Reset A A Font size: Print

United States Court of Appeals,Eighth Circuit.

Scott HAUKEREID, Individually and as Personal Representative and Administrator of the Estate of Andrew Haukereid Jr, deceased, Plaintiff–Appellant v. NATIONAL RAILROAD PASSENGER CORPORATION, doing business as Amtrak, Defendant–Appellee.

No. 14–3752.

Decided: March 08, 2016

Before MURPHY, BENTON, and KELLY, Circuit Judges. Carter Stein, argued, Little Rock, AR, for Plaintiff–Appellant. Scott H. Tucker, argued, Little Rock, AR, (Martin A. Kasten, Kristopher B. Knox, Mary K. McCarroll, Little Rock, AR, on the brief) for Defendant–Appellee.

Scott Haukereid brought this action on behalf of himself and his father's estate, alleging that the National Railroad Passenger Corporation (Amtrak) negligently caused the wrongful death of his father Andrew Haukereid because the moving train from which he fell had inadequate safety features and instructions for the crew. The district court1 granted summary judgment to Amtrak, and Haukereid appeals. We affirm.


Andrew Haukereid boarded one of Amtrak's Superliner II trains on May 19, 2012 in San Antonio, Texas destined for Chicago, Illinois to visit his sons. He was 79 years old and without any diagnosed mental, cognitive, or psychiatric conditions.

A Superliner II train car has two manual side exit doors which are adjacent to the restrooms, and each exit door has a window in its top half. The interior of each exit door is marked with a sign warning passengers not to open the door or window. To open a side exit door on this train, a person must: (1) lift a safety latch in the upper right corner of the door to an upright position; (2) push down the door handle; and (3) pull the door inward. The door windows are also secured by a latch. To open a window, an individual must lift the latch and pull the window inward. The windows and side doors serve as emergency exits and are left unlocked even while the train is moving.

While the train was underway Andrew spoke on the phone several times with his son Scott and occasionally asked crew members where the train was. At 3:39 pm Andrew called Scott and said he was on an Amtrak train bound for Chicago. About three hours later, he asked assistant conductor Louis Porch where the train was then. Around 8:55 pm, a few minutes after the crew had announced that the train had arrived in Texarkana, Andrew asked Porch if the train was near Chicago. Porch told him that it was still a long way to Chicago. Crew member Singrid Jackson reported that when she had escorted a man fitting Andrew's description from the dining car to the coach section, she observed that he had a confused look as if he did not know where he was.

At 10:17 pm Scott called Andrew again, and Andrew told him that he did not think the train went to Chicago. Scott asked him to confirm the train's destination and call him back. Andrew then walked to the dining car where Porch and conductor John Davis were seated and told them, “I don't know where I am.” Davis and Porch replied that the train was in Hope, Arkansas and asked Andrew where he was headed. When Andrew hesitated to answer, Davis checked his ticket and told him that he had plenty of time for a good nap before they reached Chicago. At 10:33 pm Andrew called Scott and told him that he had spoken with an Amtrak employee and confirmed the train was on its way to Chicago. When the train arrived in Little Rock, Arkansas at 12:13 am on May 20, Andrew asked Porch “is this Chicago?” The conductor responded that they were still twelve hours away from there.

The last person who saw Andrew alive was a fellow passenger, Bobbie Thomas. She testified that she noticed Andrew in the vestibule of his coach car sometime between 2:30 and 3:00 am on May 20, the date he disappeared. According to Thomas, Andrew was facing the right side door and the window on the left exit door was open behind him. As Thomas went in and out of the restroom, she had to walk around Andrew who did not acknowledge her. She described his behavior to be “odd,” as if “his mind was someplace else, not there with him.” Andrew never did arrive in Chicago. His body was found several weeks later about 20 feet east of the railroad tracks in Clay County, Arkansas. No one witnessed him leave the train or observed that the train doors were not properly functioning. Nor did any crew member find any doors or windows open or unsecured.

Since Amtrak was established on May 1, 1971, at least 76 passengers have exited moving trains between stops. Of these 76 incidents, at least 29 of the missing passengers appeared to have been confused or disoriented. One of Haukereid's experts, George A. Gavalla, opined that if Amtrak had informed its employees about the number of accidents involving elderly passengers falling from its trains and distributed information to its crew members on how to recognize signs of cognitive impairment, Amtrak could have prevented Andrew from falling to his death. Gavalla also testified that it is more probable than not that Andrew's accident would have been prevented had Amtrak installed door status indicators to alert crew and passengers whenever a dangerous exit door was opened, or had Amtrak installed an interlocking system on its Superliner passenger cars to prevent exit doors from being opened while a train was moving.

While Andrew was still missing, Amtrak completed a missing person investigation. Amtrak employees spoke with his family members and noted in a preliminary investigation report that his family had been “worried about him since he suffer[ed] from possible dementia although he had not been medically diagnosed.” Nevertheless, Andrew's sons and siblings testified that he had not had any cognitive or mental impairments. His sons reported that they were unaware of any prior instances of their father getting lost or becoming disoriented. Andrew's domestic companion Carole Moreno told an Amtrak detective that he was starting to become forgetful although he was generally alert and lucid. She feared Andrew was developing memory loss and possible dementia. Amtrak employees who completed a company police department missing person report for the incident listed Andrew's “reason for leaving the train” as “Possible Dementia.”

During its investigation Amtrak also interviewed its employees. On May 24 conductor Davis reported that because Andrew had repeatedly asked where he was, Davis had the impression that “[s]omething just wasn't quite right with [Andrew].” Several days later, Davis modified his statement to explain that he had not interacted with Andrew long enough to know if there was anything wrong with him. According to a neurological examination of Andrew on May 17, 2011, no mental, cognitive, or psychiatric problem was found. Less than a month before Andrew boarded the train, an urgent care doctor who examined Andrew for jaw pain on April 20, 2012 had observed similarly that Andrew's judgment and mood appeared normal and that he had been “oriented to time, place and person.”

On April 2, 2013 Scott Haukereid brought this action against Amtrak, alleging that it had negligently caused his father's death. The district court concluded that although a reasonable jury could find that Amtrak breached its duty “by not passing along information between crews about [Andrew's] confusion, not being more vigilant about his circumstances, and not providing more training to crew members about the exit dangers faced by older travelers, especially at night,” a jury would have had to speculate as to how Andrew exited the train and as to the proximate cause of his death. The district court granted summary judgment for Amtrak, and Haukereid appeals.


We review a “district court's grant of summary judgment de novo, viewing the facts in the light most favorable to the nonmoving party and giving that party the benefit of all reasonable inferences that can be drawn from the record.” Minn. ex rel. N. Pac. Ctr., Inc. v. BNSF Ry. Co., 686 F.3d 567, 571 (8th Cir.2012). Summary judgment is appropriate if no genuine dispute exists “as to any material fact and the movant is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(a). We “may affirm the judgment on any basis supported by the record.” Hohn v. BNSF Ry. Co., 707 F.3d 995, 1000 (8th Cir.2013).

Scott Haukereid claims negligence and therefore bears the burden of showing by a preponderance of the evidence that Andrew's injuries were caused by some negligent act or omission of Amtrak. See Mangrum v. Pigue, 359 Ark. 373, 198 S.W.3d 496, 503 (Ark.2004). The parties disagree as to whether the record contains sufficient evidence for a jury to find that Amtrak proximately caused Andrew's death. Under Arkansas law, proximate cause is defined “as that which in a natural and continuous sequence, unbroken by any efficient intervening cause, produces the injury, and without which the result would not have occurred.” Neal v. Sparks Reg'l Med. Ctr., 2012 Ark. 328, 422 S.W.3d 116, 120–21 (Ark.2012). Moreover, a jury should “not be left to speculation and conjecture in deciding between two equally probable possibilities.” St. Louis Sw. Ry. Co. v. Pennington, 261 Ark. 650, 553 S.W.2d 436, 441–42 (Ark.1977). A party also may not “draw one inference from another, or ․ indulge presumption upon presumption to establish a fact.” Glidewell v. Arkhola Sand & Gravel Co., 212 Ark. 838, 208 S.W.2d 4, 8 (Ark.1948) (internal quotation marks omitted).

When the facts are viewed in the light most favorable to Haukereid, the circumstantial evidence in the record supports the inference that Andrew accidentally fell out of a right side exit door on the train which he or someone else had opened. His fall was likely accidental because the record evidence does not suggest either that he intended to commit suicide or that a third party pushed him off the train. Moreover, his body was found on the right side of the train, and he was last seen standing near an exit door. His height made it unlikely that he fell through an exit door window which would have required him first to step up on something like a stool, and the crew did not find anything like that near the windows.

While there is evidence in the record which could support Haukereid's theory that Andrew accidentally fell out of the train's side exit door, the record does not support his contention that Amtrak proximately caused Andrew's death. Haukereid contends that Amtrak failed to equip the exit doors with door status indicators or an interlocking door system. We dealt with a similar fact pattern in Donnelly v. Nat'l R.R. Passenger Corp. (Amtrak ), where no one witnessed a passenger exit an Amtrak train and fall to her death. See 16 F.3d 941, 943 (8th Cir.1994). We explained in Donnelly that Amtrak's “failure to lock [or better secure] the door could not have caused [the passenger] to exit the train” because, like the exit doors on Andrew's train, in order to open an exit door required lifting its latch, then operating the door handle and pulling the door open. Id. at 945. Thus, “failure to lock the door could not have opened the door.” Id. Anyone lifting the latch on the exit door would have served as an “efficient intervening cause” in the process. See Neal, 422 S.W.3d at 121. Haukereid attempts to distinguish the intervening cause argument in Donnelly by arguing that such a cause must be intentional. Since Andrew was mentally infirm, he should not be held accountable for opening the door. Haukereid's argument fails for two reasons. First, a jury may not “indulge presumption upon presumption to establish a fact” as would be required here since the parties dispute whether Andrew was cognitively impaired at the time he was on the train and there was no evidence showing who may have opened the exit door from which Andrew apparently fell. See Glidewell, 208 S.W.2d at 8. Second, the proximate cause analysis in Donnelly was not limited to the mental state of whoever lifted the door latch.

We conclude that Amtrak's failure to install door status indicators to alert crew that an exit door was open was not the cause of Andrew's death because they would not have prevented his falling. See Donnelly, 16 F.3d at 945. Haukereid's train safety expert Gavalla testified that door status indicators could have only given crew brief notice that a door was open and would not necessarily have prevented Andrew from exiting the train. Thus, even assuming that Amtrak had a duty to secure the exit doors better, Haukereid did not show as a matter of law that the lack of additional door safety features proximately caused Andrew to exit the train.

Haukereid also has failed to present a genuine issue of fact showing that Amtrak proximately caused Andrew's death by failing to instruct its crew adequately on hazards for confused passengers. Under Arkansas law, a common carrier is “required to exercise towards its passengers the highest degree of care which a prudent and cautious man would exercise and that which is reasonably consistent with the mode of conveyance and practical operation of its trains.” Capital Transp. Co. v. Howard, 217 Ark. 333, 229 S.W.2d 998, 1000 (Ark.1950). Although a common carrier owes a heightened duty of care to its passengers, the record does not demonstrate that Amtrak would have or should have known that Haukereid was unfit to travel alone. Amtrak crew members interacted with Andrew on five occasions, once leading him back to his seat and four times informing him where the train was and where it was headed. Even viewing the evidence in the light most favorable to Haukereid and based on these limited interactions with Andrew, Amtrak crew could not have understood that he was confused or somehow cognitively impaired. In fact, Haukereid's own expert testified that Andrew's behavior was not concerning enough to cause Amtrak crew to remove him from the train as they might when a passenger's behavior is endangering him. Another expert offered by Haukereid admitted that Andrew's interaction with conductor Davis would not have put Amtrak on notice that he was at risk of exiting the train. Moreover, assistant conductor Porch testified that many passengers who do not appear mentally impaired ask “where we're going or where we are all the time.” Assigning Andrew a personal monitor or hiring more crew on overnight trains would exceed Amtrak's legal duty as a common carrier since it “is not an absolute insurer of the safety of its passengers” and adding employees to watch a passenger not diagnosed with a cognitive impairment would not be “reasonably consistent with the mode of conveyance and practical operation of [Amtrak's] trains.” See id.

Even if Amtrak had better trained its crew members about the exit dangers faced by older passengers at night, Haukereid did not raise a disputed issue of fact as to whether such training could have prevented Andrew from exiting the train. Haukereid claims that with adequate training Amtrak could have prevented Andrew's death because he could have received a “Keep in Sight” check card on his seat, requiring crew to monitor him regularly. According to assistant system general trainmaster Michael Bonner, a seat check card does not obligate crew members to restrain or keep track of a passenger. Instead, the card serves as a visible reminder that the person in the seat may need assistance at some point during the trip. Such a card would not have prevented Andrew from exiting the train by himself. Moreover, Amtrak cannot force a passenger to accept an accommodation or special service that he or she has not requested. See Americans with Disabilities Act, 49 C.F.R. § 39.21(a).

We conclude from this record that the district court did not err by granting Amtrak summary judgment on Haukereid's negligence claim.


Haukereid also appeals two of the district court's discovery rulings which are reviewed for gross abuse of discretion. See Level 3 Commc'n, LLC v. City of St. Louis, Mo., 540 F.3d 794, 796 (8th Cir.2008). Even if “a party can demonstrate a gross abuse of discretion by the trial court ․ the complaining party must also demonstrate prejudice.” Gov't of Ghana v. ProEnergy Serv., LLC, 677 F.3d 340, 345 (8th Cir.2012) (internal quotation marks omitted).

Haukereid claims that the district court violated Federal Rule of Evidence 26 by denying his motions which would have compelled Amtrak to produce 11 investigation reports about passengers who had allegedly exited trains through windows of exit doors and to produce a Rule 30(b)(6) witness to testify about “prior incidents involving passengers exiting moving trains operated by Amtrak.” Haukereid contends that these reports and the testimony he sought are relevant to prove that Andrew did not fall out of a window. Under Fed.R.Civ.P. 26(b)(1), a party “may obtain discovery regarding nonprivileged matter that is relevant to [its] claim or defense and proportional to the needs of the case.” Here, the district court did not grossly abuse its discretion by denying Haukereid's discovery requests because the record already contained evidence about the unlikelihood of a window exit. See Fed.R.Evid. 403. Moreover, Haukereid has not shown that if his discovery requests had been granted, the evidence would have affected the issue of proximate cause. See Gov't of Ghana, 677 F.3d at 345.


For all these reasons the judgment of the district court is affirmed.

This case asks us to make a very close call. I agree with the court that door status indicators or further instruction of the Amtrak crew would not have prevented Andrew Haukereid's death. Therefore, Haukereid's case rests on proving a breach of Amtrak's duty to better secure the exit doors, and a showing that the lack of additional door safety features proximately caused Andrew to exit the train.

The court holds that, as in Donnelly, the “failure to lock the door could not have opened the door,” and that “[a]nyone lifting the latch on the exit door would have served as an ‘efficient intervening cause’ in the process.” 16 F.3d at 945. The court rejects Haukereid's attempt to distinguish Donnelly first because it believes that a finding that Andrew was cognitively impaired along with the lack of evidence of who opened the door requires the jury to “indulge presumption upon presumption” to find that the opening of the door was unintentional. In my view, however, this case is distinguishable from Donnelly on its facts. There was no evidence in Donnelly that the decedent was confused, and no evidence to narrow down the possible ways of exiting the train. Donnelly, 16 F.3d at 945. In fact, we stated there that “[t]he circumstances surrounding [the decedent's] exit from the train are wholly unknown to the parties,” and that it was “equally probable that [the decedent] voluntarily exited the train or that someone pushed [the decedent] out of the train.” Id. Other instances where Amtrak has avoided liability are similarly sparse on facts. See, e.g., Harris v. Nat'l Pass. R.R. Corp., 79 F.Supp.2d 673 (E.D.Tex.1999).

Here, by contrast, the court has established that suicide, homicide, and a window exit were not equally probable scenarios. I agree that “there is evidence in the record which could support Haukereid's theory that Andrew accidentally fell out of the train's side exit door.” Ante at 6. Moreover, there is also evidence that Haukereid was cognitively impaired, not just a presumption. Haukereid's interactions with the crew—repeatedly asking them where he was in a manner that showed he was not fully aware of place or the passage of time—indicate confusion. There is evidence in the record to show that he got up to use the restroom each night around the time that he was seen in the train hallway. Passenger Bobbie Thomas described Andrew's behavior as odd and felt his mind was elsewhere. Andrew's domestic companion also feared that Andrew was developing memory loss and possible dementia.

The court recognizes that the parties dispute whether Andrew was cognitively impaired. But the evidence given that he was not impaired—that he had not been diagnosed with dementia or Alzheimer's and that an urgent care doctor had checked a box indicating normal mood while examining Andrew for jaw pain—is hardly conclusive evidence that Andrew did not suffer from some mental deficits. While the jury must “not be left to speculation and conjecture in deciding between two equally probable possibilities ․ it is not necessary that the evidence exclude every other reasonable hypothesis. It is only necessary that there be evidence that would tend to eliminate such other causes as may fairly arise from the evidence,” and reasonable inferences may be drawn from circumstantial evidence. St. Louis Southwestern Ry. Co. v. Pennington, 261 Ark. 650, 553 S.W.2d 436, 441–42 (Ark.1977) (in banc). Taking these facts in the light most favorable to Andrew, I believe there are enough facts to prevent the jury from having to “indulge presumption upon presumption to establish” that Andrew was impaired and the opening of the door was unintentional. Gildewell v. Arkhola Sand & Gravel Co., 212 Ark. 838, 208 S.W.2d 4, 8 (Ark.1948).

The court also holds that the proximate cause analysis in Donnelly was not limited to the mental state of whoever lifted the door latch. In my view, if Andrew did not intentionally open the door, then his action of opening it would not amount to an efficient intervening cause under Arkansas's proximate cause standard. If Andrew was impaired, then the reason he opened the train's emergency door was because he could. If such was the case, I do not believe Andrew “affirmatively operated the door,” and Amtrak's failure to better secure the door may be found to have proximately caused Andrew's death.

“[P]roximate causation is usually a question of fact for a jury.” Neal v. Sparks Regional Med. Ctr., 2012 Ark. 328, 422 S.W.3d 116, 121 (Ark.2012). In Arkansas, “[t]he mere fact that other causes intervene between the original act of negligence and the injury for which recovery is sought is not sufficient to relieve the original actor of liability if the injury is the natural and probable consequence of the original negligent act or omission and is such as might reasonably have been foreseen as probable.” Stecker v. First Commercial Trust Co., 331 Ark. 452, 962 S.W.2d 792, 796 (Ark.1998) (jury question on proximate causation where doctor failed to report possible child abuse and child was subsequently harmed again). “[W]hen the voluntary acts of human beings intervene between the defendant's act and the plaintiff's injury, the problem of foreseeability is still the same: Was the third person's conduct sufficiently foreseeable to have the effect of making the defendant's act a negligent one?” Hartsock v. Forsgren, Inc., 236 Ark. 167, 365 S.W.2d 117, 118 (Ark.1963) (citing Harper & James, The Law of Torts, § 20.5; Rest., Torts, § 447).

The court acknowledges that Amtrak knew that over its 45 year history, 76 people have exited moving trains between stops, and at least 29 of those individuals were confused. Amtrak also had knowledge, through experience and previous tort claims alleged against it, that the emergency door from which Andrew most likely exited was fairly easy to open, even when the person opening it is confused. It was secured only by a simple dog latch in the top right hand corner of the door. Once the dog latch is open, the door may be opened by pushing down on the door handle. Making the door harder to open would likely have “cause[d] [Andrew] not to do the act” of walking off the train. Id. There was enough historical evidence of similar incidents to conclude that Amtrak may have breached a duty to Andrew, and that evidence likewise presents a jury question on proximate cause.

Tort plaintiffs who are confused or disoriented have escaped summary judgment or adverse jury verdicts on negligence claims even where they took the final action before their accidents. See, e.g ., Northrup v. Archbishop Bergan Mercy Hosp., 575 F.2d 605, 606–08 (8th Cir.1978) (district court did not err in refusing to reverse jury verdict for plaintiff where confused decedent fell after hanging from ledge outside of unsealed hospital window); see also Williams v. Melby, 699 P.2d 723, 725, 728–29 (Utah 1985) (reversal of summary judgment for defendant where plaintiff had arranged furniture in dangerous fashion and then fell through bedroom window while disoriented). Moreover, in cases regarding unwitnessed train exits, there seems to be an important difference between circumstances that “are wholly unknown to the parties,” Donnelly, 16 F.3d at 945, and a circumstance in which there is evidence of confusion. See, e.g., Harris v. Nat'l Pass. R.R. Corp. ., 79 F.Supp.2d 673, 678 (E.D.Tex.1999) (“Unlike the present case, there was evidence [in the Kilsgaard case, where the court found plaintiff's death was proximately caused by Amtrak,] that the victim might have been disoriented and mentally confused because of his propensity towards epileptic seizures. Here, there is no evidence that either Harris or Gust were confused, disoriented, or disabled in any way so that they confused the vestibule door for another door.”).

If the jury could find that Andrew was in fact impaired and that his unwitting exit through the emergency side door was the most likely of circumstances, they could also find that Amtrak proximately caused his fall. To state otherwise would suggest that no plaintiff may ever succeed in proving a negligence case against a common carrier where the decedent's exit goes unwitnessed or where someone other than an Amtrak employee opens the door or window from which the decedent falls—even where the defendant has breached a duty to the plaintiff. I do not believe such a suggestion is warranted. The district court found a dispute of material fact on whether Amtrak had breached a duty to Andrew, in large part based on his advanced age and the evidence of his confusion. That evidence should also be available to the jury to draw “reasonable inferences” concerning proximate causation. While I believe this is a close case, on balance, I would conclude that Haukereid has created a dispute on a genuine issue of material fact concerning causation. I respectfully dissent.

MURPHY, Circuit Judge.

Copied to clipboard