GALLAGHER, Senior Judge:
In this ease we answer the following two certified questions of law from the United States Court of Appeals for the District of Columbia Circuit:
Under District of Columbia law, may a plaintiff recover against a defendant who has negligently or recklessly destroyed or allowed to be destroyed evidence that would have assisted the plaintiff in pursuing a claim against a third party?
If a plaintiff may proceed under such a theory, what standard of proximate cause must he meet?
Certification of Question of Law, 113 F.3d 1285, No. 96-7182 (D.C.Cir.1997).
In response to the first certified question, we hold that negligent or reckless spoliation of evidence is an independent and actionable tort in the District of Columbia. In response to the second certified question, we hold that in order to demonstrate that the defendant’s actions proximately caused the harm alleged, plaintiff must show, on the basis of reasonable inferences derived from both existing and spoliated evidence, that (1) the plaintiff’s ability to prevail in the underlying lawsuit was significantly impaired due to the absence of the spoliated evidence; and (2) there had been a significant possibility of success in the underlying claim against the third party. We also set forth the required elements of this cause of action and establish an equitable system for calculating damages.
I.
On November 4, 1988, appellant Ronnie Holmes was driving a Chrysler Dodge 600 automobile rented to him by appellee Ame-rex Rent-A-Car when he was involved in a head-on collision with another automobile. Appellant alleges that the engine of the rented vehicle intruded into the passenger compartment causing severe and permanent injuries.
Following the collision, Amerex took possession of the wrecked automobile. Appel-' lant’s attorney made efforts to ensure that Amerex would hold the car for inspection, and on March 30, 1989, Amerex informed appellant’s attorney that the car would be held for sixty days. Amerex'later extended the deadline to June 15, 1989. On June 14, 1989, Amerex agreed to sell the car to appellant for $200.
On the same day, however, another agent of Amerex sold the car to a salvage yard. On June 15, 1989, the very next day, a body
shop engineer severed the front of the car and removed the engine. Appellant was informed of the sale and destruction of the wrecked car approximately one week later. According to an expert in accident reconstruction, severing of the car made it, “impossible to determine within a reasonable degree of certainty whether or not the vehicle had design, manufacturing and/or maintenance defects which proximately caused Mr. Holmes’ injuries.”
On November 2, 1991, appellant filed suit against Chrysler and Amerex in the District of Columbia Superior Court. The original complaint alleged,
inter alia,
that Chrysler had negligently designed the engine of the car, and that Amerex had negligently maintained the car. Chrysler removed the case to the United States District Court for the District of Columbia. On May 26, 1992, appellant voluntarily dismissed Chrysler from the lawsuit and filed an amended complaint naming Amerex as the sole defendant.
The amended complaint against Amerex dropped the negligent maintenance claim and replaced it with six other claims, including negligent spoliation of the evidence and tor-tious interference with a prospective civil action by spoliation of the evidence. Amerex answered the complaint and filed a motion for summary judgment.
On April 26, 1995, the District Court granted summary judgment in favor of Ame-rex with regard to appellant’s claims of,
inter alia,
negligent spoliation and tortious interference with a prospective civil action by spoliation of the evidence. The District Court held that if this court would recognize a cause of action for negligent or reckless spoliation of evidence, it would follow an approach to proximate causation that would not allow appellant to recover on the facts as presented.
Appellant filed an appeal to the United States Court of Appeals for the District of Columbia Circuit. That court decided that it could not “determine whether the district court’s entry of summary judgment was proper without knowing whether the District of Columbia would recognize such a claim and, if so, the standard of proximate cause that it would apply.” Certification of Question of Law, No. 96-7182,
supra.
Those two questions of law were certified to this court.
II.
The first issue for consideration is whether the District of Columbia will allow a plaintiff to recover against a defendant who has negligently or recklessly destroyed or allowed to be destroyed evidence that would have assisted the plaintiff in pursuing a claim against a third party. We hold that a plaintiff may pursue such an action under District of Columbia law.
A.
This is a case of first impression in the District of Columbia. However, we have twice dealt with a situation where a defendant negligently or recklessly destroyed evidence needed by the plaintiff in a civil suit between the two parties. In those situations, we held that “a fact-finder may be permitted to draw an adverse inference from the failure of a party to preserve evidence within his exclusive control.”
Williams v. Washington Hosp. Ctr.,
601 A.2d 28, 31 (D.C.1991) (citing
Battocchi v. Washington Hosp. Ctr.,
581 A.2d 759, 766-67 (D.C.1990)).
In doing so, we recognized that a plaintiff has a legally pro-tectable interest in the preservation of evidence required for securing recovery in a civil case.
B.
We think it reasonable to extend this protection to a situation where the spoliator is not a party to the underlying lawsuit. New torts are recognized when an interest requiring protection from unreasonable interference is identified. “The common thread woven into all torts is the idea of unreasonable interference with the interests of others.” W. Page Keeton, et al., PROSSER and Keaton on the Law of ToRTS § 1, at 6 (5th
ed.1984). A plaintiff advancing a novel claim in this jurisdiction therefore will not necessarily be precluded from recovering.
[I]f we are in one of the “open spaces” in the law of this jurisdiction we must fill it as well as we can, with a view to the social interests which seem to be involved and with such aid as we can get from authorities elsewhere and from “logic, and history, and custom, and utility, and the accepted standards of right conduct.” We cannot evade this duty; for unless we establish a right in the plaintiff we establish a privilege or immunity in the defendant. The fact that “the question is novel in this jurisdiction” does not mean that the plaintiff cannot recover.
Clark v. Associated Retail Credit Men of Washington, D.C.,
70 App. D.C. 183, 185, 105 F.2d 62, 64 (1939) (citations omitted).
Some remedy, therefore, should be available to those whose expectancy of recovery has been eliminated or severely hampered through the negligent or reckless acts of another. In the third-party defendant scenario presented in this case, however, the already recognized remedy of permitting an adverse inference against the spoliator would serve no purpose. “[Bjecause sanctions may not be levied upon a disinterested, independent third party, an independent tort action for negligent spoliation of evidence is the only means to deter the negligent destruction of evidence and to compensate the aggrieved party for its destruction.” John K. Stipan-cich, Comment,
The Negligent Spoliation of Evidence: An Independent Tort Action May Be the Only Acceptable Alternative,
53 Ohio St. L.J. 1135, 1141-42 (1992).
For these reasons, we are willing to provide independent legal protection against negligent or reckless spoliation of evidence. In recognizing the independent tort action, we do not exercise a legislative function; rather, we perform our duty to decide “a question of policy” well within “the framework of traditional and accepted negligence principles.”
Williams v. Baker,
572 A.2d 1062, 1072, 1073 (D.C.1990) (en bane) (citation omitted).
III.
The elements of this newly recognized tort must conform to the peculiar remedy being sought. To prevail in a tort action generally, a plaintiff must show by a preponderance of the evidence that the defendant breached a legally cognizable duty to the plaintiff, and that the breach was the proximate and legal cause of ascertainable damages to the plaintiff. Keeton,
supra
§ 30, at 164-65;
Powell v. District of Columbia,
634 A.2d 403, 406 (D.C.1993). The elements of an independent action for negligent or reckless spoliation of evidence, however, must be informed by a careful balancing of interests required by the unique features of this tort.
A. Duty and Breach
There is no general duty in the common law to preserve evidence in a third-party spoliation situation. “Absent some special relationship or duty rising by reason of an agreement, contract, statute, or other special circumstance, the general rule is that there is no duty to preserve possible evidence for another party to aid that other party in some future legal action against a third party.”
Koplin v. Rosel Well Perforators,
241 Kan. 206, 734 P.2d 1177, 1179 (1987). For a spoliation claim to succeed in negligence, therefore, the plaintiff must establish the existence of such a “special relationship” that creates a duty to preserve the evidence for use in the future litigation.
When negligence is the basis of the suit alleging an economic injury resulting from the destruction of evidence, a duty on behalf of the defendant arising from the relationship between the parties or some other special circumstance must exist in order for the cause of action to survive.
Id.; see Tomas v. Nationwide Mut. Ins. Co.,
79 Ohio App.3d 624, 607 N.E.2d 944, 947-48 (1992) (citing
Koplin,
supra);
Bondu v. Gurvich,
473 So.2d 1307, 1312 (Fla.Dist.Ct.App.1984) (holding that an action
for
negligent spoliation of evidence, like any negligence action, requires showing “the existence of a duty recognized by law requiring the defendant to conform to a certain standard of conduct for the protection of others including the plaintiff”).
For the purposes of this certified question, we are assuming that a duty of care existed which was derived from a contractual relationship to transfer ownership of the car wreckage from appellee to appellant, and that the duty was breached by appellee’s actions. Absent a duty arising from such a relationship, the cause of action would not lie.
B. Proximate Cause
The second certified question asks us to decide what standard of proximate cause must be met by the plaintiff to prevail under an independent action for negligent or reckless spoliation of evidence. We conclude that a plaintiff must show, based on reasonable inferences derived from both existing and spoliated evidence, that the underlying lawsuit was significantly impaired, that the spoliated evidence was material to that impairment and that the plaintiff enjoyed a significant possibility of success in the underlying claim.
Generally, proximate cause requires a “reasonably close causal connection between the conduct and the resulting injury.” Kee-ton, § 30, at 165. That standard does not adequately protect the interests of a plaintiff in a third-party spoliation claim, however, because the very purpose of an independent action for spoliation of evidence lies in the inability of the plaintiff to prove proximate causation to the proper degree of certainty required in the underlying suit.
To prevail under this tort action in standard negligence tort law, for example, appellant in this case would be required to show by a preponderance of the evidence that the destruction of the wrecked automobile actually resulted in the loss of a civil recovery. To make this required showing of causation, appellant would have to demonstrate by a preponderance of the evidence that he would have prevailed in the underlying products liability suit against Chrysler. Were he able to make such a showing, however, there would be no need for an independent action for spoliation.
We agree with the Illinois appellate court which decided that “it would be too heavy a burden on a plaintiff to show that he or she would have won with the missing evidence. Such a showing would be nearly impossible because judges or juries cannot evaluate the value of evidence that they cannot see.”
Petrik v. Monarch Printing Corp.,
150 Ill. App.3d 248, 103 Ill.Dec. 774, 784, 501 N.E.2d 1312, 1322 (1986). Further, there would be an inequity in preventing a plaintiff from recovering because of his inability, allegedly caused by the defendant, to prove his underlying case. “[T]he most elementary conceptions of justice and public policy require that the wrongdoer shall bear the risk of the uncertainty which his own wrong has created.”
Bigelow v. R.KO. Pictures, Inc.,
327 U.S. 251, 264-65, 66 S.Ct. 574, 580, 90 L.Ed. 652 (1946) (damages were uncertain).
A plaintiff, therefore, should not be required to make a showing of causation by a preponderance of the evidence in order to prevail under this new tort action. Such a standard would be both impractical and inequitable for this tort.
On the other hand, the policy underlying the requirement of proof of causation in negligence actions is still valid and should not be ignored to the detriment of the defendant. “Negligent conduct in itself is not such an interference with the interests of the world at large that there is any right to complain of it, or to be free from it, except in the case of some individual whose interests have suffered.” Keeton,
supra
§ 30, at 165. The defendant should not be forced to pay damages to a plaintiff who had only a frivolous underlying claim; some threshold showing of causation and damage is required. “[TJhere must, at the very least, be some proximate relationship between the failure of success in the underlying action and the unavailability of the destroyed evidence.”
Tomas, supra,
607 N.E.2d at 949.
These countervailing interests—the plaintiffs interest in recovery for lack of ability to prove the underlying case, and the defendant’s interest in only compensating plaintiff for defendant’s harmful negligence or recklessness—must be balanced in determining the standard of proof for causation in an independent action for negligent or reckless spoliation of evidence.
This adds a unique characteristic to the tort. Not only must the plaintiff show that an expectancy of recovery was harmed, but also that such an expectancy realistically existed. Specifically, proximate cause must include two showings. First, it must be shown that the defendant’s actions proximately caused some level of impairment in the plaintiff’s ability to prove an existing underlying civil claim. Second, in order to show that defendant’s actions proximately caused any damages, it must be shown that plaintiff’s underlying claim was, at some threshold level, meritorious.
Appellee urges us to adopt the approach to proximate cause established in Ohio. With regard to the first prong of proximate cause, impairment of ability to advance the underlying claim, the Ohio approach requires the plaintiff to either attempt and lose the underlying case or demonstrate that the underlying claim is precluded.
Tomas, supra,
607 N.E.2d at 948;
see Petrik, supra,
103 Ill.Dec. at 779, 501 N.E.2d at 1317 (“a spoliation claim could not be brought until after the underlying lawsuit was lost”). The plaintiff’s expectancy must be eliminated in order to make a threshold showing of causation.
With regard to the second prong of proximate cause, the merit of the underlying claim, this approach requires the plaintiff in the spoliation suit to show that it was more probable than not that the plaintiff would have succeeded in the underlying suit.
Tomas, supra,
607 N.E.2d at 948. “Plaintiff’s burden is to present evidence from which it reasonably could be inferred that it is more probable” than not that the underlying suit would have been successful.
Id.
A plaintiff must also show that the destruction of the particular evidence had an impact on the merits of the underlying case.
[T]here must, at the very least, be some proximate relationship between the failure of success in the underlying action and the unavailability of the destroyed evi-dence_ Speculation based upon possibility is too tender a reed upon which to base a claim for relief.
Id.
at 949-50. This approach most closely tracks traditional tort principles by requiring the plaintiff to make the standard showing of proximate causation by a preponderance of the evidence.
The Ohio approach is effective in that it sets a high standard for protecting the interests of the defendant. Certain parts of this approach are appealing. Specifically, this approach requires the plaintiff to show a nexus between the destroyed evidence and the impairment or preclusion of the underlying suit.
However, we are unpersuaded that this approach adequately considers the interests of the plaintiff. As stated earlier, were a plaintiff able to make a showing of success in the underlying suit by a preponderance of the evidence there would be no need for an independent action for spoliation. The tort is designed to help plaintiffs who are unable to prove the underlying case to the ordinary standard of proof, so requiring plaintiffs to do just that in pursuing a spoliation claim is illogical. The Ohio approach sets the standard of proximate cause too high and greatly weakens the tort it is attempting to recognize. Furthermore, the requirement that a plaintiff either pursue and lose the underlying claim or demonstrate that the underlying claim is precluded is too harsh. Requiring plaintiffs to pursue futile lawsuits to completion and withholding relief from those plaintiffs whose lawsuits have, been severely hampered, but not precluded, by spoliation of evidence ignores the plaintiff’s interest in securing a reasonable recovery for a lost or impaired expectancy.
Appellant, on the other hand, urges us to follow the precedents set in California and Florida. In California, a plaintiff must show a “reasonable probability” that the plaintiff would have successfully litigated the underlying case if the spoliated evidence were available.
Smith v. Superior Court,
151 Cal. App.3d 491,198 Cal.Rptr. 829, 835-36 (1984). In addition, the California courts have identified several factors, originally delineated for the tort of negligent interference with prospective economic advantage, as relevant in determining whether a plaintiff has stated a cause of action for negligent spoliation:
(1) the extent to which the transaction was intended to affect the plaintiff, (2) the foreseeability of harm to the plaintiff, (3) the
degree of certainty that the plaintiff suffered injury, (4) the closeness of the connection between the defendant’s conduct and the injury suffered, (5) the moral blame attached to the defendant’s conduct and (6) the policy of preventing future harm.
Velasco v. Commercial Bldg. Maintenance Co.,
169 Cal.App.3d 874, 215 Cal.Rptr. 504, 506 (1985) (quoting
J’Aire Corp. v. Gregory,
24 Cal.3d 799, 157 Cal.Rptr. 407, 410, 598 P.2d 60, 63 (1979)). In California, however, there is no clear requirement that the spoliation of evidence impaired the plaintiff’s ability to succeed in the underlying lawsuit.
The Florida courts have dispensed with the requirement of demonstrating that the underlying suit was somewhat meritorious, and only require that the plaintiff show a “significant impairment” in the plaintiffs ability to successfully prosecute the underlying case. In Florida, the elements of a cause of action for negligent spoliation of evidence are:
(1) existence of a potential civil action, (2) a legal or contractual duty to preserve evidence which is relevant to the potential civil action, (3) destruction of that evidence, (4) significant impairment in the ability to prove the lawsuit, (5) a causal relationship between the evidence destruction and the inability to prove the lawsuit, and (6) damages.
Continental Ins. Co. v. Herman,
576 So.2d 313, 315 (Fla.Dist.Ct.App.1990).
The California approach focuses on the second prong of proximate cause, that the underlying claim was meritorious, to the exclusion of the first prong, that the underlying claim be impaired by the spoliation. The Florida approach does the opposite. In this respect, both of these approaches are deficient. Taken together, however, they create an effective approach to the tort. A plaintiff must demonstrate that the underlying claim was “significantly impaired” by the spoliation, and that the underlying claim had a “reasonable probability” of success.
We take issue, however, with the California court’s terminology. A “reasonable probability” of success still requires a showing that success was “probable.” In our view, this comes too close to requiring that the plaintiff show success by what amounts to “a preponderance of the evidence.” As discussed above, this standard is too high. Instead, we believe we should require that the plaintiff show that the underlying lawsuit enjoyed a significant possibility of success. We choose this term as it implies a showing higher than the already recognized standard of “significant evidence” but lower than the standard of “preponderance of the evidence.” The plaintiff must demonstrate a substantial and realistic possibility of succeeding, but need not cross the threshold of demonstrating that such success was more likely than not, something that would be realistically impractical of proof. We also adopt the Ohio concept of requiring a nexus between the spoliated evidence and the impairment of the underlying lawsuit.
We hold, therefore, that in order to demonstrate causation, a plaintiff must demonstrate that (1) the underlying claim was significantly impaired due to the spoliation of evidence; (2) a proximate relationship exists between the projected failure of success in the underlying action and the unavailability of the destroyed evidence; and (3) the underlying lawsuit would enjoy a significant possibility of success if the spoliated evidence were still in existence.
C. Damages
“The most difficult aspect of a spoliation of evidence tort is the calculation of damages.”
Petrik, supra,
103 Ill.Dee. at 782, 501 N.E.2d at 1320. This is because the tort does not allow for standard calculations of damages to the proper degree of certainty. Generally, to prove damages, “a plaintiff must not only establish the fact of damages with reasonable certainty, but must also prove the amount of damages with reasonable certainty.”
Smith, supra,
198 Cal.Rptr. at 835.
In the case of spoliated evidence, however, “It would seem to be sheer guesswork, even presuming that the destroyed evidence went against the spoliator, to calculate what it would have contributed to the plaintiff’s success on the merits of the underlying lawsuit.”
Petrik, supra,
103 Ill.Dec. at 782, 501 N.E.2d at 1820. While this might seem to preclude the tort action, the Supreme Court has noted that the speculative nature of damages should not necessarily operate to bar recovery altogether.
Where the tort itself is of such a nature as to preclude the ascertainment of the amount of damages with certainty, it would be a perversion of fundamental principles of justice to deny all relief to the injured person, and thereby relieve the wrongdoer from making any amend for his act.
Story Parchment Co. v. Paterson Parchment Paper Co.,
282 U.S. 555, 563, 51 S.Ct. 248, 250, 75 L.Ed. 544 (1931).
There remains a question of how to calculate uncertain damages. In such situations, “the jury may make a just and reasonable estimate of damage based on relevant data.”
Bigelow, supra,
327 U.S. at 264-65, 66 S.Ct. at 580;
Edmund J. Flynn Co. v. LaVay,
431 A.2d 543, 550 (D.C.1981). Mathematical certainty is not required.
Flynn, supra,
431 A.2d at 550. Such estimation is essential to calculation of uncertain damages. However, even a “just and reasonable estimate” is hobbled by the lack of information presented to a spoliation claim jury. There are several possible solutions.
One possibility is to award the plaintiff the entire amount of damages that the plaintiff would have received if the original lawsuit had been pursued successfully. 3 JEROME H. Nates,
et al.,
Damages in ToRT Actions § 19.33(3) (1997);
see Petrik, supra,
103 Ill.Dec. at 782-83, 501 N.E.2d at 1320-21. The problem with this approach, however, is that it ignores defendant’s interest in not providing the plaintiff with a windfall recovery. “If this method is utilized, there is the potential that the plaintiff would benefit more in an instance of spoliation than he might have in the underlying suit.” Nates,
supra,
§ 19.33(3).
A second possibility is only to award the plaintiff any costs and fees incurred in pursuit of the original suit. This method, however, fails to “adequately ‘punish’ the offending parties or adequately compensate the party who is now apparently unable to pursue its cause of action.” Nates,
supra,
§ 19.33(3). Under this approach, the plaintiff’s interest in securing the precluded recovery is ignored.
Another possibility, one that attempts a balancing of interests in light of the inherent uncertainties of proof relevant to this tort, was suggested by an Illinois appellate court. This approach requires that damages be discounted to account for uncertainties.
One possible measure of damages, therefore, could be the damages that would have been obtained in the underlying lawsuit, multiplied by the probability that plaintiff would have won the suit had he had the spoliated evidence.
Petrik, supra,
103 Ill.DeC. at 782, 501 N.E.2d at 1320.
This compromise system would apportion risk between the two parties in an equitable fashion. On the one hand, the plaintiff’s interest in recovering the expected but precluded sum is protected because recovery is allowed with lower standards of proof for causation and damages. On the other hand, the defendant’s interest in only compensating a plaintiff for actual loss is protected because the recovery will be lessened by the uncertainties involved. Both parties, then, accept some of the risk of windfall or shortage necessitated by the uncertainty inherent in proving this tort.
The problem with this method is in the difficulty of “proving what the spoliated evidence would have demonstrated and the extent to which it would have changed the outcome.” Nates,
supra,
§ 19.33(3). However, that problem is endemic to the tort of spoliation, and we are unconvinced that the problem is sufficient to overcome the overall fairness of the discounted damages approach.
We prefer to adopt the method suggested by the Illinois court in
Petrik, supra,
103 Ill.Dec. at 782, 501 N.E.2d at 1320, and hold that in an action for negligent or reckless spoliation of evidence, damages arrived at through just and reasonable estimation based on relevant data should be multiplied by the probability that the plaintiff would have won the underlying suit had the spoliat-ed evidence been available. For example, hypothetically, if a jury determined that the expected recovery in an underlying suit was $200,000, and that there was an estimated
sixty percent probability that a plaintiff would have recovered that amount if the underlying suit had not been impaired or precluded due to the spoliated evidence, then the award of damages would be $120,000, or sixty percent of $200,000.
To illustrate the practical application of this damages assessment method, we consider the methods used to apportion fault in comparative negligence jurisdictions. One popular method involves the following steps:
First, determine the amount of plaintiffs damages without reference to the question of his negligence. Second, decide what proportion or percentage of the total combined negligence of both parties causing plaintiffs injury consisted of plaintiffs negligence. Finally, reduce by this proportion or percentage the amount of damages you [the jury] would have awarded plaintiff had plaintiffs negligence not' contributed to his injury.
HenRY Woods and Beth DeeRE, Comparative Fault § 20:2 (3d ed. 1996 & Supp.1997); see 45 U.S.C.A. § 53 (1997); Comparative Negligence Manual § 13:17-20 (3d ed. 1995 & Supp.1997). In applying this method to the spoliation tort, however, the apportionment of percentages must be adjusted to reflect the likelihood of success of the underlying action and not the proportional fault of the parties. This is especially important as the District of Columbia remains a contributory negligence jurisdiction.
The jury must be instructed in accordance with the apportionment principles we have discussed.
IV. Conclusion
In response to the first certified question, we hold that there now exists in the District of Columbia an independent action for negligent or reckless spoliation of evidence. In response to the second certified question, we hold that in order to demonstrate proximate causation, a plaintiff must show that the plaintiffs ability to prove the underlying lawsuit was significantly impaired due to the absence of the spoliated evidence, and that tile plaintiff had a significant possibility of succeeding in the underlying lawsuit.
We further hold, in light of the unique considerations surrounding this tort, that the elements of a cause of action for negligent or reckless spoliation of evidence are: (1) existence of a potential civil action; (2) a legal or contractual duty to preserve evidence which is relevant to that action; (3) destruction of that evidence by the duty-bound defendant; (4) significant impairment in the ability to prove the potential civil action; (5) a proximate relationship between the impairment of the underlying suit and the unavailability of the destroyed evidence; (6) a significant possibility of success of the potential civil action if the evidence were available; and (7) damages adjusted for the estimated likelihood of success in the potential civil action.
The clerk shall certify this answer to the United States Court of Appeals for the District of Columbia Circuit.
So ordered.