National Railroad Passenger Corp. v. McDavitt

804 A.2d 275 (2002)

NATIONAL RAILROAD PASSENGER CORPORATION, Appellant,
v.
Ray E. McDAVITT, Appellee.

Nos. 98-CV-1881, 99-CV-195.

District of Columbia Court of Appeals.

Argued September 28, 2000. Decided July 25, 2002.

Joseph S. Crociata, with whom Stephanie S. Ryan was on the brief, Washington, DC, for appellant.

Lawrence M. Mann, Washington, DC, with whom David A. Richman was on the brief, for appellee.

Before STEADMAN, RUIZ and GLICKMAN, Associate Judges.

GLICKMAN, Associate Judge:

Ray E. McDavitt filed a complaint under the Federal Employers' Liability Act (FELA), 45 U.S.C. §§ 51-60, against the National Railroad Passenger Corporation (Amtrak) for negligence after he was injured in a train derailment in the Washington Terminal coach yard near Union Station in the District of Columbia. Amtrak denied fault and blamed McDavitt for the accident. The case was tried to a jury, which found Amtrak negligent and McDavitt contributorily negligent. Apportioning liability under FELA's comparative negligence framework, the jury found two-thirds of McDavitt's damages attributable to Amtrak's negligence and the remaining one-third McDavitt's own responsibility. The jury put a value of $975,000 on McDavitt's damages; it evidently credited McDavitt's evidence that he was permanently and totally disabled by the injuries he sustained in the derailment. In accordance with the jury's allocation of responsibility, the trial judge reduced the total damages by one-third and entered judgment for McDavitt in the amount of $650,325.

Amtrak contends on appeal that the trial judge erred in denying its motions during and after trial for judgment as a matter of law or for a new trial, in making certain rulings on evidence, and in giving certain instructions to the jury. We affirm the judgment against Amtrak on the issue of liability, including the allocation to Amtrak of responsibility for two-thirds of McDavitt's damages. As to the determination of those damages, however, we reverse and remand the case for a new trial.

I.

Amtrak claims that it was entitled to judgment as a matter of law on McDavitt's negligence claims under FELA because there was no legally sufficient evidence from which a reasonable jury could find that Amtrak breached any applicable standard of care. See Super. Ct. Civ. R. 50(a)(1). In evaluating such a claim on appeal, we apply the same strict legal standard as a trial court does in ruling on the motion in the first instance. We must view the evidence in the light most favorable to the non-moving party (here, McDavitt), and give that party the benefit of every reasonable inference. See Vuitch v. Furr, 482 A.2d 811, 813 (D.C.1984). "The court must take care to avoid weighing the evidence, passing on the credibility of witnesses or substituting its judgment for that of the jury." Corley v. BP Oil Corp., 402 A.2d 1258, 1263 (D.C.1979). If "the case turns on controverted facts and the credibility of witnesses, the case is peculiarly one for the jury." Id. (quoting Aylor v. Intercounty Constr. Corp., 127 U.S.App. D.C. 151, 155, 381 F.2d 930, 934 (1967)). Judgment as a matter of law is proper only if the evidence is so clear that a reasonable jury could fairly come to but one conclusion. Id.

A.

On the afternoon of May 2, 1996, someone in the Amtrak control tower at Washington Terminal, known as "K-Tower," radioed permission to locomotive engineer Ray McDavitt to move his Virginia Railway Express (VRE) train number 323 from the coach yard south into Union Station. The VRE train was parked on track 4e adjacent to "dwarf" signal 491 (so called because the signal was only two and a half feet high), and its authorization to proceed was contingent "on signal indication." This meant that the engineer, McDavitt, needed to wait until signal 491 changed from "stop" (two horizontal red lights) to "slow clear" (two vertical green lights) before he could embark.

McDavitt testified that over the next few minutes he leaned out of the locomotive cab four times to check signal 491. The first three times the signal was still red, and he did not move. The fourth time he checked it, McDavitt said, the signal had turned green, and he set the train in motion. Moving south on track 4e at approximately thirteen miles per hour, VRE 323 passed through several switches, including one — identified as switch 482 — that was aligned against the train's movement. Although the conflicting alignment of this switch would have been visible from the cab of his locomotive, McDavitt did not notice it as he arguably should have and did not apply the brakes. The train broke the switch and rolled on until it arrived at switch number 472, some 537 feet from its starting point at signal 491. Unlike the earlier switches that VRE 323 had encountered, number 472 was a derail switch, designed to prevent collisions between trains traveling on conflicting routes by derailing one of them. Switch number 472 also was aligned against McDavitt's southward movement on track 4e. McDavitt again did not notice the misalignment as he approached the switch. His train derailed and traveled another ninety-two feet before it braked to a stop. McDavitt was thrown about in the cab and injured.

At the time of this accident, signal 491 and the switches on track 4e were part of an integrated, computerized system for controlling the movement of trains at Washington Terminal. Amtrak personnel located in K-Tower operated this system, communicating with trains by radio and assigning tracks and routes by means of an enter/exit, or "NX," machine. The layout of K-Tower and the appearance and operation of the NX machine were shown to the jury in photographs and videotapes that McDavitt and Amtrak each offered in evidence.

Fifteen feet in length, the NX machine had a lower and an upper panel. The lower panel was a control panel displaying a diagram of the tracks with lights, buttons and switches. To establish a train route in the terminal, such as a route for McDavitt's train into Union Station, an operator would push a button at the train's designated starting point, such as signal 491 in this case. The panel would light up to show all available routes from that point — routes not in conflict with those already assigned to other trains or for other purposes. The operator then would select a particular route from those available by pushing a button at the intended exit point on that route, such as McDavitt's intended destination in Union Station. The NX machine would align the switches on the selected route, taking into account any established delay periods for alignment changes, and then — after all the switches were aligned properly — change the signals on the route to permit the train to proceed. The progress of the train then would be displayed on the upper panel, an electrical "model board" of the tracks, by lights along each track that would change from white to red as the train moved past each signal on its path. Any train moving through the terminal area would show up on the board, even if it was moving without having first received a designated route or permission from K-Tower to proceed.

As designed, the NX machine would not establish conflicting routes. It would not change a signal from red to green until after all switches on the route were aligned. Furthermore, built-in delays ensured that if a signal were to turn red after a train passed it by, the switches on the route would remain aligned with the train's movement through them for two to two-and-a-half minutes — enough time for the train to get safely through the route, given the short distances involved in the terminal area. In other words, the integrated signal and switch system was designed so that signal 491 should not have turned green to permit McDavitt to move VRE 323 until all the switches ahead of him on track 4e were aligned in his favor. Once that happened and signal 491 changed to allow McDavitt to go, the switches should have remained aligned in his favor for at least two minutes even if signal 491 changed back to red or K-Tower established a conflicting route after he departed. If, therefore, McDavitt saw a green signal as he said he did, and if the system was functioning as designed, the accident should not have been possible. Either McDavitt testified untruthfully (as Amtrak contended) or the system malfunctioned.

If there was a system malfunction, it evidently had to do with signal 491 prematurely turning green by itself, because the NX operators in K-Tower still were waiting for another train to complete its travel on a conflicting route and had not yet established a route for VRE 323 to follow from signal 491. Amtrak investigators who examined the relay box switches after the derailment confirmed that K-Tower had not sent an electronic impulse to signal 491 to cause it to turn green. Yet when the investigators tested signal 491 and placed a twenty-four hour watch on it, they found nothing wrong with it.^[1] McDavitt, too, offered no evidence that signal 491 was defective, nor any explanation of why it turned green for him. Over Amtrak's objection, McDavitt did present evidence that other signals in the integrated system controlled from K-Tower had turned red or green for no apparent reason on two earlier occasions. This evidence was not offered or admitted to prove that signal 491 malfunctioned, however, but only to show that Amtrak was on notice that it might need to take reasonable precautions against such mishaps.

McDavitt testified that after he saw Signal 491 turn green, he announced to his conductor that "VRE 323 has a slow clear out of the yard. Here we go." The conductor testified that he was "70 to 80 percent" sure that he heard this message over the radio, and he "believe[d]" that K-Tower ordinarily listened to such communications. None of the K-Tower personnel acknowledged having heard McDavitt's announcement, however, and if he made it, no one responded to it. Amtrak's train director, Daniel Moffitt, who was in charge of K-Tower operations and working there at the time of the derailment, testified that K-Tower monitored the Amtrak road channel and "most likely" would have heard McDavitt if he had radioed that VRE 323 was about to move. Although K-Tower also listened to other radio channels, and there was sometimes interference, the Amtrak road channel was set at a higher volume than the others. [Id. at 127] Moffitt added that he personally spent "a lot of [his] time talking on the Amtrak road channel." [Id. at 127] An assistant train director who was working with Moffitt in K-Tower agreed that a radio message from McDavitt could have been heard there. She also confirmed that if McDavitt's message had been heard, the train director could have communicated with him immediately by radio and told him to stop.

Over the next thirty-six seconds, as measured by an on-board event recorder, VRE 323 proceeded down track 4e toward its rendezvous with the derail switch. During that span of time, red lights lit up sequentially on the upper panel of the NX machine, displaying the train's progress. Moffitt saw that these red lights were still lit after the accident. He acknowledged that if he had noticed the red lights when they went on, he "probably" would have known at once that it was McDavitt's VRE train, and "absolutely" would have done something, such as radio McDavitt, to stop his unauthorized movement.

McDavitt testified that he could have braked the train to a halt in five seconds at the thirteen mile per hour speed it was traveling. From the time VRE 323 left signal 491, therefore, there was a window of approximately half a minute in which someone in K-Tower might have prevented the derailment by ordering McDavitt to stop. No one made that call. No one in K-Tower admitted to having seen the movement of McDavitt's train reflected in the red lights on the upper panel of the NX machine. Along with Moffitt, there were four other Amtrak employees working in K-Tower, all of them generally facing the NX machine. Two of these employees, both assistant train directors, were operating the machine that afternoon. There was no evidence that they were distracted by other operations or duties.

B.

FELA allows a railroad employee to recover damages for work-related injuries "resulting in whole or in part" from the negligence of the railroad's agents or from "any defect or insufficiency" in the railroad's equipment due to its negligence. 45 U.S.C. § 51. See Keranen v. Nat'l R.R. Passenger Corp., 743 A.2d 703, 711-12 (D.C.2000); McMillan v. Nat'l R.R. Passenger Corp., 648 A.2d 428, 432 (D.C.1994); Fogg v. Nat'l R.R. Passenger Corp., 585 A.2d 786, 789 (D.C.1991). Thus, although a FELA plaintiff must prove a failure on the part of the railroad to use reasonable care under the circumstances, "a relaxed standard of causation applies under FELA." Consol. Rail Corp. v. Gottshall, 512 U.S. 532, 543, 114 S. Ct. 2396, 129 L. Ed. 2d 427 (1994). The plaintiff may prevail if the railroad's negligence "played any part, even the slightest," in causing the plaintiff's injury. Rogers v. Missouri Pac. R.R. Co., 352 U.S. 500, 506, 77 S. Ct. 443, 1 L. Ed. 2d 493 (1957).^[2]

The trial judge in a FELA action is "bound" to apply that relaxed standard in deciding whether the plaintiff has presented a prima facie case:

Judicial appraisal of the proofs to determine whether a jury question is presented is narrowly limited to the single inquiry whether, with reason, the conclusion may be drawn that negligence of the employer played any part at all in the injury or death. Judges are to fix their sights primarily to make that appraisal and, if that test is met, are bound to find that a case for the jury is made out whether or not the evidence allows the jury a choice of other probabilities.

Id. at 506-07 (footnote omitted). "Congress vested the power of decision in these actions exclusively in the jury in all but the infrequent cases where fair-minded jurors cannot honestly differ whether fault of the employer played any part in the employee's injury." Id. at 510 (footnote omitted).

Negligence, for purposes of the FELA, may be defined as "the failure of a railroad's agents to do what a reasonable and prudent man would ordinarily have done under the circumstances of the situation." McMillan, 648 A.2d at 432 (quoting Anderson v. Atchison, T. & S.F. Ry., 333 U.S. 821, 823, 68 S. Ct. 854, 92 L. Ed. 1108 (1948)) (internal quotation marks and brackets omitted). While the question of negligence in a FELA action is ultimately one of federal law, the plaintiff must establish the usual common law elements of duty, breach, foreseeability and causation in order to prevail. See Keranen, 743 A.2d at 711-12; McMillan, 648 A.2d at 432.

In his closing argument, McDavitt presented the jury with more than one theory of negligence on the part of Amtrak. The trial judge then instructed the jury on the particulars of each theory. At that time, however, neither Amtrak nor McDavitt requested that the jury return findings on each theory on a special verdict form. Instead, the jury returned only a general verdict in which it found that Amtrak had been negligent.

We must affirm the denial of Amtrak's motions for judgment as a matter of law so long as the evidence, viewed in the light most favorable to McDavitt, was sufficient to justify submission of any one of his theories of negligence, even if the evidence did not justify submission of the other theories. See Nimetz v. Cappadona, 596 A.2d 603, 608 (D.C.1991) (adopting "the rule that a defendant who fails to request a special verdict form in a civil case will be barred on appeal from complaining that the jury may have relied on a factual theory unsupported by the evidence when there was sufficient evidence to support another theory properly before the jury").^[3] "The `mere theoretical possibility that the jury based its decision' on one theory of negligence rather than the other ... is not basis enough to call its verdict into question." George Washington Univ. v. Lawson, 745 A.2d 323, 329 (D.C.2000) (quoting Nimetz, 596 A.2d at 607).

The theory of negligence that commended itself to the trial judge when he denied Amtrak's motions for judgment as a matter of law was founded on the failure of K-Tower to monitor the unauthorized movement of McDavitt's train. McDavitt argued that K-Tower personnel should have been alerted to that movement by hearing his radio communication ("VRE 323 has a slow clear out of the yard. Here we go.") and seeing the lights turn red on the NX machine, and that their inattentiveness and consequent failure to radio him to stop before he derailed were negligent.^[4]

Viewing the evidence in McDavitt's favor as we must, we agree with the trial judge that it was proper to submit this theory of liability to the jury. The jury could have found that the K-Tower personnel responsible for overseeing the movement of trains in the Washington Terminal coach yard were notified by radio and the NX machine that VRE 323 was proceeding without their required authorization, but paid no heed. The jury also could have found that if the K-Tower personnel had not ignored critical information that literally was staring them in the face for half a minute, they would have been able to stop VRE 323 before it derailed. From such findings the jury reasonably could have concluded that Amtrak's agents failed to do "what ... reasonable and prudent [persons] would ordinarily have done under the circumstances of the situation," McMillan, 648 A.2d at 432 (citations omitted), thereby causing McDavitt to sustain injury.

We are not persuaded by Amtrak's arguments that McDavitt failed to present evidence of any industry standard of care specifically requiring railroad terminal control tower operators to monitor the movement of trains that have not been cleared for specific routes, and that expert testimony was necessary to establish such a standard of care. The plaintiff in a negligence action must put on expert testimony to establish the applicable standard of care if the subject dealt with is so distinctly related to some science, profession, or occupation as to be beyond the ken of the average layperson. See District of Columbia v. Arnold & Porter, 756 A.2d 427, 433 (D.C.2000). In the present case, for example, we have little doubt that expert testimony on the standard of care could have been required to establish McDavitt's theories that Amtrak was negligent in selecting and siting dwarf signal 491, in not taking appropriate measures to protect against signal malfunctions, and in relying on allegedly outdated technology. See supra, n. 4. Each of those matters depends heavily upon technical or specialized knowledge that a lay jury would not be expected to possess. See, e.g., Collins v. Seaboard Coast Line R.R., 675 F.2d 1185, 1194-95 (11th Cir.1982) (requiring expert testimony in assessing a train's safe speed limit and the danger of a railroad crossing); Rice v. Cincinnati, N.O. & Pac. Ry. Co., 920 F. Supp. 732, 737-38 (E.D.Ky. 1996) (requiring expert testimony on the adequacy of locomotive cab design).

On the other hand, "we have refused to require expert testimony when the issue before the jury did not involve either a subject too technical for lay jurors to understand or the exercise of sophisticated professional judgment." District of Columbia v. Hampton, 666 A.2d 30, 36 (D.C.1995). "Proof of a deviation from the applicable standard of care need not include expert testimony where the alleged negligent act is within the realm of common knowledge and everyday experience." Toy v. District of Columbia, 549 A.2d 1, 6 (D.C.1988) (internal quotation marks and citation omitted). In the present case, the trial judge who heard and saw what the jury did concluded that no special expertise was necessary to evaluate whether K-Tower personnel had a duty to respond when unauthorized train movements in the Washington Terminal coach yard were reported by radio and the NX machine. We agree that the scope of the duty of railroad traffic controllers to attend to notifications of unauthorized train traffic in the confined space of a rail yard is not so esoteric as to be beyond the ken of the average layperson. It is a subject amenable to common sense, to which the lay jurors in this case could "apply their own experience in deciding how any reasonably prudent person would have acted under the circumstances." Hampton, 666 A.2d at 36 n. 13 (quoting Washington Hosp. Ctr. v. Martin, 454 A.2d 306, 309 (D.C.1982)).^[5]

II.

Amtrak argues that the trial judge erred in admitting over its objection evidence concerning two incidents in 1995 in which complaints were made about signals controlled from K-Tower. One of the incidents, the so-called "dropped signal" event, was written up in an Amtrak investigative report. In September 1995, according to the report, an engineer proceeding past a "slow clear" signal radioed K-Tower that the next signal on his route suddenly (and frighteningly) changed to "stop" when his train was less than a car's length away from it. Investigators were unable to determine what caused the signal to "drop" from green to red in this manner. The report quoted a statement attributed to an Amtrak supervisor whose duties included maintaining the signal system that "this type of situation does happen from time to time and may never happen again." This supervisor testified at trial and denied making that statement, but he admitted on cross-examination that there had been times when he did not know what caused a signal to drop. The other incident, a "false proceed" event, was recounted by Dean Denno, an Amtrak assistant train director, during his August 1997 deposition. In his deposition testimony, which was read into the record at trial, Denno recollected only that "a couple of years ago roughly speaking, [a train proceeded] southward beyond H signal bridge without any display route in the tower and the engineer reported that he had some signal to proceed and I don't recall the details." Signal 491, the signal that allegedly turned green for McDavitt in May 1996, was not involved in either of these earlier incidents.

The trial judge ruled that because the signals in the 1995 events were part of the same signal system controlled from K-Tower as was signal 491, those events were sufficiently similar to the alleged malfunction of signal 491 to be admissible for the limited purpose of the issue of notice and foreseeability. Accordingly, the judge carefully instructed the jury not to consider the 1995 events as evidence that signal 491 malfunctioned in 1996, but only on the question of whether Amtrak had notice of "prior alleged malfunctions" that should have caused it to take "reasonable precautions against future mishaps."^[6] Given this theory of admissibility, we cannot say that the trial judge abused his discretion in admitting evidence of the two 1995 incidents.

Whether Amtrak had notice that the signals in its system around the Washington Terminal coach yard might malfunction as signal 491 allegedly did was relevant — in fact, it was critical — to one of McDavitt's theories of negligence, the theory that Amtrak should have taken precautions to detect and correct problems with signal 491 or otherwise protect him from its malfunction. See, e.g., Gallose v. Long Island R.R. Co., 878 F.2d 80, 85 (2d Cir.1989) ("[I]f an employer learns or should learn of a potential hazard, it must take reasonable steps to investigate and to... protect its employees or it will be liable when injury occurs."). Hence, it was appropriate for McDavitt to seek to establish that comparable signal malfunctions had come to Amtrak's attention previously. "Evidence of prior incidents is generally admissible to show a defendant's notice or knowledge of a dangerous condition that causes an injury." District of Columbia v. Doe, 524 A.2d 30, 34 (D.C.1987).

To be admissible, the "prior incidents should have occurred under substantially similar temporal as well as physical conditions," but "[t]his requirement of similarity of circumstances . . . is `much relaxed' when the evidence is merely proffered to show notice" as opposed to other legitimate purposes.^[7]Doe, 524 A.2d at 34-35 (citations omitted). "Since all that is required is that the previous injury or injuries be such as to call defendant's attention to the dangerous situation that resulted in the litigated accident, the similarity in the circumstances of the accidents can be considerably less than that which is demanded when the same evidence is used for one of the other valid purposes." 1 McCormick on Evidence § 200 at 708 (John W. Strong ed., 5th ed.1999) (footnote omitted).^[8]See also Borden, Inc. v. Florida E. Coast Ry. Co., 772 F.2d 750, 755 (11th Cir.1985) (concluding that evidence of vandalism against a signal switch less than a mile from the site of a derailment caused by similar vandalism six months later should have been admitted because it might have "allow[ed] the jury to draw a reasonable inference concerning [the railroad's] ability to foresee this type of vandalism and its results").

The 1995 incidents were not too remote in time from the incident involving McDavitt in May 1996, and an unexplained signal conflict allegedly occurred in each instance. In the "dropped signal" event, the signal reportedly changed to red in conflict with the fact that a route had been established and the previous signal was green. In the "false proceed" event, the signal reportedly showed green in conflict with the fact that no route had been established at all. In McDavitt's case, signal 491 allegedly showed green too, in conflict with the fact that no route had been established. In the absence of any explanation for what should not have happened given the design of the system, the jury reasonably could conclude that the 1995 events put Amtrak on notice that its signals might have malfunctioned and foreseeably might do so again in 1996.

We admit that the notice question in this case is a close one. Amtrak points out differences between the 1995 events and the claimed malfunction of signal 491. In the "dropped signal" event, the signal turned red, not green. It was not interlocked with other signals and switches on its route, and it could have been changed manually by an operator in K-Tower. Signal 491, on the other hand, could not have been changed to green manually without a route having been established. The "false proceed" event was uncorroborated, and the account of it was vague. The testimony did not specify the exact location of the signal, whether it was interlocked, or whether the cause of the "false proceed" sign was investigated and determined. We think, however, that these considerations did not preclude the admission of the reported 1995 events for the limited use to which they were put. Even with all the differences Amtrak identifies, the fact that Amtrak had reports of two unexplained "wrong" signal events in 1995 still arguably placed the company on notice that signal malfunctions foreseeably could result in accidents unless it took appropriate precautions. The trial judge could reasonably determine that the differences, which Amtrak was able to develop at trial, went to the weight to be given the prior events rather than their admissibility. See Jones & Laughlin Steel Corp., 348 F.2d at 400-01.^[9]

III.

Amtrak argues that the trial judge erred in excluding evidence of McDavitt's disciplinary record as a railroad employee, which Amtrak contends was relevant to McDavitt's claim for lost wages. The judge ruled that "the prejudice and the [risk] of confusion far outweigh[ed] the probative value of this evidence." Although this ruling was a discretionary one, we agree with Amtrak.

A.

McDavitt claimed to be totally disabled by physical and cognitive injuries that he sustained in the derailment. In addition to damages for pain and suffering and loss of ability to enjoy life, he sought substantial pecuniary damages for lost wages and lost earning capacity. His economic expert witness, Thomas Roth, rendered the opinion that McDavitt's net wage loss between May 1996 when he was injured and November 1998 when the case was tried amounted to $140,668. Roth further opined that the present value at the time of trial of McDavitt's future lost earning capacity was $721,614, for a total economic loss of $862,282. Roth based these opinions on the assumption that if McDavitt, who was then 53 years old, had not become disabled, he would have been able to pursue "a complete and full work life" as a locomotive engineer until he retired at age 65. Roth also assumed that McDavitt's wages as an engineer would have grown at the rate of three percent per year until his retirement.

Amtrak argued that McDavitt's disciplinary record undermined Roth's assumptions. Amtrak proffered that McDavitt had the following record during his railroad career before the derailment:

Date Infraction Sanction May 1995 Running a stop signal Sixty-day suspension March 1990 Running through a switch aligned against Seven-day suspension his movement 1984 Driving under the influence while on call Termination of for work employment 1980 or 1981 Collision in a train yard One-week suspension

Amtrak further proffered that in light of this record, its industry and expert witnesses would testify that McDavitt's culpability in the May 1996 derailment would render him unemployable as a locomotive engineer in the future.

Roth did not take McDavitt's disciplinary history into account in his wage loss projections. For example, McDavitt's wages in 1995 — the last full year before his accident — were lower than in prior years because of his two-month suspension. Roth disregarded the 1995 income dip, however, because he assumed that McDavitt's unexplained absence from work in 1995 was an "aberration."

The trial judge's ruling barred Amtrak from using McDavitt's disciplinary record to challenge Roth's assumption that but for his accident, McDavitt would have gone on to have a full and uninterrupted career as a locomotive engineer. The ruling also prevented Amtrak from eliciting testimony based on McDavitt's record from its own industry and expert witnesses.^[10]

B.

In personal injury actions, including such actions under FELA, the loss of future earnings — or, more precisely, the loss of future earning capacity — "is a distinct item of damages, which if properly proved at trial, may result in recovery for the plaintiff." District of Columbia v. Barriteau, 399 A.2d 563, 567 (D.C.1979); accord, Sinclair v. Long Island R.R., 985 F.2d 74, 78 (2d Cir.1993) (FELA action). The "amount that the injured party would have earned but for the injury," Barriteau, 399 A.2d at 567 n. 6, is not susceptible to precise measurement. Generally speaking, where there is a history of employment to consider, the future earnings loss may be gauged by comparing "the demonstrated earning capacity of the injured party prior to the injury ... projected over the remaining working life of the injured party" (taking into account foreseeable adjustments, such as expected wage increases) with "what the injured party can now be expected to earn in light of diminished physical capacity." Id.

In evaluating lost earning capacity, the plaintiff's occupational abilities, industriousness, work habits and experience are relevant. See, e.g., Bower v. O'Hara, 759 F.2d 1117, 1123 (3d Cir.1985) (holding that proffered testimony of former employer that plaintiff was an unreliable worker "was plainly relevant to the issue of damages, as it bore on his claim for loss of earning capacity"); see generally 2 Stuart M. Speiser, Charles F. Krause & Alfred W. Gans, The American Law of Torts § 8:27 at 629-32 (1985) (discussing the factors bearing on the determination of lost earning capacity).^[11] The plaintiff's work history, good or bad, bears on whether, how, and how long the plaintiff probably would have been employed had he not been injured.

Thus, McDavitt's railroad disciplinary record was relevant to his claim for lost earnings. His history of rule infractions and resulting interruptions of employment indicated, at least arguably, that his future as a locomotive engineer was clouded and uncertain even before he became disabled. That history therefore undercut the optimistic assumptions of McDavitt's economist and could have been used by Amtrak's expert to support a lower lost earnings projection. Cf. Hughes, 391 A.2d at 263.

The trial judge did not think otherwise. He excluded evidence of McDavitt's disciplinary record not because he considered it irrelevant, but because he deemed it unfairly prejudicial on the issue of liability.^[12] The judge was concerned, for example, that the jury might reason that if McDavitt ran a stop signal a year before (not to mention his older but comparable infractions), it was more likely that he did so on the present occasion as well.

"Where evidence has probative value and is thus relevant, but possesses the potential for prejudicial misuse by the jury, the trial judge, in exercising his discretion as to its admission, must weigh the probative value against the prejudicial impact." Punch v. United States, 377 A.2d 1353, 1358 (D.C.1977). "[W]e owe a great degree of deference" to the balance that the trial judge strikes. Johnson v. United States, 683 A.2d 1087, 1095 (D.C.1996) (en banc). The judge's discretion to exclude evidence because the jury might misuse it is circumscribed, however. Relevant evidence may not be excluded on this ground unless the danger of unfair prejudice "substantially" outweighs its probative value. Id. at 1099 (adopting Federal Rule of Evidence 403 with no exception for "other crimes" evidence). This standard favors "the admission of as much relevant evidence as reasonably possible." Id. Probative evidence "should not be excluded because of `crabbed notions of relevance or excessive mistrust of juries.'" Allen v. United States, 603 A.2d 1219, 1224 (D.C. 1992) (en banc) (quoting Riordan v. Kempiners, 831 F.2d 690, 698 (7th Cir.1987)).

The trial judge in this case identified a legitimate concern. Ordinarily, it is proper to exclude evidence of other accidents or instances of negligence when such evidence is offered merely to show a general propensity to be negligent and thereby "enhanc[e] the probability of negligence on the occasion in question." 1 McCormick, supra, § 189. But the balance may tilt in favor of admitting evidence of other negligent acts if that evidence is offered for another, proper purpose. See id. at § 200. Cf. Johnson, 683 A.2d at 1092 (reiterating that evidence of other crimes is inadmissible to show disposition to commit the crime charged, but may be admissible if offered for a "substantial, legitimate purpose" other than to show propensity). Amtrak did not seek to introduce evidence of McDavitt's past infractions to show his propensity for negligence. Rather, Amtrak sought to introduce that evidence for its bearing on his lost earning capacity. That was a material issue at trial to which McDavitt's record had considerable relevance. There was some risk that the jury would take McDavitt's past infractions as evidence that he was disposed to run red lights and probably did so in this case, but the trial judge had tools available to reduce that risk. The judge could have directed Amtrak's counsel not to argue propensity to the jury (which counsel did not seek leave to do), and he could have instructed the jury explicitly against misusing the evidence of McDavitt's record.^[13] If the judge doubted the effectiveness of those palliative measures by themselves, he could have taken the additional step of bifurcating the trial so that liability and damages would be tried separately. Although no party requested it, bifurcation to avoid prejudice is authorized by Super. Ct. Civ. R. 42(b). See, e.g., Cravens v. County of Wood, 856 F.2d 753, 755 (6th Cir.1988) (discussing situations in which "bifurcation is appropriate"). In a bifurcated trial, the judge could have excluded the evidence of McDavitt's past infractions from the liability phase of the proceeding to ensure that the jury would not consider his record in assessing fault. The judge then could have admitted the evidence in the damages phase only, limiting the jury's consideration of McDavitt's record to the issue to which it was relevant.

In short, McDavitt's disciplinary record was relevant to the material issue of lost earning capacity, and the trial judge had means to counter the risk of the jury's misuse of that record. We conclude that the danger of unfair prejudice did not substantially outweigh the probative value of the evidence. The evidence of McDavitt's record should not have been excluded.

The jury found that the total amount of money that would fairly compensate McDavitt for his injuries was $975,000. In making that finding, the jury did not break down the total figure into components for each type of injury that McDavitt sustained. We have no way of knowing how much, if any, of the $975,000 was for lost earnings.

We are constrained, therefore, to vacate the damages award in its entirety. We remand this case for a new trial limited to the ascertainment of McDavitt's damages, at which Amtrak shall be permitted to introduce competent evidence concerning McDavitt's disciplinary record.^[14]

Except as to damages, the judgment of the trial court is affirmed.

So ordered.

NOTES

[1] Amtrak's signal department carried out various other tests and inspections of the switches and the track circuitry. No problems or signs of malfunction were found in those areas.

[2] FELA is a comparative negligence statute. Contributory negligence on the part of the plaintiff does not bar recovery, but results in a proportionate reduction of the damage award. See 45 U.S.C. § 53.

[3] "Litigants like [the defendant] who wish to challenge the sufficiency of the evidence as to some, but not all, specifications of negligence must present an appropriate record for review by asking the jury to make separate factual determinations as to each specification." Nimetz, 596 A.2d at 607 (quoting McCord v. Maguire, 873 F.2d 1271, 1274 (9th Cir.1989)). In Newell v. District of Columbia, 741 A.2d 28, 33 (D.C.1999), we elaborated on Nimetz, holding that "in requesting a special verdict from, counsel must state the request with sufficient precision to indicate the specific interrogatories that should be included in the special verdict form, object to their noninclusion, and include the proposed special verdict form in the record on appeal."

[4] McDavitt also argued, first, that Amtrak's selection and placement of dwarf signal 491 was negligent, because it required him to lean out of his locomotive cab in order to see the signal, which he claimed was unsafe. [Supp. Rec. 4 at 241] Second, McDavitt argued that Amtrak was negligent in failing to take measures to protect against the malfunction of signal 491 after at least two previous incidents in which other signals had changed for no apparent reason. [Id. at 242-44] Third, McDavitt argued that Amtrak was negligent in relying on allegedly outdated technology to control the movement of trains in the Washington Terminal area. [Id. at 247] Amtrak makes a persuasive case on appeal that the evidence was insufficient to justify a finding of liability on any of these alternative theories.

[5] Amtrak cites us to Anderson v. Nat'l R.R. Passenger Corp., 866 F. Supp. 937, 940 (E.D.Va.1994), in which the district court rejected the claim that a dispatch center in Jacksonville, Florida, should have reacted to an abnormal switch signal indication in time to prevent an Amtrak passenger train from derailing in Newport News, Virginia. We think that case is distinguished from this one by its very different facts. Among other things, at the Jacksonville center a single dispatcher monitored the movement of trains throughout the rail system on eight large television screens that covered a far larger territory than was displayed on the single NX machine in K-Tower. Id. at 941. Additionally, "extending to Plaintiffs all benefit of the doubt respecting the location of the train when the switch was reversed, the record [in Anderson] establishe[d] that even the most attentive dispatcher could not have prevented the derailment," id., as the train was moving too fast and would have taken too long to stop once it began to brake.

[6] On appeal Amtrak complains that the wording of this instruction was faulty in that there was no evidentiary predicate for characterizing the past signal events as "malfunctions" or "mishaps." See Gebremdhin v. Avis Rent-A-Car System, Inc., 689 A.2d 1202, 1204 (D.C. 1997) ("Jury instructions must have an evidentiary predicate."). While Amtrak objected unsuccessfully to the trial judge's decision to instruct the jury that it could consider the 1995 events on the issue of notice, it did not object with its current specificity to the particular wording that the judge employed. Assuming for the sake of argument that Amtrak's objection to that wording is preserved nonetheless, we are not persuaded by it. Contrary to Amtrak's contention, the instruction did not "improperly direct[] the jury to consider the prior incidents as malfunctions or mishaps." As we discuss infra, we are satisfied that the evidence of the 1995 events was a sufficient predicate for the instructions the judge gave. If the language arguably might have been more precise, the deviation from perfection was not material.

[7] In common with other types of evidence, proof of prior similar incidents "is also subject to the reasonable discretion of the trial court as to whether the defendant is taken by unfair surprise and as to whether the prejudice or confusion of issues which may probably result from such admission is disproportionate to the value of such evidence." Jones & Laughlin Steel Corp. v. Matherne, 348 F.2d 394, 400 (5th Cir.1965). We do not perceive these concerns to be present in this case.

[8] "Other valid purposes" for admitting evidence of prior incidents in a negligence case include proving the existence of a particular dangerous condition, the fact that the dangerous condition caused the injury, or the risk that the defendant's conduct created. See id. at 704-06.

[9] On another notice question, Amtrak contends that the trial judge erred in allowing McDavitt to present new evidence for the first time in rebuttal. In support of his claim that Amtrak negligently selected and sited dwarf signal 491, McDavitt testified in his case-in-chief that another crew had parked his train within six to eight feet of the signal, making it impossible for him to see the signal without leaning out of his locomotive cab. Over Amtrak's objection, the trial judge allowed McDavitt to testify in his rebuttal case that he had complained previously to Amtrak about the practice of parking VRE trains so close to the dwarf signal. Amtrak argues that the judge should have precluded this testimony because it did not rebut any evidence adduced in the defense case concerning lack of notice. We are unpersuaded. "The scope of rebuttal testimony is committed to the sound discretion of the trial court." Oxendine v. Merrell Dow Pharm., Inc., 506 A.2d 1100, 1113 (D.C. 1986). McDavitt's rebuttal testimony was related to the testimony he gave in his case-in-chief. Amtrak was not prevented from responding to it; indeed, following the rebuttal case, the trial judge granted Amtrak's motion to reopen its case to enable it to introduce photographs of the dwarf signal in response to McDavitt's claims concerning its size and location. We perceive no abuse of discretion on the part of the trial judge in allowing McDavitt to testify on rebuttal as he did.

[10] The trial judge did permit Amtrak to bring out the fact that McDavitt was disciplined in 1995 — but not the underlying reason (running a stop signal) or the sanction that was imposed (suspension for sixty days) — for the limited purpose of explaining why Amtrak pulled McDavitt's engineer's license after the 1996 derailment.

[11] "[T]here is no requirement such loss [of earning capacity] need be measured in a vacuum: ordinarily considered are plaintiff's ... health, education and opportunity for education, age, intelligence, industriousness, manner of living, sobriety or temperance, frugality or lavishness or other personal characteristics which affect ability to secure business or earn money." Ehlinger v. State, 237 N.W.2d 784, 792 (Iowa 1976), quoted in 4 Fowler V. Harper, Fleming James, Jr. & Oscar S. Gray, The Law of Torts § 25.8, at 550 (2d ed.1986). Accord, Hughes v. Pender, 391 A.2d 259, 263 (D.C.1978) (holding that trial court did not abuse its discretion in excluding an economist's projection of lost earnings on the ground that the economist "had failed to consider factors such as the decedent's school grades, intelligence level, or arrest record"). In addition to the personal characteristics of the plaintiff, "[e]nvironmental factors such as the condition of the labor market, the chance of advancement or of being laid off, and the like, are [also] entitled to consideration." Harper et al., supra at 550-51 (citations omitted).

[12] The trial judge mentioned in passing a concern about confusion of the issues as well as prejudice if he admitted evidence of McDavitt's disciplinary record. The risk of confusing the issues is a proper factor for a trial judge to consider in ruling on the admission of evidence. In this case, however, McDavitt did not dispute his record, and no serious danger of issue confusion was identified.

[13] That is what the trial judge did after he permitted Amtrak to introduce the fact that McDavitt had been disciplined in 1995 for the limited purpose of explaining why it pulled his engineer's license after the derailment. See supra, n. 10. The judge instructed the jury not to use that fact in determining whether McDavitt did anything wrong in connection with the accident of May 2, 1996.

[14] As we are requiring a new trial on damages, we need not address Amtrak's argument that the evidence did not justify an instruction that the trial judge gave concerning compensation for aggravation of a preexisting condition.