Sanders v. BNSF Railway Co.

• Court: District Court, D. Minnesota
• Decided: October 24, 2019
• Docket: 0:17-cv-05106
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT DISTRICT OF MINNESOTA

Don Sanders, File No. 17-cv-5106 (ECT/KMM) Plaintiff,

v. OPINION AND ORDER

BNSF Railway Co.,

Defendant.

________________________________________________________________________ Lucas J. Kaster and James H. Kaster, Nichols Kaster, PLLP, Minneapolis, MN, for Plaintiff Donald Sanders.

Tracey Holmes Donesky, Stinson Leonard Street LLP, Minneapolis, MN, and Sally J. Ferguson, Arthur, Chapman, Kettering, Smetak & Pikala, PA, Minneapolis, MN, for Defendant BNSF Railway Co.

The Federal Railroad Safety Act, 49 U.S.C. § 20109 (“FRSA”), forbids a rail carrier from taking adverse action against an employee because the employee reported violations of, or refused to violate, any federal law relating to railroad safety. In this case, Plaintiff Don Sanders alleges that Defendant BNSF Railway Company violated the FRSA when it terminated his employment as a track inspector in April 2016. Sanders says essentially that BNSF fired him for reporting too many track defects. BNSF denies violating the FRSA and says it terminated Sanders for falsifying his payroll records. BNSF seeks summary judgment. This motion will be denied because a reasonable juror could conclude that BNSF terminated Sanders for reasons the FRSA prohibits and there is not clear and convincing evidence BNSF would have terminated Sanders regardless. I1 Sanders worked for BNSF as a track inspector. BNSF hired Sanders in 2007. Sanders Dep. 23–24 [ECF No. 79-24].2 He became a track inspector in approximately

2010. Id. at 26–28. As a track inspector, Sanders was responsible for identifying and reporting track defects to ensure that trains, equipment, and personnel could move safely. Id. at 34–35. Sanders could fix some more minor issues himself, but other defects required a maintenance crew to fix. Id. at 40. For such defects, depending on the type and severity of the defect, track inspectors might issue a “slow order”—ordering that trains move at a

slower speed over a particular section—or pull a section of track out of service. Id. at 38– 39, 46–47. In other cases, the track inspector might simply report the defect into an internal computer system called “TIMS,” to be repaired within 30 days. Id. 49–51. Sanders estimates that in his six years as a track inspector, he pulled tracks out of service “hundreds

1 In describing the relevant facts and resolving this motion under Rule 56(a), all of Sanders’s evidence is believed, and all justifiable inferences are drawn in his favor. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 255 (1986).

2 BNSF’s record citations omit spaces in places one would expect to find them. For example, it cited “Jones Decl.,¶2,” rather than “Jones Decl., ¶ 2,” “Ex.DD:T.10-11” instead of “Ex. DD at Tr. 10–11,” and “Sanders:24:3-10” instead of “Sanders Dep. 24:3-10.” This practice has been criticized in this District: “[O]mitting spaces is not an acceptable way to reduce a document’s word count.” Northbrook Digital, LLC v. Vendio Servs., Inc., 625 F. Supp. 2d 728, 733 (D. Minn. 2008). BNSF’s citations here represent an especially problematic example of this practice because BNSF represents that its opening and reply briefs have a combined count of 11,980 words—20 words shy of its allowed total. See LR 7.1(f)(1)(B); ECF Nos. 67-1, 85-1. Including improperly omitted spaces would have added well over 550 additional words to the cumulative total BNSF reported. In addition to violating at least the spirit of Rule 7.1, this practice is not fair to those parties who do not use this convention to comply with Local Rule 7.1’s word-count limitation. of times,” id. at 45, issued “hundreds” of slow orders, id. at 48–49, and reported “thousands” of more minor defects, id. at 53–54.

Sanders reported to roadmaster Blaine Hoppenrath, and she reported to division engineer Keith Jones. Track inspectors are supervised by a roadmaster. Sanders Dep. 59– 60. Except for a month or so beginning in January 2016 when Sanders moved to Northtown Yard in Minneapolis, where he was supervised by roadmaster Stephen Chartier,3 see Sanders Dep. at 107, 191–92, Sanders worked mainly out of BNSF’s Dayton’s Bluff Yard in St. Paul, where Hoppenrath was his direct supervisor from about

April 2015 onward. Hoppenrath Dep. at 13 [ECF No. 79-18]. She reported to division engineer Keith Jones. Id. at 70. During several conversations in or around November and December 2015, Jones voiced displeasure with Sanders’s work, including with his slow orders. See generally Mem. in Opp’n at 9–12 [ECF No. 77]. Jones variously expressed concern that he would

be fired as a result of the slow orders and defects reported in his territory, asked for Sanders’s help to keep him from getting fired, and expressed anger at Sanders for placing slow orders and for contacting the Federal Railroad Administration (“FRA”) with a safety question. Unbeknownst to Jones, Sanders recorded some of these conversations, and a

3 After about a month working under Chartier, Sanders chose to return to St. Paul and work under Hoppenrath. Chartier “was very blunt that he didn’t like [Sanders] and didn’t want [him] there[.]” Sanders Dep. at 192. Among other issues, Chartier and Sanders had a number of disagreements about what conditions constituted a defect, but Chartier believed Sanders “ha[d] BNSF and public safety in mind” and that his track inspection was improving. Chartier Dep. 136–39, 147–48 [ECF No. 68-1 at 63–64, 66]. detailed review of some of those conversation is helpful to decide BNSF’s summary- judgment motion.4 At multiple points in those recorded conversations, Jones speaks

angrily and uses profanity. The recordings are undated, but some can be assigned a date (or an approximate date) by referring to other documents in the record. Regardless, Sanders and BNSF agree that the recordings all were made while Sanders was a track inspector for BNSF. E.g., Jones Dep. at 190 [ECF No. 79-20 at 49]. Jones berated Sanders for talking about their conversations with others. In one recording, the date of which is not evident, Sanders told Jones that he wanted an instruction

he (Sanders) had received from Hoppenrath to inspect the main lines instead of the yard to be put in writing so that if something went wrong in the yard he would not be “thrown under the bus for it” and accused of failing to perform his assigned tasks. Kaster Decl. Ex. 13 at 0:12 to 1:04. Jones responded in an angry tone, saying: “Who’s throwing anybody under the bus? The only bus is when you get on conference calls and start throwing shit

out that you and I talk about behind the scenes. That’s the only bus that can happen.” Id. at 1:04 to 1:12. Jones expressed anger over Sanders’s slow-order reporting practices and asked Sanders to change his practices to protect Jones’s job. In another recorded conversation on November 9, 2015, Jones interrogated Sanders about two slow orders he had entered.

Kaster Decl. Ex. 14; see also Mem. In Opp’n at 9; Duginske Decl. Ex. BBB [ECF No. 68-

4 BNSF “preserves all objections to the admissibility of Sanders’s produced recordings,” Mem. in Supp. at 10 n.6 [ECF No. 67], but it does not argue that the recordings may not be considered as evidence in resolving the pending summary-judgment motion. 3 at 4–5]. Sanders explained the reason for the slow orders, and although Jones did not dispute the basis Sanders provided, he expressed frustration nonetheless, at least in part

because they had slowed the speed at which a “business car” carrying company higher-ups could travel, and thus reflected badly on Jones: Here’s the deal, is I got fucking people coming in from all over the place that are double checking how I manage the railroad. And they’re tell— and they’re almost wanting to take me off this fucking job and put me on another one. So I’m about to lose my job, my family’s welfare, and everything else because I can’t get a team of people that want to work with me, and want to go around me or, or do shit that just seems to be the flavor of the day, and nobody wants to work with me, this and that. So where do I start to get people to look at my best interests at heart and not think that anybody’s gonna get fired because we know when they [the defects] were put out there but they weren’t reported? Everybody’s failing here. Everybody’s failing, but you know who’s going to ultimately pay the price? It’s gonna be me.

Kaster Decl. Ex. 14 at 1:57 to 2:44. Sanders responded that he had issued orders that he believed were required, and he did not know what to do. Sanders noted that another BNSF employee who he referred to as “Ben” had told him not to report defects to TIMS and to instead report them directly to a supervisor, which would allow the defects to be handled without leaving a paper trail. Jones replied: All I know is, is we’re— we’re behind the eight ball, my ass is in the hot seat, and I don’t know what to do. And, and granted, I know that those [defects] have been out there, but condition- wise, . . . are they really something that you’re worried about, as opposed to other stuff that you got out there? I understand that the, what the EI[5] tells you, but are you really worried about, uh, something happening?

5 What “EI” stands for is not clear from the record, but it seems to refer to a source of information regarding the status of equipment. Id. at 2:45 to 4:06. Sanders provided more detail about the extent of the problem that caused him to issue the slow order, and Jones responded: I mean I guess— I guess— I see what you— You gotta do what you gotta do, but I guess I’ll just fucking pack my bags and tell the family fuck it I’ll just go back. ‘Cuz I— I tried the partnership, I tried to let everybody understand that we’re here to help each other, and it just seems like, that, you know, we— we have things that are out of spec and I understand that. And if you want to take Ben’s stuff to heart, nobody’s out here to get anybody. And I can’t get anybody to understand that I need help from you guys before we can get brought back up. And I’m not telling you that when you see something that is a condition, a cross-level, a warp, or something that is truly measured out for a defect slow order, but I’m asking for some help . . . . Nobody wants to do anything for me, they just want to do everything against me.

Id. at 4:12 to 5:29. Jones characterized his crew members as engaging in “vindictive shit,” id. at 5:35 to 5:41, and sighed, “I don’t know. I just— I’m almost at the end of being able to hold on any much longer,” id. at 8:02 to 8:09. Later, Jones complained: It just seems like I can never have the right answers for my boss. OK? ‘What is this slow order on?’ ‘Well, it’s for this.’ So and then we get that condition fixed and then all of a sudden here comes something out of left field, and we throw something else on again. . . . I mean it’s just like, I mean we know what the EI says. There’s critical joints at a location. You know what? Big fucking deal. We got other, bigger fish to fry.

Id. at 10:15 to 10:45. Later in the conversation, Jones told Sanders: All I can do is express where I sit, and then you gotta do what you gotta do. It ain’t like I can say that you’re doing anything out of line, all I can say is that I just need your help right now to keep my ass from getting fired. Id. at 13:17 to 13:30. Sanders asked if he needed to “just look the other way” to keep Jones from being fired; Jones said no, indicating that instead, Sanders should verbally report

defects to Jones and let Jones make the decision about whether to enter a slow order based not primarily on safety considerations but on Jones’s own professional interests: “I’d say well, all right, I’ve got a business car coming, I’m trying to eliminate some drama, let’s, uh, let’s hold off on those [inaudible], I’ll take the heat if something happens, and— and then I’ll get people over there to [perform maintenance].” Id. at 13:30 to 14:10. Jones rebuked Sanders for contacting the FRA. In another conversation, which

seems to have occurred on November 23, 2015, Jones expressed anger that Sanders had called FRA with a question about a safety issue. See Duginske Decl. Ex. S [ECF No. 68- 1 at 136–37]; Kaster Decl. Ex. 16. Jones and Sanders had previously spoken about whether a form of track protection, a “wrap bar,” was “within the guidelines.” Jones Dep. at 227. Jones told Sanders that it was, but Sanders evidently still had reservations, and had

contacted the FRA to inquire further, which angered Jones. Id. at 227–28. BNSF permits its employees to contact FRA for guidance on interpreting and applying safety rules, but Jones felts “betrayed” that Sanders had done so. Id. at 227–29. Jones yelled, “Why would you call them? All that does is bring up red flags that either management and track inspectors are not getting along or we’re trying to [inaudible] or whatever like that. Why

would you not call me?” Kaster Decl. Ex. 16 at 0:04 to 0:18. And again, “Why in the world would we ever call FRA about anything? Unless I’m absolutely blatantly telling you to break rules or don’t do something. Why would you ever call FRA?” Id. at 1:00 to 1:14. Jones continued: I tell ya, the maddest I’ve ever been in my life is just about three weeks ago when [unintelligible] obviously is getting information from you and all the other inspectors saying I’m telling you to do stuff wrong, and he’s gonna call FRA on me. And I’ll tell you what, nobody threatens me with that right there. And now to find out that you are making phone calls directly to FRA without including me first . . . It’s not good.

Id. at 1:23 to 1:50. Later, Jones told Sanders, “Obviously you don’t really give two shits about BNSF or you and I would be working more and more together and not bringing FRA into it.” Id. at 4:03 to 4:13. Sanders explained that FRA officials know the “rulebook” and that he had been consulting with FRA on safety regulations ever since he started working as a track inspector, but Jones accused Sanders of trying to get FRA to “start sniffing around” in Jones’s territory: That’s your whole motive. You can’t tell me you don’t have a little hidden agenda to get ‘em because I don’t do everything that you tell me to do and you think that I have you going around the books and, and cutting corners and this and that, so let’s bring FRA in here and find out that we do have some non- class tracks for 30 days.

Ex. 16 at 9:25 to 9:46. Sanders denied that he had any hidden agenda, but Jones persisted: “I have never felt more betrayed than right now, [with you] calling the FRA.” Id. at 10:40 to 10:45.6

6 Around the time this conversation occurred, another track inspector in the Twin Cities Division told Sanders that Jones had told him to remove slow orders and that his own defects sometimes disappeared from BNSF’s computer system without being repaired. See Kaster Decl. Ex. 15 at 6:29 to 8:32. BNSF objects to this recording as inadmissible hearsay, though it does not elaborate or cite authority. Reply Mem. at 6 n.5 [ECF No. 85]. This recording will not be considered in connection with this motion. The Parties may, if necessary, address more fully this exhibit’s admissibility as part of the pretrial proceedings. Sanders contacted HR to report Jones, and a future meeting was planned. Shortly after Sanders’s call with Jones about Sanders calling the FRA, on November 25, 2015, Sanders called BNSF’s Human Resources (“HR”) manager, Magenta Eggertsen,7 and

asked to meet with her. Duginske Decl. Ex. T at 1 [ECF No. 68-1 at 140–41]. Eggertsen was not available immediately and conferred with Hoppenrath to find a time she could meet with Sanders. Id. Jones told Sanders in the meantime that his slow-order reporting was “embarrassing.” On December 3, 2015, Sanders recorded another conversation with

Jones in which Sanders explained why he had put a slow order on a particular section of track, and that he had previously reported the issue but it had not been repaired. Duginske Decl. Ex. S; Kaster Decl. Ex. 18 at 0:10 to 0:50. “That’s fine,” Jones responded. “All I know is, Kramer [a welder]’s not leaving anymore. He’s not leaving on time. He’s not going home. And he’s gonna weld until he fucking falls over. I’m sick of it. He can be

pissed off at me for getting onto him for lying to him— to me, but I’m done being lied to and I’m done playing games, so I don’t care.” Kaster Decl. Ex. 18 at 0:57 to 1:14. Sanders said he had told Kramer “at 3:30, the minute we found [the defect],” to which Jones replied, “Okay. That’s fine. I’m embarrassed. I’m embarrassed every time I get something from Fort Worth that says, ‘surprising,’ or ‘what else is new,’ or all this other shit because we

7 Eggertsen’s surname is now Roberson, see Eggertsen Dep. at 4 [ECF No. 79-23], and documents filed in this litigation refer to her using both names, either alone or in combination. Because her surname at the time of the events at issue was Eggertsen, the Court uses that name here. find them all at 3:30 to 5:30 at night.” Id. at 1:14 to 1:34. He further told Sanders, “I’m not saying you’re not doing your job, but as a whole, it is embarrassing for everybody.

And if you’re not embarrassed because we’re putting these out at 3:30 at night, then obviously nobody’s taking ownership in it like we should be.” Id. at 1:53 to 2:04. Jones was worried primarily about the “perception . . . we’re just causing all kinds of ruckus” and not working during the day, only finding defects at night when they cannot be as speedily repaired. Id. at 3:38 to 4:18. “Well,” Jones said toward the end of the call, “at this point I’m in survival mode. As usual.” Id. at 7:50 to 7:58.

Sanders met with HR and reported his complaints with Jones and disclosed his recordings of their conversations. On December 7, 2015, Eggertsen and Sanders met. Duginske Decl. Ex. S. Sanders described his concerns with his treatment and played his recordings for her. Id. Within a day of when Sanders filed his first formal complaint, both Hoppenrath and Jones knew Sanders had begun communicating with HR, and Jones tried

to preemptively refute what he assumed Sanders was reporting. Kaster Decl. Ex. 19 [ECF No. 80-11]. After her meeting with Sanders, Eggertsen recommended that Jones receive coaching and counseling for his inappropriate, disrespectful communications with Sanders, but concluded that no one was retaliating against Sanders. Eggertsen Dep. at 53, 72–73 [ECF No. 79-23]; Duginske Decl. Ex. T at 2. She also recommended that “a letter of anti-

retaliation is important because Mr. Sanders feels already that he is treated different[ly].” Duginske Decl. Ex. T at 2. BNSF routinely issues anti-retaliation letters to supervisors like Jones when they are accused of harassment; such letters are “a preventive measure to ensure that employees feel comfortable bringing forward concerns to HR.” Eggertsen Dep. at 83. Eggertsen communicated her findings and recommendations up the chain the same afternoon she met with Sanders. Id. at 82–83.

Hoppenrath changed Sanders’s working conditions. While BNSF had this first formal complaint from Sanders under advisement, Hoppenrath changed aspects of Sanders’s work conditions. On December 10, 2015, Hoppenrath told Sanders he could no longer work 10-hour days four days a week (as he had been doing for months, apparently without supervisor approval), and instead needed to work eight-hour days five days per week. Duginske Decl. Ex. R [ECF No. 68-1 at 135]. Hoppenrath testified that her “other

crews don’t get to put themselves on four 10s and work a four ten schedule without approval[,] . . . [and that she addressed it] once it was brought to [her] attention.” Hoppenrath Dep. at 124–25. On December 17, 2015, she told Sanders he could no longer commute to and from work in a company vehicle. Duginske Ex. BBB [ECF No. 68-3 at 5]. When Sanders complained, Hoppenrath said employees needed supervisor approval to

take work vehicles home and “he no longer had supervisor approval.” Id. Ex. R. Hoppenrath testified that the decision to no longer let Sanders commute in a company vehicle was made as a part of a larger cost-cutting effort after decreased revenue when the Bakken oil boom dwindled. Hoppenrath Dep. at 119–20, 127–29. BNSF issued an anti-retaliation and coaching letter to Jones and determined that

Sanders’s complaints lacked merit. On December 15, 2015, BNSF issued Jones the anti- retaliation letter Eggertsen had recommended, see Duginske Decl. Ex. U [ECF No. 69], and on December 22, 2015, the company issued Jones a formal coaching and counseling letter. Duginske Decl. Ex. V [ECF No. 71]. The coaching letter informed Jones that his use of profanity during his November 23 conversation with Sanders was unacceptable and that “[a]ll BNSF employees are encouraged to report safety concerns and questions

internally and externally without reprisal.” Id. The following day, December 23, 2015, Eggertson informed Sanders by letter that, although Jones’s statements to Sanders on November 23 and December 3 were “not aligned with BNSF’s Leadership Model and applicable company policies,” “BNSF’s investigation does not substantiate your allegation that . . . Jones and . . . Hoppenrath are mistreating and singling you out.” Duginske Decl. Ex. W [ECF No. 68-2 at 1]. Nevertheless, that same letter reiterated the point made to

Jones that “[a]ll BNSF employees are encouraged to report safety concerns and questions internally and externally without reprisal.” Id. Hoppenrath changed other aspects of Sanders’s working conditions, Sanders briefly transferred to another location, and Sanders lodged a second complaint with HR. On January 6, 2019, Hoppenrath told Sanders he could no longer start his day at 6:30 a.m., and

instead had to start at 7. Duginske Decl. Ex. BBB. Hoppenrath testified that this decision, like the one limiting Sanders’s ability to commute in a company vehicle, was made as part of a larger cost-cutting effort. Hoppenrath Dep. at 119–20, 127–29. Shortly after this change to Sanders’s work hours, he transferred to the Northtown territory, where his new supervisor was roadmaster Stephen Chartier. Sanders Dep. at 192. On February 24, 2016,

during that short-lived transfer, Sanders attended a meeting in which Eggertsen provided training. Kaster Decl. Ex. 21 [ECF No. 80-13]. After the training, Sanders again told Eggertsen he thought he was being discriminated against and harassed, most recently because he was not allowed to take a company vehicle home at night. Id. at 2. She considered that conversation to be an official complaint and undertook an investigation. Eggertsen Dep. at 116–19. On March 16, 2016, as Eggertsen’s investigation of Sanders’s

February 24 complaint was underway, Sanders elected to transfer back to Dayton’s Bluff, where Hoppenrath was once again his supervisor. Duginske Decl. Ex. I at 1 [ECF No. 68- 1]; Chartier Dep. at 147 [ECF No. 26-2 at 67]; Sanders Dep. at 192. After Sanders returned to work under her supervision, Hoppenrath, with permission from Jones, began to investigate Sanders for time theft. Almost immediately after Sanders returned, Hoppenrath consulted with Jones and undertook on-the-job surveillance of

Sanders. She testified that she suspected Sanders of reporting time that he did not actually work on March 18, 2016, although in her deposition she could not recall all of the details giving rise to her suspicion and further testified that she might have misunderstood the situation. Hoppenrath Dep. at 53–55, 59, 92. At any rate, Hoppenrath wanted to further investigate her suspicions, and she sought and received permission from Jones to observe

Sanders over the weekend. Id. at 59. Hoppenrath believed she discovered that Sanders over-reported his hours. Hoppenrath and Sanders present very different, but not entirely inconsistent, versions of what occurred on March 19, the first day of Hoppenrath’s surveillance. Sanders says he initially went to BNSF’s Bridal Veil property that day, arriving at about 6:20 a.m., because

Hoppenrath had instructed him to move his belongings from the stationhouse at Bridal Veil to BNSF’s property at Dayton’s Bluff. Duginske Decl. Ex. CC at 38–41 [ECF No. 68-2 at 57–59]. But Hoppenrath did not know where Sanders was; she was periodically observing at Dayton’s Bluff, and Sanders did not arrive there until around 9:00 a.m. Id. at 10, 41. When she returned to the Dayton’s Bluff parking lot at 9:15, Sanders’s personal vehicle was in the parking lot and his hi-rail8 vehicle had been started. Duginske Decl. Ex. CC at

10. She later observed Sanders returning from the field to Dayton’s Bluff at 1:15 p.m., id. at 10–11, though Sanders argues that, given his tasks and their locations that day, he could not have returned before 1:50, see generally Mem. in Opp’n at 19 (summarizing various records). Hoppenrath then observed him leaving the property at 1:53 p.m. Duginske Decl. Ex. CC at 11. She has no knowledge of what Sanders did after that time. Hoppenrath Dep. at 106–07. Sanders says what Hoppenrath observed at 1:53 p.m. was him leaving for lunch;

Sanders says that he returned to the property after lunch, Duginske Decl. Ex. CC at 41–43, and at 2:27 p.m. he emailed his supervisors his daily log report. Kaster Decl. Ex. 24 (time- stamped e-mail) [ECF No. 80-14]. He reported working from 7:00 a.m. to 3:00 p.m. that day. Duginske Decl. Ex. CC at 12. Hoppenrath relayed her observations to Jones and to Labor Relations—after all, so far as she had seen, Sanders had worked less than six hours

but reported working eight. Duginske Decl. Ex. DD at 38–39 [ECF No. 68-2 at 170–71]. At some point in those communications, it was determined that Hoppenrath would surveil Sanders the next weekend, too, to determine whether Sanders was routinely overreporting his hours, or whether what Hoppenrath concluded from her limited observations was an aberration. Id.

8 The record contains various spellings of this word, but in all instances referring to a truck “specially equipped to drive on tracks,” which Sanders used to make his inspections. See, e.g., Mem. in Supp. at 15 n.9. HR determined that Sanders’s second complaint was without merit. On March 22, 2016, Eggertsen informed Sanders by letter that she had determined his complaint that

Hoppenrath had discriminated against him by refusing to allow him to continue commuting in a company vehicle was not substantiated. Eggertsen Dep. at 116–19; Duginske Decl. Ex. AA [ECF No. 68-2 at 16]. Specifically, Eggertsen determined that Hoppenrath’s decision was consistent with BNSF’s policy because no legitimate business reason existed to allow Sanders to use a company car to commute. Duginske Decl. Ex. AA. Hoppenrath’s investigation of Sanders continued, and Hoppenrath believed she

discovered additional time theft. Hoppenrath surveilled Sanders again the next weekend, on March 25 and 26, 2016. See generally Duginske Decl. DD. She observed that Sanders worked about three hours the first day but entered eight-and-a-half hours into BNSF’s payroll system. Id. at 13–21. The following day, she concluded based on her observations and track-authority documents that he worked about two-and-a-half hours but entered nine

hours, including one hour of overtime. Id. at 27–30. For his part, Sanders insists those were merely placeholder entries and that he intended to update his time with his foreman the following Tuesday because he had a question about time codes.9 Id. at 57, 65. Sanders suggests this practice was commonplace: a coworker submitted a declaration in this case stating that employees were able to edit their time entries “prior to and even after their time

was entered and submitted to the company,” with any changes submitted after the pay

9 Sanders did later amend his time entries for March 25 and 26—after receiving investigation notices from BNSF—but those revised submissions still claimed more time than Hoppenrath believed Sanders had worked. Duginske Decl. Ex. DD at 22–23, 30–31. period reflected in the next paycheck, and that he could recall no employee other than Sanders being reprimanded for editing their time after entering it into the system. Gaylor

Decl. ¶¶ 10, 16 [ECF No. 83]. Sanders was charged with two incidents of time theft and two hearings were held. Based on Hoppenrath’s observations, though, Jones and Labor Relations decided to charge Sanders with two separate time-theft violations, one for March 19 and the other for a period “beginning on March 25.” See generally Duginske Decl. Exs. GG & HH [ECF No. 68-2 at 233–34]. The charges triggered an investigative hearing on each charge under the

process described in the collective-bargaining agreement that governed Sanders’s employment. See generally Duginske Decl. Ex. J [ECF No. 68-1 at 110]. On April 3 and 8, 2016, BNSF held hearings on each of the time-theft charges against Sanders. Duginske Decl. Exs. CC and DD. Sanders contends the way such hearings are conducted renders them “anything but impartial,” but he does not point to any way in which his hearings

deviated from those afforded any other bargaining-unit employee facing charges of misconduct. Mem. in Opp’n at 22. Hoppenrath testified at those hearings; Jones did not. Jones Dep. at 142. Sanders was vigorously represented throughout by a union delegate. See generally Duginske Decl. Exs. CC and DD. Between the two hearings, Sanders lodged a third complaint of harassment and

retaliation with HR. On April 5, 2016, after receiving notice of the charges and after the first of the two hearings had concluded, Sanders contacted BNSF’s human resources tip line to report that Jones and Hoppenrath continued to harass him and retaliate against him. Duginske Decl. Exs. KK and SS [ECF No. 68-2 at 268–70 and 284–88]. Employee witnesses contacted in the course of the investigation of this report supported Sanders’s larger point that he was subject to push-back when he reported defects or issued slow

orders, though witnesses seem to attribute some of the most egregious conduct to Chartier, Sanders’s supervisor during his short stint at Northtown, and neither Jones nor Hoppenrath was mentioned by name. See Duginske Decl. Ex. OO [ECF No. 68-2 at 276–78]. BNSF’s investigator ultimately concluded, however, that Sanders’s allegations of harassment and retaliation were unsubstantiated. Duginske Decl. at Ex. SS. That finding was communicated to Sanders by letter dated June 21, 2016, about two months after he was

terminated, as described below. Duginske Decl. Ex. TT [ECF No. 68-2 at 289]. Jones concluded that Sanders was guilty of time theft and recommended his dismissal. After the investigative hearings on both of the time-theft charges against Sanders had concluded, but while investigation of Sanders’s April 2016 tip-line report was underway, Jones determined that Sanders had committed both of the acts of time theft with

which he had been charged and recommended in each case that Sanders be dismissed from his position. Jones Dep. at 142–45. Douglas Jensen, then the general director of line maintenance for the Twin Cities division, which included the territory Sanders worked, conferred with Jones about that dismissal recommendation and also briefly reviewed the hearing transcripts before concurring with Jones’s recommendation. Jensen Dep. at 9, 60–

61, 100 [ECF No. 79-19]. Jones communicated his dismissal recommendation to the BNSF team responsible for deciding Sanders’s fate; that team decided to terminate Jones’s employment. On April 19, 2016, Jones sent two emails to “PEPA”—Policy for Employee Performance Accountability, a team within BNSF tasked with reviewing an investigative hearing record to determine whether termination is appropriate. See Detlefsen Dep. at 8–9 [ECF No. 79-

16]. Each of Jones’s emails summarized one of the time-theft proceedings and recommended Sanders’s dismissal for each charge. Duginske Decl. Exs. UU and VV [ECF No. 68-2 at 290–93]. Stephanie Detlefsen, a PEPA team member, responded to each on April 27, 2016, supporting dismissal on the basis of time theft. Duginske Decl. Exs. UU and VV. Detlefsen testified that she did “[n]ot at all” rely on Jones’s recommendations in reaching her conclusions. Detlefsen Dep. at 41–42. She was aware from the materials she

had received that Sanders previously had made a number of complaints alleging harassment and retaliation by his supervisors, and although she consulted with BNSF’s human resources area and its legal department about those complaints, she did not conduct any independent investigation of the facts they alleged, and does not know if anyone from the departments with which she consulted did, either. Id. at 43–47. Her conclusion was

reviewed by the PEPA board, comprised of BNSF vice presidents. Id. at 68–69. Detlefsen could not recall any instance in which the Board disagreed with her recommendation. Id. at 69. Two days after Detlefsen’s recommendation, on April 29, 2016, BNSF terminated Sanders’s employment, issuing a separate dismissal notice for each charge of wage theft. Duginske Decl. Exs. XX & YY [ECF No. 68-2 at 296–97]. None of the internal or

administrative post-termination remedial processes available to Sanders have been resolved in his favor. See generally Mem. in Supp. at 23–24; Reply at 1 [ECF No. 85]; Second Duginske Decl. Ex. LLL [ECF No. 86-1]. Sanders has identified other BNSF employees who he says were found guilty of time theft but whose employment was not terminated. Sanders identifies as comparators two

other employees accused by Jones and Hoppenrath of violating the same conduct rule they accused Sanders of violating, which provides that “[e]mployees must not be . . . [d]ishonest,” but whose employment was not terminated. Duginske Decl. Ex. K [ECF No. 68-1 at 111]; Mem. in Opp’n at 24–25; Jones Dep. at 75. One, a driver referred to here as B.K., was accused of claiming to have worked hours when he was still at home, and of falsifying a federally mandated drivers’ logbook. Jones Dep. at 74. B.K. received notice

of the investigation in November 2016, and discipline was imposed the following month. Kaster Decl. Exs. 35–36 [ECF No. 80-18, 80-19]. Hoppenrath testified against him as a witness at the investigation, and Jones was involved in some capacity in imposing the discipline. Hoppenrath Dep. at 59–62; Jones Dep. at 74. It is somewhat unclear from the record, but it appears that B.K.’s conduct occurred over a span of time, perhaps as long as

three months. Kaster Decl. Ex. 36. B.K. admitted his conduct and was given a “Level S Actual Suspension” and suspended 36 days, but he was not terminated. Kaster Decl. Ex. 36. B.K. never filed an HR complaint against Jones or Hoppenrath or accused them of harassment or retaliation. Jones Dep. at 214; Hoppenrath Dep. at 62. It appears from his job that he is unlikely to have been involved in inspecting for or reporting defects internally

or to FRA, and there is no evidence suggesting he made such reports. The second comparator, referred to here as J.K., was a welder who was observed in November 2015 at the job site but not doing the work he reported doing and then lying about why he was idle. Jones Dep. at 210–11; Kaster Decl. Exs. 37–38 [ECF Nos. 80-20, 80-21]. Jones only observed that behavior on one occasion but suspected that it had been going on for a while. Jones Dep. at 212. Nevertheless, J.K. did not go through a full investigation; rather, he

admitted that he had been dishonest, accepted a formal reprimand, and received coaching. Jones Dep. at 212–13; Kaster Decl. Ex. 38. He was not terminated. Like B.K., J.K. had never filed any HR complaints or reported any retaliation or harassment by Jones, Jones Dep. at 213–14, and there is no evidence in the record of him having reported any defects internally or to the FRA. Unlike B.K. and J.K., Sanders was never given an opportunity to accept responsibility for time theft and remain employed. Jones Dep. at 213. Jones

testified that was “[b]ecause of all the efforts and the time and the communication and training and being in front of the groups about accurately reporting, being where you’re supposed to be, being honest, and [Sanders] chose not to follow instructions blatantly.” Id. BNSF points to no evidence that J.K. or B.K. received different training or communications from BNSF than Sanders did regarding BNSF’s expectations of employee conduct relating

to timekeeping. Sanders’s termination reflected positively on both Jones and Hoppenrath in their 2016 evaluations. For example, Jones’ supervisor praised him: Keith, nice job by you and your team YTD. Your team is injury free, slow orders are at an all time low, relationship[s] are good, Don Sanders is no longer working for BNSF, cost control is good, [. . .] and your team is #7 on the scorecard YTD. You’ve bounced around between #1 and #10 YTD . . . here nor there, being in the top 10 is a big deal.

Kaster Decl. Ex. 8 at 9 [ECF No. 80-7]. Based on his team’s performance, at the end of 2016, Jones received a year-end bonus of $18,130 and a $2,690 merit increase in his salary. Kaster Decl. Ex. 11 [ECF No. 80-10]. In Hoppenrath’s evaluation, Jones wrote, “You are doing very w[e]ll on you[r] scorecard measures and I am proud of your improvements in

the past few months. Kaster Decl. Ex. 7 at 5 [ECF No. 80-6]. Elsewhere he praised her for “do[ing] an excellent job with the Sanders investigation and now you see the benefits of holding your people accountable.” Id. at 6. II Summary judgment is warranted “if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.”

Fed. R. Civ. P. 56(a). A dispute over a fact is “material” only if its resolution “might affect the outcome of the suit” under the governing substantive law. Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 248 (1986). A dispute over a fact is “genuine” only if “the evidence is such that a reasonable jury could return a verdict for the nonmoving party.” Id. “The evidence of the non-movant is to be believed, and all justifiable inferences are to be drawn

in his favor.” Id. at 255; Tolan v. Cotton, 572 U.S. 650, 651 (2014). A Under the FRSA, “a rail carrier ‘may not discharge . . . or in any other way discriminate’ against an employee because he lawfully and in good faith provided information relating to, or directly assisted investigation of, conduct the employee

reasonably believed violated a Federal law relating to railroad safety,” Kuduk v. BNSF Ry. Co., 768 F.3d 786, 788 (8th Cir. 2014) (quoting 49 U.S.C. §§ 20109(a)(1) and (b)(1)(A)); for “refus[ing] to violate or assist in the violation of any Federal law, rule, or regulation relating to railroad safety or security,” 49 U.S.C. s. 20109(a)(2); or for “reporting, in good faith, a hazardous safety or security condition.’” Kuduk, 768 F.3d at 788 (quoting 49 U.S.C. §§ 20109(a)(1) and (b)(1)(A)). “The statute provides that an employee may

obtain de novo review of a retaliation claim in federal court after exhausting administrative remedies.” Id. (citing 49 U.S.C. § 20109(d)(3)). To prevail on an FRSA claim, Sanders “must establish a prima facie case by showing (i) he engaged in a protected activity; (ii) BNSF knew or suspected, actually or constructively, that he engaged in the protected activity; (iii) he suffered an adverse action; and (iv) the circumstances raise an inference that the protected activity was a contributing

factor in the adverse action.” Id. at 789 (citing 49 U.S.C. § 42121(b)(2)(B)(i); 29 C.F.R. § 1982.104(e)(2)). If Sanders can make this prima facie showing, BNSF nevertheless avoids liability if it “demonstrates, by clear and convincing evidence, that [it] would have taken the same unfavorable personnel action in the absence of [Sanders’s] protected activity.” Id. (citing § 42121(b)(2)(B)(ii)) (first alteration in original).

B BNSF does not dispute that Sanders can establish the second and third elements required for a prima facie case. That is, the company does not dispute that it knew about every act that Sanders points to as protected under the FRSA, or that his termination constitutes an adverse action.10 See Mem. in Supp. at 26–36. BNSF contests the protected-

activity element (at least in part) and the contributing-factor element, and furthermore

10 Sanders’s briefing focuses on the adverse action of termination, not any other change in his work conditions. At the hearing on this motion, Sanders’s counsel confirmed that termination is the only adverse action challenged in this case. argues that it is entitled to the affirmative defense that it would have fired Sanders for time theft even if he had not engaged in protected activity.

1 Sanders points to three categories of protected conduct he undertook. First, Sanders argues that “every time [he] reported a track defect, entered a slow order, or removed a track from service because of a defect, he reported a hazardous safety or security condition under § 20109(b)(1)(A), and thus engaged in protected conduct.” Mem. in Opp’n at 29. Although BNSF argues broadly that Sanders can show “[n]o FRSA-protected activity,” it

does not seem to dispute that reporting defects, entering slow orders, and removing tracks from service because of a defect—as Sanders undeniably did here—all constitute activity that is protected under § 20109(b)(1)(A) of the FRSA.11 See generally Mem. in Supp. at 26–27 (arguing only that Sanders cannot show he engaged in activity protected by § 20109(a)(1) and (a)(2)).

Whether or not those actions were objectively reasonable is immaterial in determining whether they were protected under § 20109(b)(1)(A). To be protected under that provision, a safety report need only be “in good faith,” 49 U.S.C. § 20109(b)(1)(A) (emphasis added)—in other words, subjectively reasonable. See Samson v. U.S. Dep’t of

11 BNSF seems to argue for the first time in its reply brief—in a footnote, and without citation to authority—that an employee is not protected by the FRSA when he performs his job duties. See Reply at 9 n.6. This argument is rejected for the reasons described in detail in Brisbois v. Soo Line R.R. Co., No. 15-CV-0570 (PJS/TNL), 2016 WL 7423387, at *4 (D. Minn. Dec. 22, 2016). To summarize, the “job-duties exception” has never been adopted by the Eighth Circuit, is not supported by the text of the statute, and would undermine Congress’s intent in enacting the FRSA. Id. Labor, 732 Fed. App’x 444, 446 (7th Cir. 2017) (distinguishing (b)(1)(A)-protected activity from (b)(2)-protected activity on that basis); see also Monohan v. BNSF Ry. Co.,

No. 4:14–cv–00305–JAJ–SBJ, 2016 WL 7426581, at *4 (S.D. Iowa May 11, 2016) (recognizing that reports need not be “reasonable” or “correct,” so long as they are made “in good faith,” a term which requires no more than subjective honesty). BNSF does not dispute that Sanders acted in subjective good faith when he reported track defects, entered slow orders, and removed tracks from service.12 Second, Sanders argues that when he “confronted his supervisors because he

believed they were instructing him to underreport track defects and slow orders,” that conduct was protected under § 201019(a)(1) and (2).13 Mem. in Opp’n at 29. It is difficult to see—and Sanders does not provide any further explanation—why the former provision, which protects railroad employees when they “provide information . . . in any investigation” by certain individuals or entities regarding certain types of safety issues,

would apply here. 49 U.S.C. § 20109(a)(1). He was not speaking about Jones or Hoppenrath in connection with an investigation into whether Jones and Hoppenrath were violating safety laws. But a jury reasonably could find that the latter provision, which

12 Even if protection under (b)(1)(A) required both subjective and objective components, no record evidence undermines the objective reasonableness of all—or some critical mass—of Sanders’s reports as a matter of law.

13 Sanders also suggests that those provisions protected him when he “contacted the FRA” but, so far as can be determined from the few materials the Parties have cited in this regard, see Kaster Decl. Ex. 16, Sanders contacted the FRA to inquire whether a wrap bar was an appropriate track protection, not to provide information relevant to an investigation, as contemplated by § 20109(a)(1). protects railroad employees when they refuse in good faith to violate any safety-related law, does apply. 49 U.S.C. § 20109(a)(2).

A jury reasonably could find based on Sanders’s recorded conversations that Jones routinely pressured Sanders hard to stop entering slow orders without Jones’s prior approval, even if Sanders had concluded in good faith that a slow order was required in light of the track conditions. Jones repeatedly worried aloud, and in catastrophic terms, that Sanders’s reports were endangering Jones’s job, and as a result his family’s financial stability. He lamented that he needed “people to look at my best interests at heart” and to

stop “do[ing] everything against me.” Kaster Decl. Ex. 14 at 2:30 to 2:33, 5:28 to 5:29. He acknowledged at various times that certain conditions Sanders had reported “have been out there,” that “we have things that are out of spec,” and that “[i]t ain’t like I can say that you’re doing anything out of line.” Id. at 3:49 to 3:53, 4:42 to 4:45, 13:22 to 13:27. But he nevertheless dismissed the safety issues Sanders had identified by saying, “You know

what? Big fucking deal.” Id. at 10:39 to 10:41. He asked for Sanders’s help “to keep my ass from getting fired,” id. at 13:26 to 13:30—a danger that, as Jones described it, existed only (or at least primarily) because of the reports Sanders had been filing. Jones first suggested, and later arguably instructed, that Sanders do as another colleague had suggested and stop reporting defects to TIMS in favor of verbally reporting them to a

supervisor. Id. at 4:45 to 4:51 (“if you want to take Ben’s stuff to heart, nobody’s out to get anybody”). Jones explained that if Sanders would bring concerns straight to him and not immediately enter a slow order, Jones would then personally decide whether and when to enter a slow order or impose other track protection based primarily on making himself look good to his bosses and only secondarily based on safety. Id. at 13:50 to 14:00 (“I’m trying to eliminate some drama, . . . let’s hold off on those . . . I’ll take the heat if something

happens.”) The evidence shows that Sanders refused, at least implicitly, to do as Jones wanted. Throughout the approximately three weeks during which Sanders recorded some of his conversations with Jones, Jones continued to pressure Sanders not to enter slow orders, and Sanders continued to enter them anyway. See generally Kaster Decl. Exs. 13, 14, 16, 18. A jury could further find that Sanders made those implicit refusals “in good faith” and that they were therefore protected under § 20109(a)(2).

Third, Sanders argues that when he contacted Human Resources to report his concerns about the pressure he had received to ease up on his use of slow orders and other defect reports, that activity was protected under § 20109(a)(1), discussed above. BNSF does not dispute this. 2

Establishing a prima facie case also requires Sanders to show that his protected conduct was a contributing factor in his termination, the only adverse action he challenges here. Kuduk, 768 F.3d at 789. A contributing factor is “any factor which, alone or in connection with other factors, tends to affect in any way the outcome of the decision.” Hess v. Union Pac. R.R. Co., 898 F.3d 852, 857–58 (8th Cir. 2018) (citing Kuduk, 768 F.3d

at 791). The protected conduct need not be the only reason for the adverse action: “two causes being non-mutually exclusive is the very essence and definition of a ‘contributing’ factor.” Blackorby v. BNSF Ry. Co., 936 F.3d 733, 737 (8th Cir. 2019), reh’g denied (Oct. 11, 2019). Thus, even if BNSF honestly believed that Sanders had engaged in misconduct or violated company rules, BNSF nevertheless could be held liable under the FRSA if a retaliatory motive played a part in its decision to terminate Sanders. Id. The contributing-

factor standard is a “lenient” one, but it does require employees to show that their protected activity was a proximate cause of the adverse action. BNSF Ry. Co. v. U.S. Dep’t of Labor, 867 F.3d 942, 946 n.2 (8th Cir. 2017). The employee “must demonstrate more than a mere factual connection between his injury report and his discipline,” but need not “conclusively demonstrate [the company’s] retaliatory motive to establish a prima facie case.” Heim v. BNSF Ry. Co., 849 F.3d 723, 727 (8th Cir. 2017) (internal quotation marks omitted).

“Absent sufficient evidence of intentional retaliation, a showing that protected activity initiated a series of events leading to an adverse action does not satisfy the FRSA’s contributing factor causation standard.” BNSF v. U.S. Dep’t of Labor, 867 F.3d at 946. A plaintiff may satisfy this element through direct or circumstantial evidence. Hess, 898 F.3d at 858. Sanders says he has both.

Sanders does not have direct evidence. Direct evidence of discrimination entails “a specific link between the alleged discriminatory animus and the challenged decision, sufficient to support a finding by a reasonable fact finder that an illegitimate criterion actually motivated the adverse employment action.” Russell v. City of Kansas City, 414 F.3d 863, 866 (8th Cir. 2005) (citation omitted). It includes “comments or statements

indicating [retaliatory] intent, where those comments are made by people with decision- making authority.” Hutton v. Maynard, 812 F.3d 679, 683 (8th Cir. 2016) (citation omitted) (employment discrimination case). In this context, “‘[d]irect’ refers to the causal strength of the proof, not whether it is ‘circumstantial’ evidence.” Griffith v. City of Des Moines, 387 F.3d 733, 736 (8th Cir. 2004).

Sanders points to two direct-evidence cases he says are analogous to this one. Neither is. First, he cites Young-Losee v. Graphic Packaging International, Inc., in which the manager wadded up the plaintiff’s complaint, called it “total bullshit,” threw it in the garbage, and told the employee to leave and that he never wanted to see her again. 631 F.3d 909, 912 (8th Cir. 2011); see Mem. in Opp’n at 31–33. But the link between any of the protected activity Sanders alleges is much more attenuated than in Young-Losee. He was

not fired by Jones in the same conversation in which Jones learned of his protected activity, for example. The protected activity and the adverse action were more distant in terms of the lapse of time, the steps in the internal process, and the identity and number of the decision makers. The second case Sanders relies on, King v. Hardesty, involved a statement demonstrating racial bias (in which a white school administrator told an African

American teacher, “white people teach black kids . . . better than someone from their own race”) from which a very short line could be drawn to the administrator’s subsequent decision to end the African American teacher’s employment working one-on-one with a particular student and reassign that student to a white teacher. 517 F.3d 1049, 1058–59 (8th Cir. 2008); see Mem. in Opp’n at 31–33. Whatever a fact-finder might conclude about

the propriety of Jones’s conduct and motives here, his statements to Sanders are linked far less directly, if at all, to the decision to terminate his employment. It is therefore not legally accurate to characterize Sanders’s evidence of contributory causation as “direct.”

Sanders argues alternatively that his claim should go to a jury on the basis of circumstantial evidence. Circumstantial evidence includes things such as “the temporal proximity between the protected activity and the adverse action, indications of pretext such as inconsistent application of policies and shifting explanations, antagonism or hostility toward protected activity, the relation between the discipline and the protected activity, and the presence of intervening events that independently justify discharge.” Hess, 898 F.3d

at 858 (quoting Loos v. BNSF Ry. Co., 865 F.3d 1106, 1112–13 (8th Cir. 2017)). Here, Sanders has presented circumstantial evidence that would allow a jury to conclude that at least some of his protected activity contributed to BNSF’s decision to fire him. Jones’s hostility reasonably may be characterized as discriminatory animus. Sanders recorded numerous conversations with Jones lasting between about 11 and

23 minutes. A significant part of each conversation reflects Jones’s antagonism toward Sanders’s findings in his track inspections and Jones’s efforts to discourage Sanders from continuing to unilaterally enter so many slow orders, even if they were appropriate in light existing track conditions. See generally Kaster Decl. Exs. 13, 14, 16, and 18. Notably, in those conversations Jones seems to accept Sanders’s stated reasons for entering various

slow orders, but nevertheless expressed hostility because of how Sanders’s reports made Jones look within the company. E.g., Kaster Decl. Ex. 18 at 1:53 (“I’m not saying you’re not doing your job, but as a whole, it is embarrassing for everybody.”). Jones responded to Sanders’s activities with a pattern of open antagonism and angry outbursts. Jones’s remarks were not the type of “stray comment[s] unrelated to the decisional process” that may be disregarded in assessing evidence of animus. E.g., Saulsberry v. St. Mary’s Univ.,

318 F.3d 862, 867–68 (8th Cir. 2003) (explaining that no reasonable jury could have inferred discriminatory intent from a few isolated one- or two-word remarks made by people without a decision-making role or by an arguable decisionmaker in a context “unrelated to the decisional process”); cf. Quick v. Wal-Mart Stores, Inc., 441 F.3d 606, 609 (8th Cir. 2006) (two brief comments, one made five months before employee’s termination, did not constitute direct evidence of discriminatory motive under McDonnell

Douglas framework).14 A reasonable jury also could conclude that Jones was involved in the decision to charge Sanders with time theft and the subsequent decision to dismiss him. Although Detlefsen testified she independently concluded that dismissal was the appropriate sanction and that she did not rely on Jones’s interpretation “at all” in that regard, Detlefsen Dep. at

41–42, a jury could decide her testimony is not credible on that point. A jury might instead conclude that Detlefsen considered Jones’s relatively detailed summary of evidence and the rationale he provided for his dismissal recommendation. Jones was the division

14 Sanders has not identified evidence that Jones or Hoppenrath displayed hostility toward Sanders based specifically on his complaints to BNSF human resources. But given the undeniable relationship between those reports, in which Sanders repeatedly complained about the pressure he was getting to ease up his safety reports (e.g., when Sanders reported in April 2016 that “Keith [Jones] wants Donald to break policy and do things his way” and that “Keith’s deviation from policy could cause safety issues,” Duginske Decl. Ex. KK at 2), and the other protected activities towards which Jones expressed hostility, a jury reasonably could infer that Jones’s antagonism toward Sanders’s refusals to stop or reduce his safety reports extended to Sanders’s human-resources complaints that his supervisors were pressuring him to stop or reduce his safety reports. engineer and presumably had some knowledge about and experience with how BNSF reviews termination decisions. Rather than simply sending Detlefsen the record that was

developed in the investigative-hearing process so she could review it with fresh eyes, Jones sent a relatively detailed summary of the evidence he believed was most relevant for each charge, as well as his own assessments of Sanders’s credibility and an explanation of his own dismissal recommendations. Duginske Decl. Exs. TT, UU, and VV. A jury might reasonably infer that an experienced division engineer would have no reason to provide such detailed summaries, credibility assessments, and discipline recommendations if they

were not going to be considered in the PEPA review process. Detlefsen testified that she could not remember a single instance in which the larger PEPA review board had disagreed with her recommendation, see Detlefsen Dep. at 69, which might reasonably lead a jury to conclude that review of the decision by the board was not independent of her recommendation and, therefore, not independent of Jones’s.

Because the evidence would permit a reasonable jury to infer that Jones was involved in the decision to terminate Sanders, this case presents a different situation than the Eighth Circuit faced in Gunderson v. BNSF Ry. Co. See 850 F.3d 962 (8th Cir. 2017). In Gunderson, the Eighth Circuit explained that even where a railroad-employee plaintiff’s immediate supervisors had expressed hostility toward the employee’s protected activity,

the employee nevertheless could not satisfy the contributory-factor element of an FRSA retaliation claim because there was no evidence that the immediate supervisors did anything more than forward a complaint they had received about the employee to their general manager, and that it was the general manager—against whom no evidence of hostility or antagonism existed—who initiated the investigation and decided, in consultation with his managers and human resources, to dismiss the employee. Id. at 970.

Because here, a jury could find that Jones was both hostile toward Sanders’s protected activity and directly involved in the decisions to charge and ultimately dismiss him, Gunderson permits a finding that Sanders’s protected activity was a contributing factor in BNSF’s termination decision. Sanders has presented evidence of pretext from which a jury could infer that alleged time theft was not the real reason he was terminated. Sanders points to comparators J.K.

and B.K., both of whom received lesser discipline after facing similar accusations of dishonesty in how they reported their time. Neither J.K. nor B.K. had refused to violate any safety-related law, reported defects or slow orders against the wishes of their supervisors, assisted in any safety-related investigation, or filed complaints against their supervisors related to safety concerns. Based on this evidence, a jury could conclude that

Sanders had been disciplined more harshly than B.K. and J.K. at least in part because Sanders, unlike B.K. and J.K., had engaged in conduct protected under the FRSA. Additionally, Sanders’s co-worker testified that it was common for BNSF employees to enter placeholder time entries and to correct those entries later, and that BNSF typically did not accuse other employees who did that of falsifying their records. See, e.g., Gaylor

Decl. ¶¶ 9–16. BNSF identifies other employees who were terminated after being found to falsify their time records, but that evidence does not preclude a finding that Sanders’s firing was pretextual. BNSF has cited no evidence of whether any of those other fired individuals had, like Sanders, engaged in protected activities, and they therefore reveal nothing about whether BNSF disciplines employees engaged in protected activity more harshly than those

not engaged in protected activity. The other individuals were not fired for (or not entirely for) time theft, but for other dishonesty-related offenses like taking unapproved absences or misusing a company credit card, and unlike B.K. and J.K., it appears that at least some (and perhaps all) were supervised by someone other than Jones and Hoppenrath. E.g., Duginske Decl. Ex. CCC (identifying various allegations of dishonesty giving rise to termination and indicating that disciplined employees worked in varying locations,

including Iowa and South Dakota); cf. Dafoe v. BNSF Ry. Co., 164 F. Supp. 3d 1101, 1109– 10 (D. Minn. 2016) (explaining that a plaintiff may show pretext by pointing to a comparator employee who is “‘similarly situated in all relevant respects,’ including that he or she ‘dealt with the same supervisor, had been subject to the same standards, and engaged in the same conduct without any mitigating or distinguishing circumstances’” (quoting

Bone v. G4S Youth Servs., LLC, 686 F.3d 948, 956 (8th Cir. 2012) (alteration omitted)). In contrast to these differences, BNSF has provided no explanation why B.K. and J.K. are not appropriate comparators for Sanders here. Although BNSF asserts that the charges against B.K. and J.K. may “sound[ ] similar” but “are not identical” to those against Sanders, Reply at 8, it provides no further explanation, and none is obvious. It is unquestionably true that

a jury could reasonably conclude that the intervening events discovered through Hoppenrath’s surveillance and the investigative hearings undercut any conclusion that Sanders’s protected activity was a contributing factor to his dismissal. But a jury could also review the same evidence and conclude the opposite: that the evidence of pretext discussed above is strong enough that the intervening events of Hoppenrath’s surveillance and the company’s investigative proceedings do not constitute a persuasive, credible

explanation for why Sanders was fired. To be sure, some types of circumstantial evidence do not support a finding of contributory causation. Sanders’s protected activity and his April 29 termination were not particularly close in time. The last protected activity described in the Parties’ briefs appears to have occurred on April 5, when Sanders made his final internal HR complaint regarding Jones and Hoppenrath’s alleged ongoing harassment, but most of Sanders’s protected

activity occurred prior to Sanders’s transfer back to Dayton’s Bluff in mid-March 2016. Duginske Decl. ¶ 37 and Ex. KK. This temporal gap is too attenuated to give rise to an inference that any of Sanders’s protected activity was a contributing factor in his termination. See Smith v. Allen Health Sys., Inc., 302 F.3d 827, 836 (8th Cir. 2002) (gap of several months between protected activity and adverse action does not suggest causal

connection between the two). To the extent Hoppenrath’s decision to surveil Sanders was sparked by his protected activity, that decision standing alone does not support a finding that his protected activity was a contributing factor in the resulting discipline. See Mem. in Opp’n at 36. In BNSF v. U.S. Dep’t of Labor, the Eighth Circuit held that “a showing that protected activity initiated

a series of events leading to an adverse action does not satisfy the FRSA’s contributing factor causation standard.” 867 F.3d at 946. In other words, the mere fact that an employee’s protected activity caused his employer to more closely scrutinize the employee’s record in the hopes of finding disciplinary violations, without any more evidence, is not enough to demonstrate that the protected activity was a contributing factor in whatever disciplinary outcome results. Id. at 946–47. Furthermore, even if Sanders is

correct that he did not falsify his time records—even if BNSF’s hearing process simply got it wrong—that fact on its own does not establish pretext if BNSF possessed a good-faith belief that he committed the time theft with which he was charged. “[F]ederal courts do not sit as a super-personnel department that re-examines an employer’s disciplinary decisions.” Id. at 792 (internal quotation mark omitted); see also Foster v. BNSF Ry. Co., 866 F.3d 962, 969 (8th Cir. 2017) (recognizing that erroneous company findings do not

show retaliation, absent any evidence that the company issued the findings as an act of retaliation). In short, BNSF acting in good faith but getting its investigation wrong wouldn’t be enough to establish a prima facie case, but BNSF reaching the decision it did at least in part to retaliate against Sanders for his protected activity would. Sanders has produced circumstantial evidence from which a jury could reasonably conclude that such

a retaliatory motive existed. As for Sanders’s argument that every single one of the thousands of track defects he reported, and every single one of the hundreds of times he pulled tracks out of service or entered a slow order, see Sanders Dep. at 45, 48–50, 53–54, contributed in some way over a period of many years to BNSF’s April 2016 decision to fire him, no reasonable jury

could so conclude. Baked into Sanders’s position is the undisputed fact that he had been engaging in this form of protected activity for a very long time, and in a very large volume. But none of those prior actions ever led to adverse action being taken against him. See Brisbois v. Soo Line R.R. Co., No. 15-CV-0570 (PJS/TNL), 2016 WL 7423387, at *6 (D. Minn. Dec. 22, 2016) (recognizing that a history of reporting safety violations without suffering adverse actions suggests lack of discriminatory animus toward safety

complaints). Sanders does not point to any evidence suggesting why BNSF would take the seemingly illogical step of suddenly decided to retaliate against him for doing something it knew he had been doing for a long time. Though a jury could conclude based on Sanders’s past instances of discipline that his dismissal in April 2016 was not retaliatory, none of the cases BNSF cites on that point establishes that a history of discipline “preclude[s]” an inference of retaliation where other

circumstantial evidence of retaliatory motive exists. See Mem. in Supp. at 29. To the contrary, in the three cases BNSF cites, the court discussed the plaintiff’s history of discipline in connection with the absence of any other evidence of shifting or false explanations of pretext. See Kuduk, 768 F.3d at 792 (where “the employer was concerned about a problem before the employee engaged in the protected activity,” more than mere

temporal connection is required to present genuine issue on retaliation (internal quotation omitted)); Guimaraes v. SuperValu, Inc., 674 F.3d 962, 978 (8th Cir. 2012) (same); Loos v. BNSF Ry. Co., No. 13-cv-3373 (PAM/FLN), 2015 WL 3970169, at *6 (D. Minn. June 30, 2015) (granting summary judgment to railroad employer where plaintiff had no evidence of contributing factor and, to the contrary, plaintiff’s discipline history suggested

that the decision to fire him was the result of repeated violations rather than pretextual), rev’d on other grounds, 139 S. Ct. 893 (2019). 3 Once a plaintiff establishes a prima facie case under the FRSA, the employer avoids

liability if it can establish by clear and convincing evidence that it would have taken the same action even in the absence of the protected conduct. Kuduk, 768 F.3d at 789; Blackorby, 936 F.3d at 737. The clear-and-convincing standard is an intermediate one, located somewhere between a preponderance-of-the-evidence standard and a beyond-a- reasonable-doubt standard in terms of the hurdle it presents. Addington v. Texas, 441 U.S. 418, 425 (1979). To satisfy the clear-and-convincing standard, the party with the burden—

here, BNSF—must show that “the truth of its factual contentions [is] highly probable.” Colorado v. New Mexico, 467 U.S. 310, 316 (1984). Put another way, whatever proof BNSF offers to show that it would have taken the same action even in the absence of Sanders’s protected conduct must “instantly tilt[ ] the evidentiary scales” in its favor when compared with the evidence Sanders offers. Id.

Sanders argues that the more lenient discipline BNSF imposed on comparators B.K. and J.K. for similar violations prevents BNSF from making the showing required to obtain summary judgment on the basis of this affirmative defense. Evidence that similarly situated employees were treated differently can “do[ ] double duty for a plaintiff: It tends to show a causal link between the protected conduct and the adverse action, and it also

tends to show that the justification given by a defendant was pretextual.” Hudson v. Norris, 227 F.3d 1047, 1052 (8th Cir. 2000). As discussed above, BNSF argues that B.K. and J.K. are not similarly situated to Sanders but does not explain why, and on the record presented by this motion, the correct conclusion is that they are. Reply at 8. For the reasons described above, a reasonable jury could find that the more lenient discipline those comparators experienced, and the fact that, according to testimony from Sanders’s coworkers, Sanders

was disciplined for conduct that other employees routinely engage in, BNSF’s stated justification for firing Sanders was pretextual. See Head v. Horfolk S. Ry. Co., No. 2:15- cv-02118-RDP, 2017 WL 4030580, at *17–18 (N.D. Ala. Sept. 13, 2017) (material fact issue raised by comparator evidence precludes summary judgment on affirmative defense that employer would have taken adverse action even in the absence of protected activity). Furthermore, also for the reasons discussed above, a genuine issue of material fact exists

as to whether the review by Detlefsen and the PEPA team was sufficiently independent of Jones’s influence in the discipline process to establish that BNSF would have dismissed Sanders even if he had not engaged in the protected conduct. C BNSF argues that, if the Court does not grant summary judgment on the Complaint

in its entirety, it should at least award BNSF summary judgment as to Sanders’s ability to seek punitive damages. Mem. in Supp. at 38–39; see also Compl. at Prayer for Relief ¶ 3 [ECF No. 1] (seeking punitive damages under the FRSA). The FRSA does not articulate the standard that governs punitive damages under that statute, but courts have awarded punitive damages where the employer has acted with

reckless or callous disregard for the plaintiff’s rights or intentionally violated federal law. BNSF Ry. Co. v. U.S. Dep’t of Labor, 816 F.3d 628, 642 (10th Cir. 2016). Applying this standard in the Title VII context, the Supreme Court has held that employers may be liable for punitive damages based on the discriminatory conduct of lower-level management, but that punitive damages will not be awarded where “the discriminatory employment decisions of managerial agents . . . are contrary to the employer’s good faith efforts to

comply” with the law. Kolstad v. Am. Dental Ass’n, 527 U.S. 526, 545 (1999). A defendant that makes good-faith efforts to comply with the FRSA is not liable for punitive damages. BNSF v. U.S. DOL, 867 F.3d at 949. An employer that adopts policies and rules that explicitly prohibit retaliation, maintains a hotline or website for reporting concerns, and follows an internal review process in which the person ultimately responsible for reviewing the evidence and making a decision had no prior knowledge of the plaintiff or

his protected activity has “strong evidence of [its] good-faith efforts to prevent retaliation.” Id. Sanders’s punitive-damages claim presents a close call on summary judgment. For the reasons described above, a jury reasonably could conclude that Jones acted with reckless or callous disregard for Sanders’s rights or intentionally violated federal law by

repeatedly pressuring Sanders to refrain from entering legitimate slow orders and retaliating by having Sanders fired when he would not do so. As for whether BNSF may be liable for punitive damages as a result, it is undisputed that BNSF has many of the protections the Eighth Circuit has described as “strong evidence” showing that it was trying in good faith to prevent retaliation. At the same time, one of the key aspects of this

determination is whether BNSF did not merely have but followed an internal review process in which the person ultimately responsible for the decision was disconnected from the assertedly bad actor. Here, there are genuine issues of material fact about whether Detlefsen and the PEPA team were independent of Jones. In other cases, this issue has prevented the entry of summary judgment for BNSF on the punitive damages question, Fresquez v. BNSF Ry. Co., No. 17-cv-00844-WYD-SKC, 2018 WL 6249686, at *9 (D.

Colo. Oct. 2, 2018) (fact question regarding whether BNSF PEPA review process represented good-faith efforts to enforce corporate anti-retaliation policies precluded summary judgment on punitive damages); Cf. Smith v. BNSF Ry. Co., No. 17-cv-00977- KMT, 2019 WL 3230975, *8 (D. Colo. July 18, 2019) (fact issue regarding whether company imposed termination for the purpose of retaliation precluded summary judgment on punitive damages based on company’s good faith), and it will here also.

ORDER Based on the foregoing, and on all of the files, records, and proceedings herein, IT IS ORDERED THAT Defendant BNSF’s motion for summary judgment or, alternatively, for partial summary judgment [ECF No. 65] is DENIED.

Dated: October 24, 2019 s/Eric C. Tostrud Eric C. Tostrud United States District Court