George v. Estate of Baker

OPINION

Gonkartee Dekpah was killed when the taxicab in which he was a passenger swerved across a median in a winter storm and was crushed by an oncoming tractor-trailer truck. Dekpah’s brother, appellant Bob Makor George, brought a wrongful death action against the drivers of the cab and truck. The jury found that both drivers were negligent but that neither one’s negligence caused the accident. The court of appeals affirmed the district court’s denial of judgment notwithstanding the verdict (JNOV). George argues that the jury’s finding of negligence but not causation was perverse, that trial errors in evi-dentiary rulings and jury instructions require a new trial on liability, and that the district court erred by instructing the jury on the life expectancy of a Liberian male. We affirm in part, reverse in part, and remand.

On the morning of March 14, 2002, Dek-pah called a taxicab to take him from his home in Brooklyn Park to a medical appointment in Plymouth. Respondent Dennis Baker, driving for the Pioneer Taxi fleet, was dispatched and picked up Dek-pah. At approximately 10:49 a.m., Baker’s cab was southbound near Rockford Road on Interstate 494 when it fishtailed, crossed the 30-foot median, and entered northbound traffic roughly 10 feet in front of a semi driven by Respondent Brian Keith Losey. Losey applied his brakes but could not avoid broadsiding the taxi, killing Baker and Dekpah. George brought this wrongful death action against Baker’s estate, Baker’s employer AJD Transportation and the taxi’s owner Gabriel Russell Limongelli (collectively “Baker”), and against Losey and Losey’s employer, CRST, Inc. (collectively “Losey”).

George’s theory at trial was that Baker was negligent in failing to control his taxi and Losey was negligent in failing to stop. Baker and Losey each argued that they were not negligent and that the accident was due to inclement weather. All parties called accident reconstruction experts who offered conflicting testimony about the drivers’ respective speeds and behaviors.

The only known eyewitnesses to the accident were Losey and J.V., who was driving a box truck immediately behind Lo-sey’s semi. Losey first saw the taxi as it fishtailed in its left-hand lane prior to crossing the median but did not see what caused it to fishtail. J.V. did not see the taxi until it was leaving the median and entering oncoming traffic. Suggestions were made at trial that the taxi may have been traveling too fast or been cut off, rear-ended, or sideswiped by another car. Because weather at the scene precluded a full-fledged accident reconstruction and because the taxi was completely destroyed, none of these theories could be substantiated.

In support of the weather theory, Baker called a state statistician to testify, over George’s objection, that 481 other traffic accidents occurred on the same day in the metro area. J.V. and a Minnesota Highway Patrol trooper also testified about the inclement weather and a number of other accidents that occurred on that day. Four witnesses testified that Baker was generally a careful person and was unlikely to drive unsafely. George objected to some, but not all, of this testimony.

*5 The parties disputed the applicable standards of care. George requested an instruction that Baker, as a common carrier, was subject to the “utmost caution” standard. 1 George also requested an instruction that Losey, as a commercial motor vehicle driver, was subject to an “extreme caution” standard when driving conditions were hazardous. 2 In his closing argument, George’s counsel told the jury that “[t]he reasonable care standard does not apply here to either the taxi or the semi; both have special standards.” In response to Baker and Lose/s objection to that statement, the district court began the next day of trial with this incorrect curative instruction: “Mr. Jacobson, during his final argument yesterday, stated that the reasonable person standard did not apply to either Defendant. This is erroneous. The reasonable person standard applies to both Defendantfe].” The court then read the full jury instructions, including an instruction that Baker was subject to the utmost caution standard for the taxi, that Losey could be subject to the extreme caution standard for the semi if the jury found hazardous conditions, but that Losey would be subject to the reasonable care standard if the jury did not find hazardous conditions. 3 The court also instructed the jury that a “direct cause” is one which “has a substantial part in bringing about the accident.” 4

The parties also disputed the jury instructions regarding Dekpah’s life expec-fancy. Dekpah, who was 52 years old at the time of his death, was born in Liberia and immigrated to the United States in 1996. Baker requested that the court instruct the jury that the life expectancy of a Liberian male is 47.03 years, a figure taken from the CIA World Factbook. George objected that this information was irrelevant. The district court instructed the jury that future life expectancy of a 52-year-old African-American male is 73.4 years and that the life expectancy of a Liberian male is 47.03 years. The court further instructed that these figures were not conclusive, and that the jury should consider them along with Dekpah’s health, physical condition, and other relevant factors. 5

The jury returned a special verdict, finding that both Baker and Losey had operated their respective vehicles negligently, but that neither one’s negligence was a direct cause of the accident. The jury also found that Dekpah’s heirs were damaged in the amount of $68,000. The district court adopted the jury’s findings and dismissed the case. George then moved for JNOV or a new trial. In support of JNOV, George argued that the jury verdict was perverse. In support of a new trial, George argued that several evidentia-ry and jury instructions constituted reversible error. The district court denied both motions.

George appealed the denial of his motions. As to the motion for JNOV, the *6 court of appeals held that the jury’s verdict was not perverse and not contrary to the evidence.

As to the motion for a new trial, the court of appeals held that the district court had erred in several respects. First, the district court’s curative instruction was mistaken because Baker was a common carrier subject to the utmost caution standard. George v. Estate of Baker, No. A05-108, 2005 WL 3289451, at *4 (Minn.App. Dec.6, 2005). Next, by allowing four witnesses to testify that Baker was a careful driver, the district court had allowed character evidence offered to prove conduct on a particular occasion, in violation of Minn. R. Evid. 404(a). George, 2005 WL 3289451, at *6. Finally, by allowing the statistician to testify to the number of other accidents that occurred that day, the district court abused its discretion because no foundation was laid to establish that those accidents were substantially similar to the one that killed Dekpah. Id. at *7. But the court of appeals held that none of these errors was prejudicial because the jury found both drivers negligent and because there was adequate independent evidence in the record to support the jury’s verdict of no causation. Id. at *4, *6, *7.

George also argued on appeal that the district court erred in providing the Liberian life expectancy figure to the jury. Id. at *5. The court of appeals held that the figure was relevant and that the jury was properly instructed on its use. Id.

We granted George’s petition for further review.

I.

George first asks us to reverse the denial of his motion for JNOV and determine liability as a matter of law against Baker. George argues that no reasonable jury could have found Baker negligent without also finding that his negligence was a direct cause of the accident, and thus that the jury’s verdict was perverse.

The standard that applies to a motion for JNOV is that the evidence must be “so overwhelming on one side that reasonable minds cannot differ as to the proper outcome.” Clifford v. Geritom Med, Inc., 681 N.W.2d 680, 687 (Minn.2004). We will not disturb a jury’s answer to special verdict questions if it can be reconciled on any theory, and will set aside a special verdict answer only if it is “perverse and palpably contrary to the evidence.” Hauenstein v. Loctite Corp., 347 N.W.2d 272, 275 (Minn.1984). Specifically, where a jury has found negligence without causation, we have been generally disinclined to second-guess the verdict. 6

*7 Here, the jury could reasonably have found Baker negligent without causation for a variety of reasons. We will mention just one: the jury may have decided that Baker was negligent in passing another car given the road conditions, but that he would have lost control, whether he passed or not, because of the effect of the weather. We cannot know exactly how the jury arrived at its verdict, but we must accept it because it was potentially reasonable.

George suggests that the effect of the weather was already factored into the jury’s negligence finding and therefore should not have been considered in determining causation. George’s argument would have merit if we could conclude that the sole basis for the jury’s finding that Baker was negligent was the traffic violation that Baker committed when his vehicle crossed the center line and faced oncoming traffic. Crossing the center line is a traffic violation. Minn.Stat. § 169.18, subd. 9 (2004). Violation of a traffic regulation is prima facie evidence of negligence. Minn.Stat. § 169.96(b) (2004). Proof of a traffic violation shifts the burden to the violator to show excuse or justification for the violation or to rebut the prima facie case. Holten v. Parker, 302 Minn. 167, 177, 224 N.W.2d 139, 145 (1974). Thus, if the sole basis for finding Baker negligent had been his crossing the center line, the effect of the inclement weather would be relevant to Baker’s excuse or justification. In such a case, it would not be reasonable to also conclude that weather was an intervening cause of the accident. But, the premise for this analysis disappears when George acknowledges, as he must, that there is no way to conclude that the traffic violation was the sole basis for the verdict. George’s own arguments suggest an alternative basis for Baker’s negligence — driving too fast for conditions or passing another car in icy conditions. Accordingly, we affirm the court of appeals’ determination that the district court did not err in denying George’s motion for JNOV.

II.

George argues, in the alternative, that the trial errors mandate a new trial on the liability of Baker and Losey. Although he made this argument in his brief and at oral argument, George failed to specifically include it in the statement of the issues section of his petition for further review. He did, however, mention trial errors and his request for a new trial in the body of the petition. The first issue, then, is whether George has properly preserved for review the issues arising from his motion for a new trial.

A petition for review to this court must specify the legal issues to be reviewed. Minn. R. Civ.App. P. 117, subd. 3(a). The court will generally not address issues that were not specifically raised in the petition for review. In re GlaxoSmithKline PLC, 699 N.W.2d 749, 757 (Minn.2005); see also Northwest Racquet Swim & Health Clubs, Inc. v. Deloitte & Touche, 535 N.W.2d 612, 613 n. 1 (Minn.1995). We do, however, grant petitioners some latitude in their presentation of issues. Hapka v. Paquin Farms, 458 N.W.2d 683, 686 (Minn.1990).

We have never made clear precisely what is required to raise a particular legal issue in a petition for review. A petition for review has four parts: a statement of the legal issues sought to be reviewed, a statement of the criteria relied upon to support the petition, a statement of the case, and a brief argument in support of the petition. Minn. R. Civ.App. P. 117. A strict reading of a petition would require *8 us to consider only those issues specified in the statement of issues. A liberal reading would allow us to consider any issues, whether mentioned in the petition or not. We decline to adopt either of these extremes and conclude that this question must be considered on a case-by-case basis and includes consideration of several factors, such as the relationship of the questioned issue to those specifically highlighted in the statement of issues, the procedural history concerning that issue, and any potential prejudice to the respondent by not specifically including it in the statement of issues.

On this record we conclude that George has adequately preserved for review the trial errors raised in his new trial motion. George made alternative motions for JNOV or a new trial before the district court and brought the denial of both motions for review by the court of appeals. From the beginning the two motions have been interrelated because they are, in a sense, two sides of the same coin. If the jury’s finding on causation is perverse, because a reasonable person could not conclude that there was negligence but not causation, then judgment on liability should be entered as a matter of law. But, if the finding of no causation is potentially reasonable, then the court should review whether errors at trial could have influenced the causation finding sufficiently to require a new trial. In the argument section of his petition for further review, George emphasized this relationship by stating:

In a bit of legal jujitsu, the Court of Appeals’ opinion places Baker in an oncoming lane of traffic due to his own negligence, in order to justify denying Petitioner a new trial based on the significant errors that occurred at trial, and then, in the same opinion, places Baker in an oncoming lane of traffic due to bad weather in order to justify the jury finding of no causation. This is an absurdity that needs to be corrected. Petitioner should either have been granted a new trial on liability because of the errors occurring at trial or a finding that Baker was causally negligent as a matter of law.

George also specifically stated that the district court “erred in instructing the jury that the standard of care applicable to the taxi was the ‘reasonable man’ standard of care rather than the ‘utmost caution’ standard applicable to common carriers.” As to the procedural history, George consistently presented these alternative issues to the district court and the court of appeals. In fact, the court of appeals’ opinion reflects full consideration of both alternative motions. More important, we are satisfied that counsel for Baker and Losey were on notice of George’s arguments for a new trial because they fully addressed them in their briefs and in their arguments before us. Although George’s petition for review was not artfully drafted, we are persuaded by the consistency of George’s presentation of these alternative motions through the course of this litigation, and by Baker and Losey’s ability to respond to those arguments at all levels, that the denial of the motion for new trial is adequately preserved for our review.

III.

The court of appeals held that the trial court erred in three ways: by allowing four witnesses to testify that Baker was generally a careful driver, by allowing a state statistician to testify to the number of other accidents in the metro area on that day, and by giving the curative instruction that the reasonable person standard of care applied to Baker. The court further held that all three errors were harmless. George argues that all three errors were prejudicial and'warrant a new *9 trial on liability. We will address all three errors, first as they relate to Baker, and then to Losey.

Evidentiary Rulings.

We consider the evidentiary errors first. An improper evidentiary ruling resulting in the erroneous admission of evidence will only compel a new trial if it results in prejudicial error to the complaining party. Kroning v. State Farm Auto. Ins. Co., 567 N.W.2d 42, 46 (Minn.1997). An evidentiary error is prejudicial if it might reasonably have influenced the jury and changed the result of the trial. See Jenson v. Touche Ross & Co., 335 N.W.2d 720, 725 (Minn.1983) (applying standard to exclusion of evidence) superceded by rule on other grounds, Minn. R. Prof. Conduct 1.10(b), as recognized in Lennartson v. Anoka Hennepin Indep. Sch. Dist. 11, 662 N.W.2d 125, 134-35 (Minn.2003). But the admission of evidence that is cumulative or is corroborated by other competent evidence will be deemed harmless and will not warrant a new trial. See Estate of Lea v. Sheehan, 301 Minn. 253, 259, 222 N.W.2d 92, 97 (1974); Minn. R. Civ. P. 61.

The court of appeals held that testimony about Baker’s careful nature was evidence of character offered to prove conduct in conformity therewith, and therefore a violation of Minn. R. Evid. 404(a), but that the error was harmless because the jury found Baker negligent despite the testimony. George, 2005 WL 3289451, at *6. George argues that the error may have prejudiced the jury’s causation analysis.

The record shows that George objected to some, but not all, testimony about Baker’s careful nature, and elicited some of the testimony himself on cross-examination. Several witnesses also testified about Baker’s driving history and service as a driver in the military, and George did not object to this testimony. The cumulative weight of all this testimony would have allowed the jury to conclude that Baker was generally a careful person. Furthermore, whatever- the jury’s impression of Baker’s usual level of prudence, they found him negligent on this occasion. Accordingly, we.conclude that this character testimony could not have affected the jury’s causation analysis. We hold that the admission of this testimony was harmless error.

The court of appeals held that the district court erred in admitting the testimony of the statistician because, where evidence of similar accidents is introduced to show a common, inanimate cause, a foundation must be laid showing that the circumstances surrounding the other accidents were substantially the same as those involved in the accident in litigation. George, 2005 WL 3289451, at *7; see Haukom v. Chicago Great W. Ry. Co., 269 Minn. 542, 555, 132 N.W.2d 271, 279 (1964). The court of appeals held, however, that the error was harmless because the jury found both drivers negligent and because the record contained other evidence concerning the weather, road conditions, and other accidents. George, 2005 WL 3289451, at *7. Because the record contained ample testimony to which George did not object from which the jury could have concluded that weather caused the accident, we agree with the court of appeals and hold that this error was harmless.

Curative Jury Instruction.

The court of appeals held that the district court’s curative instruction was error as applied to Baker because, as a common carrier, he was subject to the utmost caution standard. George, 2005 WL 3289451, at *4; see Lindstrom v. Yellow Taxi Co., 298 Minn. 224, 226, 214 N.W.2d 672, 674-75 (1974); McKellar v. *10 Yellow Cab Co., 148 Minn. 247, 249, 181 N.W. 348, 349 (1921). The court further held that the error was harmless because the jury found Baker negligent despite the instruction.

Although district courts have broad discretion to fashion curative jury instructions, a court errs if it gives a jury instruction that materially misstates the law. See State v. Kuhnau, 622 N.W.2d 552, 555-56 (Minn.2001). Such an error does not necessitate a new trial unless the error was prejudicial. Rowe v. Munye, 702 N.W.2d 729, 743 (Minn.2005). A jury instruction is prejudicial if a more accurate instruction would have changed the outcome in the case. Lewis v. Equitable Life Assurance Soc’y of the U.S., 389 N.W.2d 876, 885 (Minn.1986). If the effect of the erroneous instruction cannot be determined, we will give the complainant the benefit of the doubt by granting a new trial. Rowe, 702 N.W.2d at 743; Morlock v. St. Paul Guardian Ins. Co., 650 N.W.2d 154, 159 (Minn.2002) (“A new trial is also required if the instruction was erroneous and its effect cannot be determined.”).

We agree with the court of appeals that Baker was subject to the utmost care standard and that the curative instruction was error. To determine whether that error was prejudicial, we must first decide whether the correct instruction, given later, ameliorated the error and, if not, whether the erroneous standard of care instruction could have influenced the jury’s causation analysis.

We conclude that the later instruction did not ameliorate the curative instruction. After George’s counsel correctly stated in his closing argument that the reasonable person standard did not apply to Baker, the district court began the next day of trial by announcing that George’s counsel made an erroneous statement of law and that the reasonable care standard applied to both defendants. The circumstances under which the court delivered the curative instruction made it more memorable and important to the jury than the correct statement of the utmost caution standard that was given in the midst of the court’s comprehensive instructions. The conflicting instructions may, at the least, have created confusion about which instruction to follow.

We must next consider whether this confusion could have affected the jury’s causation analysis. While the standard of care is more directly relevant to negligence, we believe that applying the wrong standard of care may influence the jury’s causation analysis because the standard of care defines the specific negligent act, which then becomes the central focus of a causation analysis.

We have recognized the close relationship between negligence and causation. In Clifford v. Geritom Med, Inc., we noted that “negligence and causation in this case are so closely related that we cannot say the jury’s confusion as to causation did not taint its understanding of negligence.” 681 N.W.2d at 688-89. See also Koenigs v. Werner, 263 Minn. 80, 85, 116 N.W.2d 73, 76 (1962) (“Here we have an intersection accident where the issue of liability may involve the closely related elements of proximate cause and negligence of the parties.”).

Minnesota applies the substantial factor test for causation. The negligent act is a direct, or proximate, cause of harm if the act was a substantial factor in the harm’s occurrence. 4 Minn. Dist. Judges Ass’n, Minnesota Practice—Jury Ins truction Guides, Civil, CIVJIG 27.10 (4th ed.1999); Restatement (Second) of Torts § 431 (1965). Factual, or but-for, causation is insufficient to establish liability in Minnesota because “[i]n a philosophical *11 sense, the causes of an accident go back to the birth of the parties and the discovery of America.” William L. Prosser, The Minnesota Court on Proximate Cause, 21 Minn. L.Rev. 19, 22 (1936). But-for causation, however, is still necessary for substantial factor causation because if the harm would have occurred even without the negligent act, the act could not have been a substantial factor in bringing about the harm. Restatement (Second) of Torts § 432 (1965); Draxton v. Katzmarek, 203 Minn. 161, 164-65, 280 N.W. 288, 290 (1938) (“If the accident * * * would have happened even if there had been an. absence of excessive speed, such speed was not a material element or substantial factor in bringing it about and hence not a proximate cause.”). The classic test for determining factual cause is to compare what actually happened with a hypothetical situation identical to what actually happened but without the negligent act. W. Page Keeton, et al., Prosser and Keeton on the Law of Torts § 41, at 265 (5th ed.1984). Because the negligent act is defined by the standard of care, an erroneous standard of care instruction will necessarir ly affect a jury’s causation analysis.

Applying this reasoning to the facts of this case, the jury might have believed (for example) that a reasonable person would have driven no more than 30 miles per hour under the conditions, but that a person exercising the utmost caution would have driven no more than 20 mph, or even would have pulled off the road. If the jury concluded that Baker was traveling 40 mph, it would have found that Baker’s negligent act was driving 10 mph faster than a reasonable person, whereas the jury could have found that Baker was driving 20 mph faster than a person exercising utmost caution, or that a person exercising utmost caution would not have been driving at all. Under the reasonable care standard a jury would compare actual events to events if Baker had driven 10 mph slower, and could possibly conclude that the accident would have occurred anyway and that but-for causation, and thus direct causation, was lacking. But the same jury, under the utmost caution standard, would compare actual events to events if Baker had driven 20 mph slower or had stopped driving entirely, and could possibly conclude that this difference would have been enough to avoid the accident, allowing the jury to find causation because the negligent act of driving 20 mph faster than a person exercising utmost care, or perhaps of failing to pull off the road, was a substantial factor in the harm.

Because it is analytically and practically possible that the erroneous standard of care instruction affected the jury’s causation analysis, we hold that the curative instruction was reversible error- and remand the case for a new trial on Baker’s liability. 7 Because we grant a new trial on this error, we need not decide whether the erroneous admission of character evidence regarding Baker’s careful nature or of the state’s statistical evidence regarding other accidents was also prejudicial.

IV.

In support of his motion for a new trial against Losey, George relies on the same evidentiary and jury instruction errors. The improperly-admitted character evidence and statistical evidence do not compel a new trial against Losey for the same reasons that they do not compel a new trial for Baker. This leaves the cura *12 tive instruction. George argues that the instruction may have prejudiced the jury’s causation analysis as to Losey by leading the jury to apply a reasonable care, rather than an extreme caution standard. The court of appeals held that the question of whether hazardous conditions existed to give rise to the extreme caution standard was a fact question for the jury, and therefore the district court’s instruction on both the extreme caution and reasonable care standards was proper. George, 2005 WL 3289451, at *4. We agree with the court of appeals that the existence of hazardous conditions was a fact issue for the jury and we decline to order a new trial on Losey’s liability.

V.

Because we remand for a new trial on Baker’s liability, we will also reach the issue of the use of the Liberian life expectancy figure. The court of appeals held that the figure was relevant to Dek-pah’s life expectancy and that the jury was properly instructed on its use. George argues that the figure was irrelevant to Dekpah’s life expectancy and that the information was biased in that it invited the jury to view Dekpah as an outsider. 8 This court has yet to decide to what extent, if any, a court should take notice of life expectancy data for foreign countries.

American mortality tables are compiled from the collective experience of insurers, who are financially motivated to ensure that the tables are accurate. Due to this inherent reliability, courts have taken judicial notice of American mortality tables as statistical facts of general and common knowledge. 29 Am.Jur.2d Evidence §§ 102-03 (1994). Certain tables were so readily recognized as reliable that courts identified them by name. See, e.g., Lincoln v. Power, 151 U.S. 436, 441, 14 S.Ct. 387, 38 L.Ed. 224 (1894) (“There is high authority for the proposition that courts can take judicial notice of the Carlisle Tables, and can use them in estimating the probable length of life, whether they were

introduced in evidence or not.”); Tollefson v. Ehlers, 252 Minn. 370, 377, 90 N.W.2d 205, 210 (1958) (accepting the American Experience Tables of Mortality); Hallada v. Great N. Ry., 244 Minn. 81, 95, 69 N.W.2d 673, 684-85 (1955) (accepting the U.S. Life Experience Table of Mortality), overruled on other grounds by Busch v. Busch Const., Inc., 262 N.W.2d 377 (Minn.1977); Thoirs v. Pounsford, 210 Minn. 462, 464, 466-67, 299 N.W. 16, 17-18 (1941) (accepting the American Experience Mortality Table). Presumably, American mortality tables include the life expectancies of immigrants who die in America.

Life expectancy figures from other countries may display few or none of these indicia of reliability. Countries that have no insurance industry, census, or other reliable data may not be able to provide accurate mortality information. Although the presence of the Liberian mortality information in the CIA World Factbook *13 might allow a court to conclude that it is authentic, i.e., it is the official Liberian information, it does not provide any information about how the information was compiled or why it should be regarded as reliable.

American mortality tables have the additional benefit of providing remaining life expectancy based on the age of the individual in question. “A mortality table consists of a schedule showing for each age the number of persons who die and the number who survive out of a known number under observation.” Henry Moir, Sources and Characteristics of the Principal Mortality Tables 1 (1932). A single life expectancy figure for an entire population, however, is considerably less meaningful. That figure will be skewed by the country’s infant mortality rate and understate the life expectancy of any individual who survives infancy. A single figure will be of no relevance to an individual who has already exceeded the life expectancy of a newborn. For these reasons, “tables which do not show the expectation at an age approximately that of the person in question should not be received.” 32A C.J.S. Evidence § 1007 (1996) (citing Decker v. McSorley, 111 Wis. 91, 86 N.W. 554 (1901)).

The standard jury instruction recognizes the need for tables that show the remaining life expectancy of a person- who is the age of the decedent. It provides:

According to life expectancy tables, the future life expectancy of a_-year old (male) (female) is _years. This means (he)(she) is expected to live to age-

4A Minn. Dist. Judges Ass’n, Minnesota Practice —Jury Instruction Guides, Civil, CIVJIG 91.85 (4th ed.1999) (emphasis added).

The Liberian life expectancy figure lacks any indicia of reliability, in general and as applied to Dekpah. First, it was not a life expectancy “table” but provides only a single life expectancy figure for all males in Liberia. Applied to Dekpah, the Liberian figure loses whatever probative value it may have had. He had outlived the stated expectancy by seven years. He had survived infant mortality. He was no longer endangered by Liberia’s internal conflicts. He did not, as far as the record reveals, carry any chronic diseases from Liberia. He had access to American health care. We fail to see how the figure could aid a jury in evaluating Dekpah’s life expectancy. We hold that its use in the jury instructions was error and we remand for a new trial on damages. 9

George invites us to hold that “absent convincing proof of relevance, life expectancy tables for residents of foreign countries should never be used to assist a jury in determining life expectancies of U.S. residents.” Although we are not prepared to rule that the district court may never take judicial notice of life expectancy data for a foreign country, we caution that such data should be received only where a foundational basis can be laid. That basis should include indicia of the general reliability of the data and a showing that the data provides a life expectancy for individuals approximately the age of the individual in question.

Affirmed in part, reversed in part, and remanded.