit, fall in or out of legal bounds—whether they are or are not enough as a matter of law to permit a reasonable jury to issue a verdict for the plaintiff under the terms of the governing legal test for causation or any other legal element. See, e.g., McBeth v. Himes, 598 F.3d 708, 717 (10th Cir.2010) (treating causation as an abstract legal question in an appeal from a denial of summary judgment based on qualified immunity).
Assured of our authority to hear this appeal, we still cannot reach its merits just yet. When assessing Ms. Walton's claim of unlawful retaliation under the First Amendment, the district court used the McDonnell Douglas heuristic to guide its analysis. See 411 U.S. 792, 93 S.Ct. 1817, 36 L.Ed.2d 668. On appeal, Mr. Powell urges us to do the same—to hold that, while developed in the Title VII context, McDonnell Douglas also applies to First Amendment retaliation claims, and to proceed to analyze this appeal through the prism of that tripartite test. So here we face another process fight, one requiring us to analyze how to analyze this appeal before we can actually analyze this appeal. And where before we couldn't agree with Ms. Walton, here we cannot agree with Mr. Powell. Several reasons conspire to lead us to conclude that McDonnell Douglas has no useful role to play in First Amendment retaliation cases.
In the first place, the tide of authority runs strongly against it. Neither the Supreme Court nor this one has ever used McDonnell Douglas to assess a First Amendment retaliation claim. See, e.g., Garcetti v. Ceballos, 547 U.S. 410, 126 S.Ct. 1951, 164 L.Ed.2d 689 (2006); Casey v. W. Las Vegas Indep. Sch. Dist., 473 F.3d 1323 (10th Cir.2007); Brammer–Hoelter v. Twin Peaks Charter Acad., 492 F.3d 1192 (10th Cir.2007). Admittedly, none of these cases expressly addresses the possibility of using McDonnell Douglas, so none might be fairly taken as having rejected the option. But it does seem notable that not a single judge in any of these cases expressed a yearning for McDonnell Douglas and all seemed well able to resolve the First Amendment claim at hand without resort to it. It's surely notable, too, that almost every circuit to have considered whether McDonnell Douglas should apply in First Amendment discrimination or retaliation cases has thought the idea a poor one. See, e.g., Dye v. Office of the Racing Comm'n, 702 F.3d 286, 294–95 (6th Cir.2012); Allen v. Iranon, 283 F.3d 1070, 1074–75 & n. 4 (9th Cir.2002); Acevedo–Diaz v. Aponte, 1 F.3d 62, 66–67 (1st Cir.1993). And notable that the only circuit with authority going the other way now seems uncertain. Compare Graning v. Sherburne Cty., 172 F.3d 611, 615 & n. 3 (8th Cir.1999), with Wagner v. Jones, 664 F.3d 259, 270 (8th Cir.2011).
Of course, we have to make our own independent judgment and cannot simply follow the tide of authority like lemmings. But we think the tide runs against McDonnell Douglas as strongly as it does for a good reason for the test has proven of limited value even in its native waters. This court has expressly declined to employ McDonnell Douglas even in Title VII cases at or after trial because of the confusion and complexities its application can invite. See, e.g., Whittington v. Nordam Grp., Inc., 429 F.3d 986, 997–98 (10th Cir.2005); see also U.S. Postal Serv. Bd. of Governors v. Aikens, 460 U.S. 711, 714–16, 103 S.Ct. 1478, 75 L.Ed.2d 403 (1983); St. Mary's Honor Ctr. v. Hicks, 509 U.S. 502, 513–15, 113 S.Ct. 2742, 125 L.Ed.2d 407 (1993). In the summary judgment context, too, where McDonnell Douglas is sometimes applied, it is only sometimes applied. We have used McDonnell Douglas in cases relying on circumstantial evidence but we will not use it in cases relying on direct evidence (and so have had to engage in the business of trying to police the often fine line between these kinds of evidence). See Trans World Airlines, Inc. v. Thurston, 469 U.S. 111, 121, 105 S.Ct. 613, 83 L.Ed.2d 523 (1985). And still then, in the narrow remaining class of (summary judgment, circumstantial-proof) cases, it may be that McDonnell Douglas is properly used only when the plaintiff alleges a “single” unlawful motive—and not “mixed motives”—lurking behind an adverse employment decision. See Price Waterhouse v. Hopkins, 490 U.S. 228, 246–47, 109 S.Ct. 1775, 104 L.Ed.2d 268 (1989) (plurality opinion); id. at 260 (White, J., concurring in the judgment); Twigg v. Hawker Beechcraft Corp., 659 F.3d 987, 998–99 (10th Cir.2011); Quigg v. Thomas Cty. Sch. Dist., 814 F.3d 1227, 1237–39 (11th Cir.2016) (collecting cases). A potentially crippling limitation given that Title VII's statutory language doesn't ever require plaintiffs to establish more than mixed motives to prevail. See 42 U.S.C. § 2000e–2(m); id. § 2000e–5(g)(2)(B); see also Desert Palace, Inc. v. Costa, 539 U.S. 90, 98–101, 123 S.Ct. 2148, 156 L.Ed.2d 84 (2003). And a limitation of perhaps special relevance here, given that a First Amendment retaliation claim also can be pursued using a “mixed motive” theory—by proving that political affiliation was a substantial or motivating (but not the only) factor behind an adverse employment action. See, e.g., Gann v. Cline, 519 F.3d 1090, 1093 (10th Cir.2008). Indeed, given so many complications and qualifications like these, more than a few keen legal minds have questioned whether the McDonnell Douglas game is worth the candle even in the Title VII context, let alone well adapted for new adventures in far-flung environments like this one. See, e.g., Coleman v. Donahoe, 667 F.3d 835, 863 (7th Cir.2012) (Woods, J., concurring); Wells v. Colo. Dep't of Transp., 325 F.3d 1205, 1226 (10th Cir.2003) (Hartz, J., writing separately); Timothy M. Tymkovich, The Problem with Pretext, 85 Denv. U.L.Rev. 503, 519–29 (2008).
Finally, our decision to avoid McDonnell Douglas is informed by the fact the Supreme Court has already suggested a more conventional and straightforward test of liability for use in First Amendment retaliation cases. In Mt. Healthy City School District Board of Education v. Doyle, 429 U.S. 274, 97 S.Ct. 568, 50 L.Ed.2d 471 (1977), the Court held that a First Amendment retaliation plaintiff must prove (using either direct or circumstantial evidence) that her political affiliation was a “substantial” or “motivating” factor behind the adverse employment action. Id. at 287. If the plaintiff succeeds in proving so much, the defendant-employer may yet prevail by way of affirmative defense if it can prove by a preponderance of the evidence that it “would have reached the same” adverse employment decision “even in the absence of the protected conduct.” Id. To be sure, the Court offered this test in the course of reviewing a decision after trial. But it seems to us to follow naturally from Mt. Healthy (and in line with conventional practice) that a defendant seeking to prevail at summary judgment must show a reasonable factfinder either would have to reject the plaintiff's claim on the merits or accept its affirmative defense. See, e.g., McCue v. Bradstreet, 807 F.3d 334, 344–47 (1st Cir.2015) (applying Mt. Healthy to First Amendment retaliation suit in summary judgment context); Eckerman v. Tenn. Dep't of Safety, 636 F.3d 202, 207–08 (6th Cir.2010) (same); see also Celotex Corp. v. Catrett, 477 U.S. 317, 331, 106 S.Ct. 2548, 91 L.Ed.2d 265 (1986) (“If the moving party will bear the burden of persuasion at trial, that party must support its motion with credible evidence ․ that would entitle it to a directed verdict if not controverted at trial.”).
Indeed, we can think of no reason why more by way of test or gloss might be needed. Yes, today motions practice, and especially summary judgment motions practice, seems to have assumed a place near the center of the legal universe: almost no one makes it to trial anymore. With that development surely comes a strong temptation to anoint summary judgment with unique significance and adorn it with special rules and procedures. But the truth is summary judgment was supposed to be that—summary. Not a maddening maze. Not a paper blizzard. Not a replacement for the trial as the preferred means for resolving disputes. But an expedited procedure, one tracking the motion to dismiss and the motion for judgment during and after trial, with the same standards for liability and the same essential processes used at all of these stages, the motions differentiated primarily and simply by the point in the life of the case when they happen to arise. See 10A Charles Alan Wright et al., Federal Practice and Procedure §§ 2713, 2713.1 (3d ed.1998). Better, we think to follow this traditional view, keep it simple, and employ the same and straightforward standard for liability (Mt.Healthy ) for all cases at all stages in the litigation rather than devise special and idiosyncratic (McDonnell Douglas ) rules that depend on what kind of proof you allege, what kind of case you allege, and where in the life of the litigation you happen to find yourself. For that alternative may surely promise more process—and more cost and more delay and more traps for the unwary—but we are unpersuaded it promises more accurate outcomes or that it enjoys any grounding in our rules, any written law before us, or traditional practice.
With that much resolved you might wonder if we must vacate the district court's summary judgment decision even before reaching its substance. After all, both the district court and Mr. Powell have analyzed this case using the McDonnell Douglas framework we hold inappropriate for First Amendment retaliation claims, and neither has addressed the case in light of the Mt. Healthy test we've outlined. But while that course is within our discretion to pursue, we don't think it necessary to pursue here. To prevent cases from needlessly bouncing back and forth between district and appellate courts, this court is entitled to affirm a district court on alternative grounds that court didn't consider if those grounds are adequate, apparent in the record, and sufficiently illuminated by counsel on appeal. See Richison v. Ernest Grp., Inc., 634 F.3d 1123, 1130 (10th Cir.2011).
Here, happily, we enjoy all of those things. Stripped of its McDonnell Douglas argot, Mr. Powell's appeal clearly raises three essential arguments. First, he claims that Ms. Walton's affiliation with Mr. Lyons wasn't protected by the First Amendment because it didn't involve a matter of “public concern.” Second, he suggests that, even if Ms. Walton's affiliation was protected by the First Amendment, a reasonable jury couldn't find that it was a substantial or motivating factor in her termination. Third, even assuming a reasonable jury could find so much, Mr. Powell submits that his actions weren't clearly forbidden by existing law at the time he took them so he's entitled to qualified immunity. In light of the facts the district court determined a reasonable jury could find (and to which we must defer under Johnson ) and the ample briefing we enjoy from both sides, we feel confident we can finally reach the merits of this appeal and confident too that we can safely reject each of these contentions without the necessity of a remand and further delay and wrangling between these parties.
Take the first argument first. A plaintiff alleging that a government employer retaliated against her for exercising her First Amendment right of political association generally must show that her conduct involved a matter of “public concern.” Merrifield v. Bd. of Cty. Comm'rs, 654 F.3d 1073, 1083 (10th Cir.2011). Originally drawn from the common law tort action for invasion of privacy, the public concern test requires a plaintiff to establish that her activity was “a subject of legitimate news interest” or “a subject of general interest and of value and concern to the public”—in short, the sort of activity someone might pursue in her capacity as a citizen as opposed to the sort of activity she might pursue in her capacity as an employee advancing a personnel grievance. City of San Diego v. Roe, 543 U.S. 77, 83–84, 125 S.Ct. 521, 160 L.Ed.2d 410 (2004) (quoted in Merrifield, 654 F.3d at 1084); see also Casey, 473 F.3d at 1328. For his part, Mr. Powell contends that the district court failed to apply the public concern test. And he argues that a correct application of it on appeal requires us to enter judgment in his favor.
Mr. Powell is only half right. It's true that the district court didn't acknowledge or apply the public concern test. But even so, Mr. Powell fails to give us sufficient reason to think the test unsatisfied here. This court has repeatedly held that a government employer who fires a protected civil service employee for “failing to endorse or pledge allegiance to a particular political ideology” generally does state a triable claim for retaliation in violation of the First Amendment's guarantee of free political association. See Gann, 519 F.3d at 1093–94 (internal quotation mark omitted); see also Bass v. Richards, 308 F.3d 1081, 1090–91 (10th Cir.2002); Dickeson v. Quarberg, 844 F.2d 1435, 1444–45 (10th Cir.1988). And the facts the district court held a reasonable jury could find fall pretty neatly within the parameters we set in Gann and related cases. After all and again, those facts show that Ms. Walton was at the time of her dismissal a civil service employee protected from political retaliation by state law; that she was a well-known ally of a political figure and his party and administration; that she was publicly criticized by the media as a “political crony” of that political figure; that a new administration came into power and immediately began attacking her political affiliations; and that the new administration cited her political affiliations in a negative light just days after deciding to fire her. These facts closely parallel those in Gann; Gann held a triable claim exists; so reasoning by analogy and with a healthy respect for precedent it would seem we're obliged to do the same here.
Still we must admit a caveat to our conclusion. The fact is Gann predates Merrifield and doesn't expressly address the public concern requirement Merrifield announced. As a result, we could imagine someone might try to suggest that Gann and similar precedents no longer fully control the question whether a public employee like Ms. Walton states a triable claim for unlawful retaliation—that Merrifield requires more. But any argument along these lines must be left to our imagination for Mr. Powell has attempted to pursue nothing like it in this appeal. In fact, he doesn't so much as mention Merrifield and we are, of course, under no obligation to pursue for him an argument for reversal that he has not pursued for himself. So while interesting questions about Gann 's controlling effect and its compatibility with Merrifield may remain for other parties to address in other cases, we are given no reason to address them here. For now and in the absence of any suggestion otherwise, we assume our precedents Gann and Merrifield can coexist peaceably enough and, on that premise, hold a triable claim remains.
That takes us to Mr. Powell's second argument. Here he raises the causation challenge we alluded to some time ago when assessing Ms. Walton's Johnson argument. Even assuming Ms. Walton can establish her dismissal involved a matter of public concern, and even accepting the facts the district court outlined, Mr. Powell contends that as a matter of law no reasonable factfinder could find Ms. Walton's political affiliation was, in Mt. Healthy 's words, a “substantial” or “motivating” factor in her dismissal.
But here again we do not see how we might agree. The facts the district court has given us, coupled with all the reasonable inferences Ms. Walton is due as the non-movant, leave us no reason to doubt that a reasonable jury could find her affiliation with the Lyons administration was a substantial or motivating factor in her dismissal. Again, we have the facts showing Ms. Walton's extensive political affiliations with Mr. Lyons, his administration, and party. Again, we have the news report critical of her civil service appointment and evidence Mr. Powell was aware of the report and found it credible. Again, we have the campaign of harassment from the first day of his new administration. And, again, we have close temporal proximity between Mr. Powell's comments at the land board meeting and the public announcement of Ms. Walton's termination. Taken together, this is more than enough to permit a reasonable jury to find that Ms. Walton's political affiliation was a substantial or motivating factor in her dismissal. Indeed, we have already held as much in similar circumstances. See, e.g., Mason v. Okla. Turnpike Auth., 115 F.3d 1442, 1450–52 (10th Cir.1997), overruled on other grounds by TW Telecom Holdings Inc. v. Carolina Internet Ltd., 661 F.3d 495 (10th Cir.2011). Neither does Mr. Powell's out-of-circuit authority suggest a different result. By way of illustration, he points to Garrett v. Barnes, 961 F.2d 629 (7th Cir.1992). But the Seventh Circuit there only and very modestly (and quite reasonably) held that an employer's mere knowledge that the plaintiff had supported his political opponent was insufficient by itself to permit a reasonable jury to find the plaintiff's affiliation was a substantial or motivating factor in his subsequent dismissal. See id. at 632–34. By its terms, that holding simply does not speak to the situation we face, where the facts could support an inference not merely that the employer knew of the employee's affiliation but acted on it.
With that, we finally arrive at Mr. Powell's final argument and here our analysis can be brief. Mr. Powell rightly notes that, to avoid a qualified immunity defense, a plaintiff must show not only that her federal constitutional or statutory rights were violated but also that they were so “clearly established” at the time of the violation that a reasonable official in the defendant's shoes would have known his conduct to be illegal. Anderson v. Creighton, 483 U.S. 635, 640, 107 S.Ct. 3034, 97 L.Ed.2d 523 (1987). And Mr. Powell contends that, even if he did violate Ms. Walton's First Amendment rights, those rights were not clearly established in 2010 and 2011, the time of the events underlying this suit. The problem for Mr. Powell is that back in 2008 Gann held that firing a civil service employee for refusing to show allegiance to a particular political cause was already a “clearly established” violation of the First Amendment. 519 F.3d at 1095–96. So unless and until someone explains why Gann doesn't control—and again Mr. Powell hasn't made the try—its holding would appear to resolve this case.
Having said so much, much surely remains for the parties to resolve on remand. On remand, Mr. Powell will finally have his chance to prove that Ms. Walton's facts are not worthy of belief. Maybe he can show Ms. Walton wasn't properly a protected employee or maybe he can show that he dismissed her for reasons having nothing to do with her political affiliations. On remand, too, Ms. Walton will have all the burdens to bear of a plaintiff seeking to establish a First Amendment violation. But those matters and more are for judge and jury to resolve at trial, just as the district court held.
GORSUCH, Circuit Judge.