290 F.3d 895
Rod GUSTAFSON and Javier Cornejo, Plaintiffs-Appellees,
Arthur JONES and Philip Arreola, Defendants-Appellants.
United States Court of Appeals, Seventh Circuit.
Argued May 15, 2001.
Decided May 17, 2002.
William R. Rettko (argued), Rettko Law Offices, Brookfield, WI, for plaintiff-appellee.
Rudolph M. Konrad (argued), Milwaukee City Attorney's Office, Milwaukee, WI, for defendant-appellant.
Before RIPPLE, MANION, and DIANE P. WOOD, Circuit Judges.
DIANE P. WOOD, Circuit Judge.
This is the second time we have been asked to consider the legal ramifications of certain actions that then-Chief of Police of the Milwaukee Police Department Philip Arreola, and then-Deputy Inspector Arthur Jones, took in November 1993 with respect to Officers Rod Gustafson and Javier Cornejo. In our first opinion, Gustafson v. Jones, 117 F.3d 1015 (7th Cir.1997) (Gustafson I), we concluded that the officers had stated a claim for retaliation in violation of their First Amendment rights that was not subject to dismissal on qualified immunity grounds. That part of the case was remanded for a trial, which took place in January 2000. The result was a jury verdict in favor of Gustafson and Cornejo, awarding each $10,000 in compensatory damages and $180,000 in punitive damages. Jones and Arreola have now appealed from the adverse verdicts. Bearing in mind the deference we owe to the jury's resolution of the contested factual issues, we affirm.
* Because the background facts became clearer at the jury trial, and because some information that was merely assumed for purposes of the Rule 12(c) judgment on the pleadings was clarified, we think it important to restate the facts in light of the full record now before us.
During the summer of 1993, the Milwaukee Police Department was concerned that it lacked the resources to respond adequately to the increased number of service calls that typically occur during the warmer time of the year. In an effort to address that problem, it instituted a policy under which officers assigned to the elite Tactical Enforcement Unit (TEU) would be split into two groups during their shifts. Some were designated as so-called "490" units, whose job would be to patrol designated districts and respond to service calls relayed through the district dispatchers. The others would retain their usual TEU designation as "700" units and perform normal TEU duties.
Prior to July 13, 1993, TEU officers assigned to 490 duty were permitted to take themselves off patrol duty to conduct follow-up investigations of crimes they had previously begun to work on, if two prerequisites were met: first, they had to have permission from their TEU supervisors, and second, they had to notify the appropriate district dispatcher of their plans. All of that changed on the night of July 13, and it is that change whose character governs the outcome of this suit. On the night of July 13, 1993, Jones was on duty as the "Field Deputy Inspector," which meant that he was in charge of the entire department. Cornejo and Gustafson were officers assigned to the TEU and they were working as partners on the night shift, 7:00 p.m. to 3:00 a.m. At roll call, they were designated squad 493 and assigned to patrol District III. They arrived in District III at approximately 8:15 p.m. Within minutes they were flagged down by Lois McDougal, a woman who a week earlier had reported to Gustafson that she and her family had been held at gunpoint by a 12-year-old member of the Vice Lords gang named Sidney. McDougal stopped the officers in order to report that her son had since told her that Sidney fired ten shots at him in an incident that occurred around July 1. McDougal then provided the officers with a number of addresses at which her son indicated Sidney might be found.
Following standard follow-up investigation protocol, Gustafson used his low-frequency radio to contact his TEU supervisors Sergeant William Skurzewski and Sergeant Gerald Filut and explained the situation. Skurzewski and Filut cleared Cornejo and Gustafson to conduct a follow-up investigation of Sidney provided they received clearance from the District III dispatcher. As instructed, the officers then contacted the dispatcher, who also cleared them to look into McDougal's information. Because of heavy rains, there were very few calls that evening. The officers visited several addresses and conducted interviews between 8:30 and 9:45 p.m. Each time they changed locations, they contacted the dispatcher and their TEU supervisors as required.
Shortly before 10:00 p.m., Gustafson and Cornejo arrived at 2453 W. Brown Street. This turned out to be Sidney's house, although only his mother was home. Once the officers explained why they were looking for Sidney, his mother became concerned that Sidney might hurt someone or himself and permitted the officers to search Sidney's room for the gun he reportedly had been wielding. Sidney's mother informed the officers that Sidney would arrive home at around 11:00 p.m. and asked that they return and arrest him before he hurt someone. The officers agreed that they would return if they were not called away to another assignment. They left the house at approximately 10:15 p.m.
When they returned to their squad car, Gustafson and Cornejo received a call from the District III dispatcher telling them to go on a side channel to talk to squad 123 from the Detective Bureau. At the same time, squad 713 called them on the low-band radio. Both calls were about the same incident; squads 123 and 713 were working an attempted homicide that Cornejo had worked on a week earlier. Squad 713 requested that Gustafson and Cornejo join them in District VII. The officers proceeded to a secure call box to explain their situation to the dispatcher.
As Cornejo got out of the car to use the call box, Gustafson got a call from Deputy Inspector Jones, `squad 197.' Without explanation or inquiry into the nature of their situation, Jones directed Gustafson to "discontinue the follow-up investigation and only take assignments from the dispatcher." The dispatcher relayed the same directive from Jones to Cornejo through the call box.
The officers were surprised at the no-questions-asked order, but they complied with it and returned to regular patrol status. They were particularly concerned that they could not continue with the Sidney investigation. Given what they had learned about Sidney and the dangers he posed, they were afraid that dropping their effort to arrest Sidney that night might put the community at risk and might expose them to potential departmental discipline or legal liability should Sidney injure someone. Gustafson and Cornejo requested that squad 713 follow up the Sidney investigation for them, but the officers in that squad refused, saying they understood Jones's order to be a clear directive to abandon the investigation.
At 11:47 p.m., Skurzewski and Filut contacted Gustafson and Cornejo and set up a meeting in District III. At the meeting, the officers asked the sergeants to follow up the Sidney investigation, but they too declined, explaining that they had been chewed out by Jones, that his order was clear, and that he had even directed Skurzewski to issue an order at roll call that 490 squads in the future were not to engage in follow-up investigations without Jones's express permission. This order represented a substantial departure from established follow-up protocol.
Jones later testified that the revised protocol was necessary to overcome the resistance of TEU officers to doing regular patrol. He issued the command to Gustafson and Cornejo on the night of July 13 because he considered the fact that they had spent the first two hours of their shift doing follow-up to be an act of resistance to Arreola's "rapid response" program, even though he had no knowledge of the nature of the follow-up they were conducting, nor did he apparently take into account the fact that they had received proper departmental authorization for every step they had taken. To the contrary, according to Skurzewski and Filut, Jones issued his `no follow-up' roll call order even after learning from the sergeants that Cornejo and Gustafson had been on the trail of a potentially armed suspect who had already fired on citizens on an earlier occasion and that they had properly cleared their investigation with the dispatcher and their supervisors.
Following their shift, Gustafson and Cornejo returned to the Department. Everyone in the TEU was talking about Jones's order to the officers and his roll call order. Supervisors and officers alike were confused and concerned about the implications of Jones's actions. The TEU officers were concerned that the roll call order put them in an untenable position. They were duty-bound to do proper follow-up investigations, and follow-up of some kind was frequently necessary. Despite that, they now understood that in every case they would be required to obtain express permission to perform follow-up work from a deputy inspector, who would likely not always be reachable. As Sergeant Michael Kuspa, one of the TEU supervisors on duty that night put it, officers feared the order `would cause problems with initial investigation in that ... there was a need to call the deputy inspector who if he was not working or we were unable to get ahold of him, if there was a need for immediate investigation to continue, we would not be able to provide that.'
The level of confusion and concern about Jones's initial order was sufficiently great that the next day the TEU's commanding officer, Lieutenant Ronald Rebernick, went over Jones's head to discuss it with Inspector Thomas Harker. Acknowledging that the order was overly restrictive, Harker permitted Rebernick to modify it somewhat. The new order prohibited 490 squads only from engaging in "self-initiated" follow-ups without first contacting their supervisors.
This revised order left both the sergeants and the officers as confused as ever. Gustafson and Cornejo remained particularly concerned. Their follow-up investigation was the result of being flagged down by a citizen. Was this self-initiated? Was it "self-initiated" if they asked for both the supervising sergeants' and area dispatcher's permission before they undertook it? And what about the Department regulations requiring timely and complete follow-up? Even under the revised order, no follow-up could be conducted without express permission from Jones. What were the officers to do at times when Jones could not be reached? Gustafson testified that he and Cornejo were also concerned that according to the order "the tactical 400 cars that were supposed to be patrol cars were not allowed to operate like all the other patrol cars in the district, ... we were just supposed to drive around and take assignments from the dispatcher."
The officers approached their sergeants with all these concerns, as well as with their particular concern about the implications of having prematurely abandoned the Sidney investigation. What if Sidney had shot someone that night? Among other concerns, Gustafson, Cornejo, and their sergeants were aware that just over a year earlier two patrol officers were fired after failing to properly follow up on information suggesting criminal conduct by Jeffrey Dahmer — a failure that the Department determined contributed to Dahmer's grisly murder of a boy. (See Balcerzak v. City of Milwaukee, 163 F.3d 993 (7th Cir.1998) (upholding summary judgment in favor of Chief Arreola and the Milwaukee Board of Fire and Police Commissioners on claims brought by two officers who were disciplined for their failure properly to investigate Dahmer's conduct in 1991 with respect to one of his victims).) Their sergeants and Lieutenant Rebernick said there was nothing they could do about the Sidney investigation or the problems created by Jones's roll call order. As one TEU sergeant testified, Rebernick informed the officers "there was nothing more to add.... The officers were to follow [the order]."
Having consulted their superiors without success, and still concerned about the untenable position that Jones's order put them in, Gustafson and Cornejo went to see the president of their union, Bradley DeBraska, "to get some answers."1 They took with them their notebooks and a supplemental investigation report on the Sidney incident that they had prepared after their July 13 shift. That report explained what they had learned through their investigation and stated that they broke off the investigation under unusual circumstances after receiving an order from Deputy Inspector Jones. Gustafson and Cornejo gave these materials to DeBraska and explained their various concerns.
After hearing out Gustafson and Cornejo, DeBraska informed them that he would do a `complete investigation' into their allegations and that if they had merit he would not only write a letter to Chief Arreola, but he would send the letter to local elected officials and the Milwaukee press. As Gustafson testified, that letter was to contain "[our] concerns that we had been limited in doing our investigation, and that the order contradicted several department rules and regulations." Although the officers were uncomfortable with the idea of going public, DeBraska convinced them that permitting him to go public with their concerns was the only way the officers could `protect [themselves] from any of these things happening, department repercussions, [and] criminal/civil liability, in the event that Sidney hurt someone.'
On July 21, 1993, DeBraska sent Arreola a two-page letter criticizing Jones's roll call order. The letter alleged that the order `contravenes a number of Milwaukee Police Department Rules and Regulations and Standard Operating Procedures' as well as the overall law enforcement objective of the Department. DeBraska expressed confusion regarding the expectations for members given the Department's past terminations of officers who did not adequately follow up investigations. As an example of the kind of problem created by the order, DeBraska described the Sidney investigation and suggested it represented a willingness by the Department to `discontinue an investigation of a serious offense, even where a known suspect was about to be arrested.' The letter did not identify Gustafson, Cornejo, or Sidney, referring instead only to Sidney's alleged conduct and the scope of the investigation. The letter closed with a request that Arreola revoke the roll call order, or at a minimum that he `obtain a legal opinion from the Attorney General' regarding officers' potential criminal liability for failing to investigate pursuant to that order and that the Department agree to hold officers harmless for any civil liability.
As he said he would, DeBraska sent copies of his letter to local elected officials and the press. In response, eight alderpersons signed a letter questioning the wisdom of Jones's order and suggesting that it be repealed. At trial, Alderwoman Suzanne Breier, one of the letter's signatories, testified that the order particularly concerned her because even at that time Jones, who had since replaced Arreola as Chief, `tried to do what he felt like doing and got away with it' and that in this case the order appeared to impose an unwarranted limitation on the investigation of serious criminal activity and thus placed Milwaukee's citizens at an unacceptable risk: "I have to say that when someone is investigating something as serious as [the Sidney incident], obviously the only thing worse is to have a death ... and I certainly would not want [the investigation] handled that way if it did happen in my district. That puts everyone in danger as far as I am concerned."
On July 22, 1993 both the Milwaukee Journal and the Milwaukee Sentinel ran articles under similar headlines, "Police union questions rule on follow-ups" and "Police union objects to order." The letter and both articles identified the order as having issued from Jones.
Arreola responded to the alderpersons' letter and the newspaper articles by instructing Inspector Harker to look into the roll call order and the events of July 13. Captain August Tjaaland was given the task of gathering information. He instructed Skurzewski and Filut to file a report on the incident. They interviewed Gustafson and Cornejo. Neither sergeant could find any evidence that the officers had violated any departmental rules or orders the night of July 13. Similarly, in reviewing the sergeants' report, Tjaaland found no evidence of any rule or order violations. Tjaaland's investigation resulted in a letter on July 23 from Arreola to the Alderpersons explaining the order and re assuring them that it merely reflected a resource allocation decision. Initially, it thus appeared that neither Gustafson nor Cornejo would ever receive so much as a verbal reprimand for their actions on the night of July 13. To the contrary, the investigation revealed only that they had played everything by the book.
Following the publication of the articles and Arreola's reply letter, everything remained normal in the TEU. No one, including the Department's Deputy Chief Koleas, testified to noticing any morale, disciplinary, or performance problems following the public disclosure of Jones's order, the evident public interest it had generated, and the general descriptions of the Sidney investigation. Koleas testified that he had no indication that the incident had any negative effect on morale, and that he had "no indication that it affected [the Departments] efficiency." He had no problems with either Gustafson or Cornejo following the incident, and there was no evidence that the publicity impaired the Department's investigation and ultimate arrest of Sidney or any other criminal suspect (Gustafson arrested Sidney on August 10, 1993). Lieutenant Rebernick testified that following the publication of the articles, there were no unusual disciplinary problems in the unit and that he had no knowledge of any member of the TEU who was less likely to follow or obey an order issued by Deputy Inspector Jones. Both Gustafson and Cornejo remained "very competent" officers and their actions had "no impact whatsoever" on how the unit "functioned and conducted its activity." Similarly, Sergeant Kuspa testified that following the publication of the information, the unit returned to "business as usual." No tactical officer was any less inclined to obey an order from Deputy Inspector Jones and both Gustafson and Cornejo remained "outstanding officers."
Because everything had been going so smoothly and neither Gustafson nor Cornejo had received anything other than positive performance reviews following the July 13 incident and the July 22 newspaper publications, members of the TEU — officers and supervisors alike — were stunned when nearly four months later, on November 12, Gustafson and Cornejo were involuntarily transferred from the TEU. Typically, officers left the TEU only if they sought a transfer or if they engaged in misconduct. Gustafson and Cornejo had not sought the transfer and they had not been formally disciplined for anything, much less something that would have merited a transfer. When they asked the TEU's new commanding officer, Captain Bialk, why they were being transferred, he told them it was "for the good of the unit." In fact, as Bialk later admitted at trial, the two were transferred because Jones was embarrassed by the newspaper articles questioning his order. According to Bialk, Jones had been talking to him for months on almost a weekly basis about his desire to transfer Gustafson and Cornejo.
When the involuntary transfer order came down, both Bialk and Rebernick objected to Jones and Arreola that Gustafson and Cornejo were outstanding officers, that there had been no problems in the TEU since June, and that they ought not to be transferred. The transfers occurred nonetheless, and the TEU's morale suffered as a result. There had been indirect threats from Arreola and Jones immediately following the publication of the articles that if the leaks to the press did not cease they would disband the entire TEU, but the officers assumed that by November the storm had passed.
Following their transfer, Gustafson and Cornejo repeatedly applied to be reassigned to the TEU. They regularly were ranked at the top of the list for assignment to the TEU and yet each time they were passed over for other officers ranked below them. At the time this case went to trial they continued to be barred from the unit. With no other relief in sight, they sued.
Gustafson and Cornejo's theory of their case was that Jones and Arreola transferred them and continued to prevent them from returning to the TEU in retaliation for engaging in speech protected by the First Amendment. At trial, Jones and Arreola both denied that the speech in question had anything to do with the transfer and also downplayed the significance of the speech. Significantly, they did not suggest that the speech related only to purely internal police department affairs and thus did not rise to the level of a matter of public concern.
Jones testified that he approached Arreola about transferring Gustafson and Cornejo on July 15 because he believed the officers had disobeyed orders and resisted their patrol assignment the night of July 13. He also testified that he was upset because he thought that the language in Gustafson's supplemental investigation report showed a disrespect for Jones's authority. Nevertheless, Jones conceded that the only discipline that could properly have been taken in response to an infraction related to disrespect was "remedial," "talking with the officers ... tell them it was inappropriate." When asked about the articles, Jones denied that it was even possible that they had anything to do with his recommendation to transfer the officers, for the simple reason that they were published well after he approached Arreola. He testified he had no idea that the information in those articles stemmed from Gustafson and Cornejo, or that they had spoken to their union. He even denied being aware that Gustafson and Cornejo had spoken about his order to other TEU officers and TEU supervisors. Finally, he testified that he was not at all embarrassed by the newspaper articles.
Notably absent from Jones's testimony was any indication that he had, or even feared, any difficulty with discipline in the TEU following Gustafson and Cornejo's internal and public complaints about his order. There was similarly no testimony that Gustafson, Cornejo, or any other TEU officer was resistant to his orders or showed any less respect for the Department's managerial hierarchy between July 22 and November 12, when the officers were transferred. Jones did not testify to any actual or anticipated problems with morale, nor did he indicate that the articles in any way inhibited the Sidney investigation or any other ongoing investigation.
Chief Arreola's defense at trial was that he merely rubber-stamped the formal recommendation he received to transfer Gustafson and Cornejo. He testified that he had no knowledge that the two officers took information to DeBraska or even that they complained to other TEU officers and supervisors about Jones's order. With respect to the letters from DeBraska and the alderpersons, Arreola testified that he frequently received such letters and that he simply routed them to other administrators for review. He testified that he could not recollect having read the report prepared by Captain Tjaaland documenting the events of July 13 and that he had no role in drafting the letter responding to the alderpersons. He denied that the newspaper articles either angered or embarrassed him because such criticism was a regular occurrence. According to Arreola, the whole affair "was not a significant event in any mind."
Like Jones, Arreola did not testify to any particular impact that the articles either had or that he thought they might have on the Department. Upon reviewing the information at trial he stated that the disclosure of a Department order, other than through the Department's Public Information Bureau, would be a violation of Department rules but "in a technical sense." He also testified that the disclosure of information from an ongoing felony investigation report "could be detrimental" to "the successful conclusion, if you will, of an ongoing investigation." Arreola, however, offered no testimony with respect to how he perceived the particular disclosures made by Gustafson and Cornejo. There is no evidence that he at any time feared that the articles compromised the Sidney investigation, much less that they actually did so. He also offered no testimony suggesting that the articles critical of Jones's order undermined discipline or morale in the Department or the TEU. Again, Arreola testified that he considered the publicity surrounding the incident to be nothing unusual. Similarly, he did not suggest that the publicity around Jones's order either made, or risked making, the TEU patrol squad program less effective in carrying out its mission.
In the end the jury chose not to believe Jones's and Arreola's claims that they were either ignorant of, or not concerned about, Gustafson and Cornejo's speech when they decided to transfer them and returned a verdict in favor of the officers. Whether or not we may find the commanding officers' accounts plausible, it was unquestionably the jury's prerogative to decide whose version of the entire incident to credit. After the verdict, Jones and Arreola moved for judgment as a matter of law or for a new trial, in the alternative; the court denied both motions. In this appeal, they do not challenge the jury's conclusion that they caused the officers to be transferred and kept out of the TEU because of their speech, nor do they challenge the jury's finding that absent the speech the officers would not have been transferred. Instead, Jones and Arreola contend that the district court erred in concluding as a matter of law that the officers' speech was protected from punishment by the First Amendment and that they were not entitled to qualified immunity. The trouble for Jones and Arreola is that they must live with the evidentiary record they created at trial, and on that record we can only affirm the district court's ruling.
Whether a government employee's speech is protected by the First Amendment is a question of law that the district court properly withheld from the jury, even though it may have required predicate factual determinations. Biggs v. Dupo, 892 F.2d 1298, 1300 n. 1 (7th Cir.1990). Although our review of the court's legal conclusions is de novo, we review the record as a whole, and we will accept the district court's conclusions of historical fact unless they are clearly erroneous. Falanga v. State Bar of Georgia, 150 F.3d 1333, 1335 (11th Cir.1998) (disputes of constitutional fact, as opposed to historical fact, reviewed de novo); Ezekwo v. New York City Health & Hosps. Corp., 940 F.2d 775, 780 (2d Cir.1991) (applying clearly erroneous standard to factual determinations in First Amendment retaliation case); Rankin v. Independent School Dist. No. I-3, 876 F.2d 838, 842 (10th Cir.1989) (historical facts reviewed under "traditional standard of review"). Cf. Ornelas v. United States, 517 U.S. 690, 699, 116 S.Ct. 1657, 134 L.Ed.2d 911 (1996) (requiring clear error review of historical facts in Fourth Amendment context). With respect to those issues decided by the jury and not relevant to the question whether the employees' speech was entitled to First Amendment protection, we view the facts in the light most favorable to the jury's verdict. Taylor v. Carmouche, 214 F.3d 788, 791 (7th Cir.2000).
There are four elements to a First Amendment retaliation claim in the employment context. First, the plaintiffs must prove that their speech was a matter of public concern. Next, they must prove that their speech played at least a substantial part in the employer's decision to take an adverse employment action against them. If the plaintiffs can carry their burden on these two elements, the defendants can only prevail if they prove by a preponderance of the evidence that the government's interest, as an employer, in efficiently providing government services outweighs the employees' First Amendment interests, or if they can prove that they would have disciplined the employees even in the absence of the speech. Klunk v. County of St. Joseph, 170 F.3d 772, 775 (7th Cir.1999). See also Hellstrom v. United States Dep't of Veterans Affairs, 201 F.3d 94, 97 (2d Cir.2000).
Because the jury's verdict has not been challenged, the only three questions before us are (1) Did the officers' speech touch on a matter of public concern? (2) If so, did Jones and Arreola carry their burden of demonstrating that their interests as employers outweighed the officers' interests in speaking out on this matter of public concern? (3) And, even if the speech was protected, was that fact sufficiently well established in November of 1993 to deprive the defendants of qualified immunity? Applying the line of cases that began with Pickering v. Board of Education, 391 U.S. 563, 88 S.Ct. 1731, 20 L.Ed.2d 811 (1968), we conclude that the district court properly answered all three questions in favor of Gustafson and Cornejo.
A. Matter of Public Concern
Whether a government employee's speech addresses a matter of public concern depends upon "the content, form, and context of [the speech] as revealed by the whole record." Connick v. Myers, 461 U.S. 138, 147-48, 103 S.Ct. 1684, 75 L.Ed.2d 708 (1983). Of these three factors, content is most important. Button v. Kibby-Brown, 146 F.3d 526, 529 (7th Cir.1998); Marshall v. Porter County Plan Comm'n, 32 F.3d 1215, 1219 (7th Cir.1994); Belk v. Town of Minocqua, 858 F.2d 1258, 1264 (7th Cir.1988). The "public concern" element is satisfied if the speech can fairly be said to relate to a matter of political, social, or other concern to the community, rather than merely a personal grievance of interest only to the employee. Connick, 461 U.S. at 146, 103 S.Ct. 1684.
Gustafson and Cornejo contend on appeal that we need not reach the merits of this issue because we previously held that their speech was on a matter of public concern in Gustafson I. The decision there, they suggest, creates binding law of the case. What we said, however, was that "[b]earing in mind that we are just evaluating the pleadings, we find that the plaintiffs adequately alleged both the content of their speech and the defendants' awareness of it." 117 F.3d at 1018. We similarly emphasized the fact that our conclusions were based on the pleadings when we discussed the "public concern" element specifically. Id. Given the fact that the disposition of this issue must rest upon all relevant aspects of the speech, we conclude that we have an obligation to revisit this issue in light of the full record. See, e.g., Connick, 461 U.S. at 148, 103 S.Ct. 1684. See also Kokkinis v. Ivkovich, 185 F.3d 840, 844 (7th Cir.1999) (legal question requires that we "delve deeper into the precise content, form, and context of speech"); Campbell v. Towse, 99 F.3d 820, 827 (7th Cir.1996) (public concern determination made based on "record as a whole").
Although the final record differs in some respects from the details of the pleadings, it is the evidence at trial that controls. Looking at that record in its entirety, we conclude once again that the officers' speech was on a matter of public concern. Indeed, it would require some mental gymnastics to see it otherwise, given the high level of public interest this incident commanded in the Milwaukee press and among its elected officials — although we address the relevance of that outside interest in a moment. With respect to content, we have long recognized that "[i]t would be difficult to find a matter of greater public concern in a large metropolitan area than police protection and public safety." Auriemma v. Rice, 910 F.2d 1449, 1460 (7th Cir.1990) (en banc). More specifically, "[i]ssues involving proper allocation of police patrols and other departmental resources ... are questions of serious public import." Campbell, 99 F.3d at 828. Such issues undoubtedly "benefit from a full airing in the public marketplace of ideas and opinions." Id.
In this case, Gustafson and Cornejo spoke directly to their fellow officers and their union president about Jones's direct order to suspend further investigation into Sidney's whereabouts and his general order limiting follow-up investigations by TEU officers on patrol assignments; those reports in turn were passed along to the press and local elected officials. As we said in our earlier opinion, "[t]his was not speech about merely personal matters; it related to how police investigations are to be conducted, and what kind of balance between individual officer initiative and central control was to be struck." Gustafson I, 117 F.3d at 1019. Although evidence of actual public interest is not dispositive of whether speech is on a matter of public concern, the fact that the press and elected officials took such a keen interest in Jones's orders is certainly relevant to the issue. See Auriemma, 910 F.2d at 1460. Alderperson Breier's testimony at trial reflected the community's concern about the possibility that Jones's orders might needlessly lead to dangerous suspects like Sidney evading capture, thereby "put[ting] everyone in danger," as she expressed it.
On appeal, Jones and Arreola do not contest that the content of the speech was of interest to the public. (Their tentative suggestion that Gustafson and Cornejo were "recklessly indifferent" to the accuracy of the information they gave to DeBraska is neither supported by any fair reading of the record, much less one favorable to the officers, nor is it relevant to the question whether the issue was a matter of public concern). Having conceded content, Jones and Arreola face a difficult task. They turn instead to the second step of the Pickering analysis and argue that this is a case in which the context of the speech should trump its content. They urge us to read the record to show that Gustafson and Cornejo's speech was motivated by their frustration with the department's decision to have TEU officers perform district patrol duty and by their fear that they might suffer departmental or legal consequences if Sidney committed some crime while at large. These motivations, appellants argue, are "purely personal" and thus the speech cannot be considered to have been on a matter of public concern. After an exhaustive review of the record, however, we find that this argument is not supported by the facts, nor is it persuasive on the law.
As a legal matter, while motive is relevant to the "matter of public concern" inquiry, we have consistently held that it is not dispositive. See, e.g., Button, 146 F.3d at 529; Campbell, 99 F.3d at 827; Cliff v. Board of Sch. Comm'rs of Indianapolis, 42 F.3d 403, 409 (7th Cir.1994). Motive matters to the extent that even speech on a subject that would otherwise be of interest to the public will not be protected if "the expression addresses only the personal effect upon the employee," Marshall, 32 F.3d at 1219 (emphasis added), or if the only point of the speech was "to further some purely private interest." Kokkinis, 185 F.3d at 844 (officer's public comments on sex discrimination in department were not protected where his sole motivation was to further a personal vendetta against police chief). We emphasize the word "only" because, while speech that is only motivated by private concerns may not be protected, "[a] personal aspect contained within the motive of the speaker does not necessarily remove the speech from the scope of public concern." Greer v. Amesqua, 212 F.3d 358, 371 (7th Cir.2000) (quoting Marshall, 32 F.3d at 1219). See also Cliff, 42 F.3d at 410; Button, 146 F.3d at 529. Thus, even if Gustafson and Cornejo were advancing some private interests when they raised concerns about Jones's orders, their claim survives as long as they also intended to bring to light what they believed to be the negative law enforcement consequences of the new policy. The jury so found, and there is ample evidence in the record to support this conclusion.
When Gustafson and Cornejo talked to the other TEU officers and then with DeBraska, they were worried about potential departmental and legal penalties if Sidney committed some new crime, but their testimony makes clear that they were also concerned about the fact that Jones's order would thwart both their efforts to get a dangerous suspect off the streets and the department's efforts in such cases more generally. It is telling in that connection that their complaints to DeBraska about Jones's orders and the fact that Sidney was able to avoid capture conveyed the concerns of many of their fellow TEU officers that Jones's roll call order would force officers to abandon future investigations involving time-sensitive information and potentially dangerous suspects.
Contrary to Jones and Arreola's assertions on appeal, this is not a case like Kokkinis. In that case, an officer with a personal grudge against his supervisor took advantage of a sex discrimination complaint he knew nothing about to embarrass his supervisor in the press. Gustafson and Cornejo had personally experienced the consequences of Jones's orders, they had discussed them with their fellow officers and supervisors, and they had real and legitimate concerns about the effect of the orders on the ability of TEU officers to carry out their law enforcement duties. Unlike in Kokkinis, there was no evidence of a malicious motive, and Jones testified that the public knowledge of his order did not embarrass him. Based on the content of the speech, the public's demonstrated interest in the issues raised by the speech, and the context within which Gustafson and Cornejo spoke, we agree with the district court that the speech touched on a matter of public concern.
B. Pickering Balancing
Even if an employee's speech is on a matter of public concern, a government employer is entitled to restrict that speech if it can carry its burden of proving that the interest of the public employee as a citizen in commenting on the matter is outweighed by the interest of the state, as employer, in promoting effective and efficient public service. See Pickering, 391 U.S. at 568, 88 S.Ct. 1731; Waters v. Churchill, 511 U.S. 661, 675, 114 S.Ct. 1878, 128 L.Ed.2d 686 (1994). The stronger the employee's interest in speaking, the more substantial a showing the state must make to justify its restriction of that speech. Waters, 511 U.S. at 675, 114 S.Ct. 1878.
Pickering contemplates a highly fact-specific inquiry into a number of interrelated factors: (1) whether the speech would create problems in maintaining discipline or harmony among co-workers; (2) whether the employment relationship is one in which personal loyalty and confidence are necessary; (3) whether the speech impeded the employee's ability to perform her responsibilities; (4) the time, place, and manner of the speech; (5) the context within which the underlying dispute arose; (6) whether the matter was one on which debate was vital to informed decision-making; and (7) whether the speaker should be regarded as a member of the general public. Greer, 212 F.3d at 371.
Pickering balancing is not an exercise in judicial speculation. While it is true that in some cases the undisputed facts on summary judgment permit the resolution of a claim without a trial, that means only that the Pickering elements are assessed in light of a record free from material factual disputes. Here, after the trial, we must conduct this inquiry in light of the full record viewed in the light most favorable to the jury's verdict. This is precisely what the Supreme Court did in Connick, where its Pickering analysis looked to the actual testimony of the employee's supervisor regarding the potential impact of the employee's speech and then evaluated the other evidence in the record to determine whether it supported the employer's fears. 461 U.S. at 150-54, 103 S.Ct. 1684. We are not entitled to speculate as to what the employer might have considered the facts to be and what concerns about operational efficiencies it might have had, once the record shows what those concerns really were. To put the point another way, this is not like "rational basis" review of state legislation, under which it is enough to imagine any rational underpinning for the law the legislature chose to enact. First Amendment rights cannot be trampled based on hypothetical concerns that a governmental employer never expressed.
This court, in Jungels v. Pierce, 825 F.2d 1127 (7th Cir.1987), reversed a district court's 12(b)(6) dismissal of a First Amendment retaliation claim because its conclusion that the employee's speech could have unduly interfered with the mission of his employer was based solely on "speculation." Id. at 1132. We held that while it was "plausible" that the employee's speech properly could have been sanctioned, the dismissal could not be sustained because "[no] substantial showing" had been made indicating the disruptive potential of the speech. Id. Likewise in Kokkinis, we sustained a grant of summary judgment in favor of the employer only after finding that "[e]vidence in the record ... reflects that Mr. Kokkinis' statements caused embarrassment to his superiors and coworkers and that his relationships with them deteriorated." On this basis we found the chief of police's testimony that he thought Kokkinis's comments reflected negatively on the Department and would affect morale among the officers to be reasonable and uncontested. 185 F.3d at 846.
Even accepting the proposition that a police department is a paramilitary organization built on relationships of trust and loyalty, and as such the judgment of police officials regarding the disruptive nature of an officer's speech is entitled to considerable — although by no means complete — deference, Waters, 511 U.S. at 677, 114 S.Ct. 1878, Jones and Arreola offered no evidence at trial even hinting that they were punishing the officers for their speech. Moreover, there is no evidence that Gustafson and Cornejo's speech had any disruptive effect on the department, nor is there evidence that, despite the absence of any actual disruption, Jones and Arreola reasonably believed it would have future disruptive consequences. Numerous witnesses, including the department's Deputy Chief, testified to the contrary that following publication of the articles, everything remained normal in the TEU. The officers in the TEU, including Gustafson and Cornejo, continued to respect Jones's authority, they obeyed his orders, and the unit functioned efficiently and effectively.
In response to this factual finding, Jones and Arreola now argue on appeal that an employer is allowed to act on the basis of the "potential disruptiveness" of an employee's speech and that it need not wait for "events to unfold to the extent that the disruption of the office and the destruction of working relationships is manifest before taking action." Connick, 461 U.S. at 151, 103 S.Ct. 1684. This is an accurate statement of the law, but it comes too late in the day for this case. At trial, neither Jones nor Arreola testified that they decided to transfer Gustafson and Cornejo because they feared the speech would cause future disruptions. (Had they done so, the plaintiff officers would have had an opportunity to present evidence to the contrary.) Instead, Jones flatly denied any knowledge that Gustafson and Cornejo had ever complained about his order, either to other officers, the union, or the press. He also denied that the articles embarrassed him, and he offered no testimony that he considered the public challenges to the orders (whether made by Gustafson and Cornejo or anyone else) to be a threat to his ability to manage or to Department morale. He did testify that Gustafson and Cornejo's supplemental investigation report contained inappropriate implicit challenges to his authority, but according to his own testimony this misdeed warranted no more than an informal verbal reprimand.
Arreola, meanwhile, denied being fazed at all by the articles or their content. He told the jury that he was not even certain he had read the articles. According to Arreola, such criticisms were commonplace and of no particular consequence. Nowhere did he testify that he feared that the public criticism of Jones's order would undermine the efficiency or morale of the TEU or the departmental command hierarchy. Although he testified that public disclosure of the order was "technically" a violation of Department policy and that disclosure of confidential information regarding an ongoing investigation could be disruptive to the successful conclusion of that investigation, he did not testify that such transgressions in general would justify a transfer from the TEU, much less that Gustafson and Cornejo's particular disclosures warranted a transfer. He conceded he had no indication that their disclosures of information regarding the Sidney investigation in any way hampered their ability to apprehend Sidney.
The absence of any evidence that either Jones or Arreola considered Gustafson and Cornejo's speech potentially disruptive is particularly telling in this case, given that four months had passed without any evidence of ill effects from the speech before the transfers took place. Mere assertions of a generalized potential for disruption are in any event insufficient, see Hulbert v. Wilhelm, 120 F.3d 648, 655 (7th Cir.1997) (explaining that under Connick "mere incantation of the phrase `internal harmony in the workplace' is not enough to carry the day"), but where, as here, substantial time has passed without incident, it naturally becomes more difficult for an employer to satisfy its burden of proving that punishment on the basis of anticipated disruption was reasonable. Had such evidence been presented, we would obviously have a different case; but it was not.
On appeal, Jones and Arreola pin their hopes on Kokkinis. They offer a plausible story about how, given that (1) Gustafson and Cornejo are police officers, (2) their speech challenged a lawful order of one of their superior officers, and (3) they disclosed confidential information to non-Department personnel, the Department might have believed the officers' speech would be disruptive even after four months of calm. Once again, the critical difference is that Kokkinis did not rely on a story told for the first time on appeal; it rested instead on a proper summary judgment record that contained testimonial evidence from the chief of police that he was concerned about the efficiency and morale of the police department, as well as on the testimony of other officers that supported the reasonableness of the chief's beliefs. See also Lickiss v. Drexler, 141 F.3d 1220 (7th Cir.1998) (reversing district court's grant of summary judgment to plaintiffs because reasonable inference could be drawn from evidence in the record that police supervisors were concerned about the disruptive effects of the officer's speech). In the end, this is a case of failure of proof, and should be taken as no more than that. Jones and Arreola did not meet their burden of proving by a preponderance of the evidence that any of the first three elements of the Pickering balancing test supported restricting the officers' speech.
The other critical elements of the Pickering balancing test also weigh against Jones and Arreola on this record. In particular, we have said that "[t]he manner and means of the employee's protestation are key considerations in balancing an employer's and employee's interests under Pickering." Greer, 212 F.3d at 371. Gustafson and Cornejo began by raising their concerns about Jones's orders by going up the internal chain of command, speaking first with their sergeants and then with their lieutenant. Not only could these supervisors not address their concerns, they themselves were confused and concerned that Jones's roll call order would interfere with TEU officers' ability to conduct appropriate follow-up investigations. It was only after Lieutenant Rebernick informed them that there was nothing to be done about the orders that Gustafson and Cornejo went to DeBraska. The district court concluded that Gustafson and Cornejo had properly taken their concerns up the chain of command before seeking out DeBraska, and there is ample support for this conclusion in the record. See Wright v. Illinois Dep't of Children & Fam. Servs., 40 F.3d 1492, 1504 (7th Cir.1994) (recognizing that an employee's position is stronger under Pickering where she has followed authorized procedures and appealed to appropriate authorities).
The officers went to DeBraska to seek his advice on how to respond to the orders. They provided him with copies of Jones's roll call order and their Sidney investigation report as well as an accurate description of the events that had transpired over the previous days. Although they may have been mistaken, neither believed that providing such information to their union president violated the Department's confidentiality policy, and the information they provided was centrally relevant to their concerns. They subsequently approved DeBraska's suggestion that he disclose portions of that information to the press and public officials, but only after concluding that disclosure was necessary given the nature of their concerns and their supervisors' unwillingness to do anything further about the orders.
Contrary to the appellants' suggestion, this case is not at all like Greer, in which we found that a fire department was entitled to punish a firefighter who "fired off his news release" about his boss's alleged favoritism without taking advantage of any of the available internal complaint procedures and without making any effort to learn the facts of the situation he was complaining about. 212 F.3d at 371-72. Gustafson and Cornejo might have tried other avenues before agreeing to DeBraska's plan to go public, and they might have taken better care not to disclose unnecessary specifics of the Sidney investigation to their union chief (details which notably were never disclosed to elected officials or the press), but unlike in Greer, the "manner and means" of the officers' speech in this case was not so objectionable as to weigh against them in the Pickering analysis.
Like the district court, our review of the evidence persuades us that Jones and Arreola have not carried their burden under Pickering of demonstrating that their interest, as employers, in punishing Gustafson and Cornejo for their speech outweighs the officers' considerable interest in speaking out about the potential risks created by Jones's orders.
C. Qualified Immunity
Last, Jones and Arreola contend that even if we find that transferring Gustafson and Cornejo on the basis of their speech was not justified, they are entitled to qualified immunity because there was no clearly established law that would have put them on notice of this fact in November of 1993. Again, we disagree.
As we pointed out in Gustafson I, the key elements of this case have been clear for years: a public employer may not retaliate against an employee who exercises his First Amendment speech rights, including in particular through a transfer to a less desirable position, and speech about police protection and public safety raises a matter of public concern. 117 F.3d at 1021. Indeed, this court held as early as 1979 that a public employer may not retaliate against an employee's exercise of First Amendment rights by a retaliatory transfer to a different position, even if there is "no loss of pay, seniority or other rights." McGill v. Board of Educ. of Pekin Elem. School Dist. No. 108, 602 F.2d 774, 780 (7th Cir.1979); see also Walsh v. Ward, 991 F.2d 1344, 1345 (7th Cir.1993) ("McGill recognizes a proposition that cannot be denied: an employer can penalize past speech and discourage future speech by assigning a worker to an undesirable job. Dissenters exiled to Siberia (or the equivalents found within many bureaucracies) quickly get the message, even though the new postings carry the same salary and title."). As for the "public concern" point, speech related to law enforcement and police policies that have an impact on public safety was recognized as speech touching on a matter of public concern, even if the speaker was partly motivated by purely personal concerns. Auriemma, 910 F.2d at 1460 (public safety is matter of public concern); Belk, 858 F.2d at 1264 (collecting cases indicating that to be motivated in part by personal stake in outcome of dispute does not necessarily deprive speech of First Amendment protection). We repeat again that at all times relevant to this case, the law with respect to both the public concern and the anti-retaliation elements of these claims was well established.
On a different record, there might have been a question about how clear it was that the public employer could not punish employees who exercised First Amendment rights, if the employer cited efficiency concerns as its reason. But this case does not present that question. Instead, the issue is whether any employer could have thought it was entitled to punish an employee for speech on a matter of public concern where the speech caused no actual disruption of any kind for four months, and where the employer neither articulates a belief that the speech has the potential to be disruptive in the future, nor has evidence to support the reasonableness of such a belief. We need look no further than Connick to know that the answer to that question is no. The law to that extent was clearly established, and thus the district court properly denied qualified immunity to the defendants.
We emphasize again that our decision to affirm the district court's ruling on the First Amendment issue, and the jury's conclusions about the facts that gave rise to this case, is a narrow one that is driven in large part by the trial strategy that the defendants themselves chose. At trial, Chief Arreola and Deputy Inspector Jones elected not only to deny that they were motivated to transfer Gustafson and Cornejo because of their speech, but also to deny having any knowledge of the speech (Jones) or any concerns regarding it (Arreola) at the time they recommended that the officers be transferred. Both denied that the speech embarrassed them and neither offered any testimony suggesting that the officers' speech either created, or created the potential for, the kind of disruption that would have warranted punishing speech that is on a matter of public concern. Downplaying the significance of the speech was consistent with — although not necessary to — the defenses they offered at trial. Having created this record, they are stuck with it. Gustafson and Cornejo established that their speech was on a matter of public concern and the weight of the evidence supports the conclusion that their speech could not justifiably be punished under Pickering. We therefore Affirm.
Jones and Arreola argue in their brief that Gustafson and Cornejo pleaded falsely that an anonymous source had leaked the relevant documents to the newspapers. In fact, the complaint actually says only that Gustafson and Cornejo spoke to officials and that, "not at their direction," actions were taken which resulted in adverse publicity. We do not read Gustafson and Cornejo's statement as a sweeping denial of their involvement in the publicization of the affair as opposed to an allegation that another person (DeBraska) took the lead in publicizing the affair. At trial they admitted that they were the ones who gave this information to DeBraska — a position consistent with the statement in the pleadings. While the jury could have concluded otherwise, it had before it both the accurate information and the arguably contrasting information from the pleadings. Because the jury thus made its decision upon a full record, we can find no justification for rejecting its credibility determination