One would guess that the chances are pretty slim that the work of a 17th century French poet would find its way into a Chicago courtroom in 2009. But that’s the situation in this case as we try to make sense out of what has been dubbed the “cat’s paw” theory. The term derives from the fable “The Monkey and the Cat” penned by Jean de La Fontaine (1621-1695). In the tale, a clever — and rather unscrupulous — monkey persuades an unsuspecting feline to snatch chestnuts from a fire. The cat burns her paw in the process while the monkey profits, gulping down the chestnuts one by one. As understood today, a cat’s paw is a “tool” or “one used by another to accomplish his purposes.” Webster’s Third New International Dictionary (1976). More on this a little later.
Vincent Staub sued the Proctor Hospital of Peoria, Illinois, under the Uni
*651
formed Services Employment and Reemployment Rights Act (USERRA), 38 U.S.C. § 4301
et seq.,
after he was discharged from his position as an angiogra-phy technologist. An Army reservist, Staub alleged that the reasons given— insubordination, shirking, and attitude problems — were just a pretext for discrimination based on his association with the military. A jury sided with Staub, and the district court denied Proctor’s renewed motion for judgment as a matter of law or for a new trial. On appeal, Proctor argues that the court gave a faulty instruction regarding the “cat’s paw” theory and, in connection with that error, improperly admitted evidence of animus by nondecision-makers. The cat’s paw theory, which we will discuss later in more detail, is a way of proving discrimination when the decision-maker herself is admittedly unbiased; under the theory, the discriminatory animus of a nondecisionmaker is imputed to the decisionmaker where the former has singular influence over the latter and uses that influence to cause the adverse employment action.
Brewer v. Board of Trustees of University of Illinois,
Staub was a veteran member of the United States Army Reserve. Like all reservists, he was a part-time soldier, spending the bulk of his hours in the civilian world. For Staub, that meant employment as an angio tech for Proctor. Balancing work and military duties can be a complicated task, but Staub apparently managed. For a while, at least. In late 2000, some 10 years after he was hired, things began to grow a little tense. 1
It was around that time that Janice Mu-lally, second in command of the Diagnostic Imaging Department, began to prepare the department work schedules. Staub would notify Mulally of his drill and training obligations, which occupied one weekend per month and two weeks during the summer. Before Mulally took over scheduling, Staub had weekends off. But Mu-lally placed Staub back in the weekend rotation, creating conflicts with his drill schedule. Mulally did this even though she had advance notice of Staub’s military obligations, and when Staub approached her about the issue she became agitated. Beginning in 2000, the scheduling conflicts were only “occasional,” but Mulally’s attitude reflected a deeper problem. Mulally responded to Staub’s questions by throwing him out of her office and saying she “didn’t want to deal with it.” Staub found some relief by going to department head Michael Korenchuk, yet it was far from complete. Sometimes Mulally would *652 change Staub’s schedule after Korenchuk spoke with her, but other times she would post a notice on the bulletin board stating that volunteers were needed to cover the drill weekends, portraying Staub as irresponsible. And occasionally Mulally made Staub use his vacation time for drill days or scheduled him for additional shifts without notice. Mulally made her reasons plain: She called Staub’s military duties “bullshit” and said the extra shifts were his “way of paying back the department for everyone else having to bend over backwards to cover [his] schedule for the Reserves.” And it came as no surprise that Korenchuk did little to remedy the situation. Although Korenchuk only commented about Staub’s reserve duties on a “couple different occasions,” these comments were none too subtle. Korenchuk characterized drill weekends as “Army Reserve bullshit” and “a b[u]nch of smoking and joking and [a] waste of taxpayers!’] money.”
Bad as that was, things became worse in 2003. In February of that year, Staub was called to active duty for a period of up to one year. Though unforeseen circumstances cut the tour short at the 92-day mark, Staub’s return home was less than pleasant. Korenchuk told one of Staub’s coworkers, Amy Knoerle, that Mulally was “out to get” Staub. Knoerle was at a loss because she saw nothing in Staub but a hard worker and team player. However, she noticed that whenever Staub approached Mulally about drill obligations, Mulally would roll her eyes and make sighing noises.
Knoerle left her post in July of 2003, to be replaced by Leslie Sweborg. Two weeks into the job, Sweborg met Mulally and another coworker, Angie Day, for drinks after work. Expecting nothing more than casual chit-chat, Sweborg was shocked when the conversation turned to Staub. Mulally was blunt: “She said that [Staub’s] military duty had been a strain on the[ ] department” and “she did not like him as an employee.” So Mulally asked Sweborg “to help her get rid of him.” Sweborg refused. In her opinion, Staub was always competent and professional, and there was no reason for such animosity.
Day, on the other hand, shared Mulally’s dislike of Staub. In a departmental meeting on December 9, 2003, Day said Staub failed to train her properly and always seemed to “disappear” when help was needed. Sweborg defended Staub — saying he was a solid trainer and “just as available as any other tech in the department” — but Mulally shot her down. Mu-lally told Sweborg she “didn’t know what [she] was talking about.” 2
In any case, the tide was turning against Staub, and his military obligations were at least peripherally involved. On January 9, 2004, Staub received an order to report for “soldier readiness processing,” a precursor to another round of active deployment. Staub gave a copy of the order to both Korenchuk and Mulally, and Korenchuk became apprehensive. He asked Staub several times per week when he would have to ship out. Day had resigned by this point, leaving Sweborg and Staub as the only two angio techs. If Staub went on active duty, Korenchuk would have to use “rent-a-techs,” placing strain on the department’s budget.
*653 One might think this enhanced Staub’s job security, but not so. On January 27, 2004, Mulally gave Staub a written warning accusing him of shirking his duties. A bit of background information is necessary to understand what Staub allegedly failed to do, because it bore no connection to angiography. The Diagnostic Imaging Department (headed by Korenchuk) was divided into two units: one unit for angiog-raphy, and a far larger unit for traditional diagnostic imaging services like radiology, mammography, ultra sounds, CAT scans, and MRIs. Though they normally stuck to their speciality, angio techs were trained to work in both units and therefore had the ability to help out with radiology and the like when the need arose and the circumstances permitted. According to Mulally, however, Staub didn’t respect that arrangement.
On the morning of January 26, a worker from general diagnostics called Mulally to see if any of the angio techs were free to help out. Mulally in turn called Sweborg and asked if she and Staub had any patients. Sweborg said they didn’t and, according to Mulally, later admitted they were completely free from 8:30 to 9:45 a.m. Nevertheless, Sweborg and Staub failed to lend a hand. Mulally therefore issued Staub a written warning — Sweborg received one the next day — noting that he had already been warned about this behavior several times. But Staub and Sweborg dispute all that. They say they had an angio patient at 8:30 a.m., and although that case was ultimately cancelled by the doctor, they learned of the cancellation only 15 minutes prior to the call for help in general diagnostics, to which they immediately responded. Further, according to Sweborg and Staub, they had never before been instructed to report automatically to general diagnostics if they did not have angio cases. Staub thus refused to sign the warning, and he asked Korenchuk (who approved the action) why he was being targeted. Korenchuk said Mulally “had all of the pertinent facts,” and he just signed the warning “to get her off of his back.” So the warning stood, and so did its instructions for the future. Going forward, Staub was to “report to Mike [Ko-renchuk] or Jan [Mulally] when [he] ha[d] no patients and [the angio] cases [we]re eomplete[d].” He would also “remain in the general diagnostic area unless [he] specified to Mike or Jan where and why [he would] go elsewhere.”
This discipline of Staub emboldened Mu-lally. Shortly afterwards she called Staub’s Reserve Unit Administrator, Joseph Abbidini, in Bartonville, Illinois. Mu-lally had called Abbidini on a prior occasion to confirm that Staub was actually a member of the Reserves, but now she wanted to know if Staub could be excused from some of his military duties. Mulally asked Abbidini if Staub really had to attend two-week training in the summer because he was needed at work. Abbidini stated that the training was mandatory. Most Reserve members have outside employment, he explained, so excusing Staub would set an ugly precedent. Mulally’s response? She called Abbadini an “asshole” and hung up. (Again, we add that we are, as we are required to do at this stage of the proceedings, taking all facts in the light most favorable to Mr. Staub.)
After all this, there can be little dispute that Mulally didn’t like Staub, and that part of this animus flowed from his membership in the military. But it was Day’s beef with Staub — not Mulally’s — that would- ultimately get the ball rolling towards termination. On April 2, 2004, Day had a meeting with Korenchuk, Linda Buck (vice-president of Human Resources), and R. Garrett McGowan (chief operating officer). Day was upset with Korenchuk because she complained to him *654 about Staub and he did nothing in response. Day said she had difficulty working with Staub, he would “absent himself from the department,” and he tended to be “abrupt.” After Day left the room, Koren-chuk, Buck, and McGowan discussed what they should do. This wasn’t the first time McGowan had heard about “availability” problems involving Staub, so he told Ko-renchuk to work with Buck to create a plan that would solve the issue. They never found time to do that — Staub ran into trouble again and was fired three weeks later on April 20. 3
The day of reckoning started out normally enough. Staub and Sweborg worked together in the angio department all morning, finishing up around lunchtime. Hungry, yet mindful of the prior warning, Staub walked to Korenchuk’s office to tell him that he and Sweborg were going to lunch. But Korenchuk wasn’t there. So Staub walked back to the angio suite, placed a call to Korenchuk’s office, and left a voice mail informing him they were off to the cafeteria. Staub and Sweborg returned 30 minutes later and went to work on some leftover filming. Korenchuk showed up a few moments afterwards, demanding to know where they had been. He said he was “looking all over” for Staub, and Staub’s explanation — that they were only at lunch and left a voice mail— appeased him little. Korenchuk escorted Staub down to Buck’s office in Human Resources, picking up a security guard along the way. Korenchuk had met with Buck earlier in the day — informing her that Staub failed to report in as instructed and couldn’t be located — so the decision to terminate was already made. As Staub walked into the room, Buck handed him his pink slip. The guard then escorted him off the grounds. Sweborg was not disciplined, though she resigned a few days later out of disgust.
According to the written notice, Staub was discharged for failing to heed the earlier warning instructing him to report to Korenchuk whenever he had no more work in the angio department and otherwise to remain in the general diagnostics area. Without the January 27 write-up, Day’s April 2 complaint, and the event on April 20 — all of which involved unavailability or “disappearances” — Buck said she would not have fired Staub. Buck’s testimony makes it clear that although she relied on Korenchuk’s input, the ultimate decision was hers. Korenchuk “reluctantly agreed” with her decision, but it was her call to make.
Beyond consulting Korenchuk and reviewing the more recent incidents, Buck relied on past issues with Staub in making her decision. She said she heard “frequent complaints” about Staub during her first year with Proctor, 2001. And she knew of two workers who resigned because of Staub in 2002: an angio tech quit because Staub made her feel like “gum on the bottom of his shoe,” and a registered nurse gave up for similar reasons. What’s more, a recruiter told Buck she had difficulty attracting workers to angio because Staub “had a reputation.” Among other things, he was known for flirting with medical students.
Admittedly, however, Buck failed to speak with other angio techs who worked with Staub, including Sweborg, and she *655 had no idea that Mulally and Day wanted Staub fired. But Buck did review Staub’s employee file, including the good (like his most recent annual evaluation) and the bad (like the January 27 write-up).
Staub filed a grievance following the termination. Staub insisted that the January 27 write-up — containing the command he allegedly violated — was fabricated by Mulally to get him in trouble. Buck did not follow up with Mulally about this claim — though she did discuss it with another Human Resources official — and she did not investigate Staub’s contention that Mulally was out to get him because he was in the Reserves. Instead, Buck stuck with her initial assessment: Staub, despite technical competency, didn’t work well with others and deserved to be fired for insubordination. Staub’s involvement with the military played no role in her analysis.
Against this backdrop, Staub faced an uphill battle in his USERRA discrimination suit. Like other employment discrimination legislation, USERRA prohibits adverse action based on a prohibited criterion, in this case military status. 38 U.S.C. § 4311(a), (c)(1). But also as with other discrimination legislation, a plaintiff suing under USERRA does not win by showing prohibited animus by just anyone. He must show that the decisionmaker harbored animus and relied on that animus in choosing to take action. Since Buck was the decisionmaker and there was no evidence she had a problem with Staub on account of his membership in the Reserves, Staub was out of luck under the traditional rubric. But that doesn’t mean he had no case at all.
Deploying the cat’s paw theory, Staub sought to attribute Mulally’s animus to Buck, and therefore to Proctor. He posited that Mulally fed false information to Buck (i.e., that he dodged work on January 26, 2004); that Mulally was motivated to do this because he was a member of the Army Reserves; and that Buck relied on this false information (without vetting it any meaningful way) in deciding to fire him. The case made it to trial on this theory, where the jury apparently found it convincing, returning a verdict in Staub’s favor. Pursuant to the parties’ stipulation, Staub was awarded $57,640 in damages. The court then denied Proctor’s renewed motion for a new trial or judgment as a matter of law.
Proctor argues on appeal that the district court mishandled the cat’s paw theory (both in terms of instructing the jury and admitting certain evidence), and also that the evidence was insufficient. Before reaching those arguments, however, we must address Staub’s invitation to review this not as a cat’s paw case, but as a traditional discrimination suit. Staub attempts to skirt the cat’s paw analysis by arguing that Buck was not the only decision-maker. Staub says Buck shared that responsibility with Korenchuk, who admittedly made some anti-military remarks. If that were the case, it would not matter whether Buck acted as the cat’s paw, because the jury could find that a decision-maker
himself was
biased.
See
38 U.S.C. § 4311(c)(1);
Lewis v. City of Chicago,
In
Brewer,
we applied the cat’s paw concept to discrimination law. That case dealt with the “Machiavellian world of permit parking at the University of Illinois’s Urbana-Champaign campus, and the ill fortune of a student who became involved in it.”
where an employee without formal authority to materially alter the terms and conditions of a plaintiffs employment nonetheless uses her “singular influence” over an employee who does have such power to harm the plaintiff for racial reasons, the actions of the employee without formal authority are imputed to the employer and the employer is in violation of Title VII.
Id. at 917. And we noted that this influence may be exercised by, among other things, “supplying misinformation or failing to provide relevant information to the person making the employment decision.” Id.
So Brewer’s case looked strong under this formula — there was evidence that Thompson was racist, and that she influenced Hendricks’s decision by withholding the fact that she told Brewer he could park where he liked. But, alas, we held against Brewer. It wasn’t fair to impute Thompson’s animus to Hendricks, we concluded, because Hendricks looked into the situation for herself; though she “listened to the information Thompson relayed to her,” she “did not simply rely on it.” Id. at 919. From this we derived a simple rule to prevent the cat’s paw theory from spiraling out of control:
[W]here a decision maker is not wholly dependent on a single source of information, but instead conducts its own investigation into the facts relevant to the decision, the employer is not liable for an employee’s submission of misinformation to the decision maker.
Id. at 918. By asking whether the deci-sionmaker conducted her own investigation and analysis, we respected the role of the decisionmaker. We were, and remain to this day, unprepared to find an employer liable based on a nondecisionmaker’s animus unless the “decisionmaker” herself held that title only nominally. If the decision-maker wasn’t used as a cat’s paw — if she didn’t just take the monkey’s word for it, as it were — then of course the theory is not in play.
*657
We affirmed this principle in
Metzger v. Illinois State Police,
Measured against this precedent, we And that there was insufficient evidence to support a verdict against Proctor under the cat’s paw theory. But before we explain why, it is necessary to comment on the way the trial court handled the matter. We do not fault the court much for its approach — the judge certainly did an admirable job given the dearth of case law— but we agree with Proctor that the division of labor between jury and court, if not the jury instruction itself, was legally defective.
Just before the case went to trial, Proctor filed motions in limine seeking to exclude evidence of military animus on the part of individuals, principally Mulally, not involved in the decision to terminate. The court denied the motions, however, reasoning that the evidence was essential to the cat’s paw theory. But the court agreed that animus by a nondecisionmaker is only relevant if she exercised singular influence over the decisionmaker, and it instructed the jury to that effect. The instruction read as follows:
The Defendant is a corporation and can act only through its officers and employees. Animosity of a coworker toward the Plaintiff on the basis of Plaintiffs military status as a motivating factor may not be attributed to Defendant unless that co-worker exercised such singular influence over the decision-maker that the co-worker was basically the real decision maker. This influence may have been exercised by concealing relevant information from or feeding false information or selectively-chosen information to the person or persons who made the decision to discharge Plaintiff.
If the decision maker is not wholly dependent on a single source of information but instead conducts its own investigation into the facts relevant to the decision, the Defendant is not liable for a non-decision maker’s submission of misinformation or selectively chosen information or failure to provide relevant information to the decision maker. It does not matter that much if the information has come from a single, potentially biased source, so long as the decision maker does not artificially or by virtue of her role in the company limit her investigation to information from that source.
This instruction, we think, is unwieldy — a fact-driven instruction would have been far more useful 4 — but not technically wrong. It captures the essence of Brewer, telling the jury that it can only consider nondecisionmaker animosity in the case of singular influence, and even then that the employer is off the hook if the decisionmaker did her own investigation. So we reject Proctor’s challenge to the instruction itself. But a court faced with the cat’s paw theory case should not just give an instruction and ask the jury to sort it all out. The court has a critical task to perform before giving the instruction or admitting evidence of nondecision- *658 maker animus- — preferably at the summary judgment or in limine stage of the proceedings. Namely, the court should determine whether a reasonable jury could find singular influence on the evidence to be presented. If there is not sufficient evidence to support such a determination, then the court has no business admitting evidence of animus by nondecisionmakers. Admitting this sort of evidence would not only be technical legal error; it would likely be prejudicial due to the jury’s tendency to associate the nondecisionmaker’s remarks with the employer, fairly or not.
As we say, however, we do not fault the court here for failing to perform this task. Nothing in
Brewer, Metzger,
or our other cases on the cat’s paw,
see, e.g., Byrd v. Ill. Dep’t of Pub. Health,
Because the trial court in this case did not follow this procedure, Staub’s abundant evidence of Mulally’s animosity was erroneously admitted into evidence. And this error was prejudicial because the strongest proof of anti-military sentiment came from the improperly admitted evidence.
The normal remedy for prejudicial evidentiary error is a new trial.
See United States v. Garcia,
USERRA states that “[a] person who ... has performed ... service in a uniformed service shall not be denied initial employment, reemployment, retention in employment, promotion, or any benefit of employment by an employer on the basis of that ... performance of service....” 38 U.S.C. § 4311(a). A plaintiff
*659
suing under this Act must “demonstrate that he suffered an adverse employment action and that the adverse action was motivated in part by his military service.”
Maher v. City of Chicago,
The story told by the evidence is really quite plain. Apart from the friction caused by his military service, the evidence suggests that Staub, although technically competent, was prone to attitude problems. The fact that he made some friends along the way (such as Swe-borg) doesn’t diminish the fact that he offended numerous others for reasons unrelated to his participation in the Reserves. So, when Staub ran into trouble in the winter and spring of 2004, he didn’t have the safety net of a good reputation. Even if Staub behaved reasonably on the day of his discharge and' the January 27 write-up was exaggerated by Mulally, his track record nonetheless supported Buck’s action. Most importantly, Buck took this action free of any military-based animus, which Staub admits. And the cat’s paw is not applicable — even setting aside the evi-dentiary error — because a reasonable jury could not find that Mulally (or anyone else) had singular influence over Buck. To the contrary, the evidence established that Buck looked beyond what Mulally and Ko-renchuk said — remember, Korenchuk supported the firing only “reluctantly” — and determined that Staub was a liability to the company. We admit that Buck’s investigation could have been more robust,
e.g.,
she failed to pursue Staub’s theory that Mulally fabricated the write-up; had Buck done this, she may have discovered that Mulally indeed bore a great deal of anti-military animus. But the rule we developed in
Brewer
does not require the decisionmaker to be a paragon of independence. It is enough that the decision-maker “is not wholly dependent on a single source of information” and conducts her “own investigation into the facts relevant to the decision”
Brewer,
We ReveRse and Remand with instructions to enter judgment in favor of Proctor.
Notes
. Though we start our narrative in 2000 — as that is when Staub alleges his military-based problems arose — Staub's employee file was already thick by that date. In 1998, Proctor fired Staub (temporarily, as it turns out) for refusing to work past his scheduled shift. Staub was reinstated after filing a grievance, but with certain conditions in place. Those conditions addressed perceived weakness in availability, attitude, and communication. For instance, Staub was to communicate with his supervisor whenever he left his work area, and he was warned that "any insubordination, immature behavior, unprofessionalism or lack of support of [a] management decision[ ] w[ould] be grounds for immediate dismissal.” This is relevant, as we shall see, because Proctor says Staub’s latest termination was due to similar insubordination and shirking of duties.
. Apparently she did, at least to some degree. The very next day, Korenchuk completed Staub’s annual evaluation, giving him an impressive score of 97.53 out of 100. To be fair, though, Staub's evaluations from years past were not so sparkling; his attitude, professionalism, and ability to work well with others were noted as constant weaknesses. See supra n. 1.
. Though it is clear Staub was fired on the 20th, there is some ambiguity as to whether the triggering event occurred that day. Staub, Sweborg, and Korenchuk said the event occurred on the 20th, but Buck claimed it happened on the 19th and said she mulled the situation over for a day before settling on a decision. We find the collective recollection of Staub, Sweborg, and Korenchuk more credible, but either way it has no effect on the outcome of the case.
. Instead of saying "the Defendant,” for instance, the court could have said "Proctor,” replaced "decision maker” with “Buck,” and so on. Just as juries are ill-equipped to construe legalese — aren’t we all! — they are more likely to understand instructions that use facts rather than abstract terms.