Appellant landlord seeks reversal of the judgment entered in favor of the appellee (plaintiff) in an action for unlawful eviction. He argues that the complaint failed to state a cause of action inasmuch as in evicting appellee, he was simply exercising the right of a landlord to use self-help to evict a tenant, a right recognized at common law and, according to appellant, still viable under the case law of this jurisdiction. Resolution of this question turns on whether or not the various statutory remedies for reacquiring possession have abrogated the common law right of self-help. In Section I, we hold that those statutory remedies are exclusive (overruling our prior decision to the contrary) and that an action will lie in tort against a landlord who evicts a tenant without legal process. In Section II, we analyze the issue of prospective versus retroactive application, and give this overruling decision partially retroactive application. In Section III, we treat the issue of punitive damages. We affirm the judgment of the trial court except as to the award of punitive damages.

Appellee was a tenant by the month in a house owned and managed by appellant. He evicted her by packing up and removing all her possessions. Appellee and her daughter had to pay for a motel before finding other lodging. The complaint *783sought exemplary or punitive damages in addition to actual damages of more than one thousand dollars. The trial court awarded Ms. Johnson one hundred dollars for actual damage to and loss of furniture, and three hundred dollars in punitive damages. This appeal followed.

I.

Appellant argued at trial that the complaint should be dismissed for failure to state a cause of action. He claimed, as noted above, that in the District of Columbia, the common law right of a landlord lawfully entitled to possession1 to evict a tenant by effecting a peaceful2 reentry survived the enactment of specific statutory remedies for reacquiring possession. Two cases, Burford v. Krause, 89 F.Supp. 818 (D.D.C. 1950), and Snitman v. Goodman, D.C.Mun.App., 118 A.2d 394 (1955), were cited in support of this position. The trial court rejected this view, adopting instead the reasoning of Chief Judge Greene as set forth in Wheeler v. Thompson, 98 D.W.L.R. 41 (D.C.Gen.Sess., L&T No. 103875-69, Jan. 9, 1970). Before turning to these authorities, an outline of the historical background of this dispute may be helpful.

A.

Under early common law, a landlord was privileged to enter upon his land and recover it by force, using violence if necessary. However, this privilege was modified as early as 1381, when a statute was passed making such forcible entry a criminal offense.3 This criminal statute was accepted as part of the common law, or reenacted, by nearly all of our states.4 In addition, most states provided a summary procedure whereby a landlord could quickly reacquire possession from a defaulting tenant with the aid of judicial process.5

*784These “forcible entry and detainer” statutes, both civil and criminal, have been variously interpreted. In England, the Statute of Richard II was held not to provide any basis for a civil action for trespass to the land, since a tenant, having no rightful claim to the land, could not be injured by its deprivation; the landlord would be liable for criminal prosecution, but the tenant had no claim for damages. However, in Newton v. Harland, 1 Man. & G. 644, 133 Eng.Rep. 490 (1840), it was held that an action for assault and battery would lie when force was used upon the person of the occupant in the course of the eviction. Thus a distinction was developed between forcible entry, which was a criminal offense but not a civil wrong against one wrongfully in possession, and forcible eviction, which was actionable civilly as an invasion of the occupant’s interests in bodily security and exclusive custody and control of his chattels.6 Before being overturned in England in 1920 (Hemmings v. Stoke Poges Golf Club, 1 K.B. 720 (1920)), this view gained a wide following among the states. The current English view is that the privilege extends to the use of reasonable force both to reenter the premises and evict the occupant and his possessions.7

American jurisdictions have been neither consistent nor uniform in their approach to these issues.8 Thus some states still adhere to the current English view that the landlord may use reasonable force to evict the tenant and his possessions without being civilly liable.9 Some jurisdictions grant the landlord the right to reenter if he can do so peaceably.10 A large number of states have adopted the rule that the landlord must in all cases, resort to the courts to dispossess a tenant or be liable to the tenant for forcible eviction.11

*785These varying results are, in effect, reflections of the judicial response to two separate questions: (1) does the criminal forcible entry and detainer statute provide a basis for a civil action as well; and (2) is the summary possessory action created by statute an exclusive remedy. It is the second inquiry which directly concerns us here. As noted, appellant’s view is that under the case law of the District of Columbia, the statutory remedy is merely an alternative, and the self-help doctrine has been abrogated only to the extent of requiring that it be exercised without violence. Because the entry and eviction here were accomplished peaceably, appellant claims that he may not be held liable in tort, simply for failure to resort to judicial process.

B.

Although the question has arisen infrequently, appellant’s position is not without support in the case law of this jurisdiction. The first case squarely to address the issue was Burford v. Krause, supra, which involved a tenant of commercial property who sought to recover compensatory and punitive damages from his landlord who reentered the premises after breaking a lock and removed the plaintiff’s possessions to a warehouse. The court noted that the common law rule permitting the use of “any force necessary to expel the tenant” had been “qualified to prevent physical violence, force and breach of the peace.” Id. at 819. Nonetheless, it concluded that:

The presence of the local forcible entry and detainer statutes does not supplant the common law right of self help in peaceably entering into premises on which a lease has expired where such entry is accomplished without breach of the peace. They merely afford additional means to which the parties may resort and under some circumstances may prove more orderly and less costly than the method employed in the case herein. [Id. at 820.]

The Burford case was relied upon by our Municipal Court of Appeals five years later in Snitman v. Goodman, supra, which affirmed a directed verdict on behalf of the defendants in a suit brought by the former tenant of a parking lot to recover possession thereof. In considering whether the legislatively created summary proceeding was the exclusive remedy, the court quoted at length from Burford, and then continued:

In Lindner v. District of Columbia, D.C. Mun.App., 32 A.3d 540, 543, we said that “unless so declared in express terms or by necessary implication, a statutory remedy is merely cumulative and does not abolish an existing common law remedy.” There is no express provision in the statute under consideration which abolishes the landlord’s common law right and we see no necessary implication to that effect. [118 A.2d at 397.]

A second basis for the Snitman ruling was the fact that the criminal forcible entry and detainer statute forbids

(1) forcible entry and (2) unlawful detention by force although entry was without force. It appears to us that this statute recognizes that one who is lawfully entitled to possession has a right of entry if he can make such entry without the use of force. [118 A.2d at 397-98 (emphasis in original).]

However, more recent decisions suggest that the Snitman approach is not widely favored even in this jurisdiction. Thus the circuit court in an action brought by a commercial tenant for wrongful eviction, trespass and breach of contract observed that punitive damages might be allowed for trespass and wrongful eviction where the landlord engaged in “the dark-of-night lock-changing method of eviction . instead of invocation of the process of the law for that purpose.” Camalier & Buckley-Madison, Inc. v. Madison Hotel, Inc., 168 U.S.App.D.C. 149, 162 n. 93, 513 F.2d 407, 420 n. 93 (1975).

*786The only reported decision addressing the right of self-help in the residential context is Wheeler v. Thompson, supra, upon which the trial court relied in the instant case. The tenant, who had been forcibly evicted from her apartment by the landlord, sought an order restoring her to the premises and protecting her possession from further disturbance except through normal court processes. The landlord, relying on Snit-man v. Goodman, supra, argued that he was not limited to court action to dispossess a tenant. Unable to overrule Snitman, the trial court chose to distinguish it factually from the case before it, and held that for public policy reasons, a landlord of residential property could not rely on self-help to evict his tenant. We find Chief Judge Greene’s reasoning persuasive.

It is true, of course, as the Snitman court observed, that a statutory remedy is, as a rule, merely cumulative and does not abolish an existing common law remedy unless so declared in express terms or by necessary implication. However, we think that the summary procedure provided by Congress12 did, by necessary implication, abolish the common law right of self-help. We reach this conclusion for a number of reasons.

First of all, the salutary purposes of the statute would be totally defeated by acceptance of appellant’s view. One of the motivations for providing a summary pos-sessory action was “to avoid resort to self-help and force, condoned at common law.” Tutt v. Doby, 148 U.S.App.D.C. 171, 174, 459 F.2d 1195, 1198 (1972).13 Yet, as this and earlier cases have demonstrated, resort to self-help will be avoided or obviated only if the statutory remedy is deemed to be exclusive. Since it cannot be presumed that the Congress would do a futile thing, it must have intended to abrogate fully the common law.14

Moreover, considerations of the public interest would compel this interpretation even if logic and statutory interpretation did not. To sanction the use of self-help in our densely populated city, chronically plagued with serious housing shortages, would be to invite and sanction violence. This consideration alone has prompted many other jurisdictions to hold that a landlord must in all cases resort to the courts to dispossess a tenant.15

Acceptance of appellant’s view would also deprive the tenant of his opportunity to assert various equitable defenses and be accorded certain rights mandated by the statutes and case law of this jurisdiction.16 We note in this regard that landlord-tenant law has changed substantially in the intervening years since Snitman.

More general grounds of public policy also support this result. As Chief Judge Greene observed in Wheeler v. Thompson, supra:

Those who, for one reason or another, have grievances against society are often *787told that they must seek redress through established political, legislative, and judicial processes, notwithstanding their impatience with what they regard as a lack of perfection in these processes and a lack of sufficient speed in their operations. That is as it should be. Self-help, whether it takes the form of the seizure of buildings, the blocking of streets, personal violence and disruption, or any other acts which interfere with the rights of others, are justly condemned by a law-abiding society in which legal and political safeguards against oppression and abuse exist. . . . This feudal doctrine was not found to be an unalterable obstacle by the courts in those states which have abandoned the so-called common law rule; there is no reason why that rule should prove to be immutable here. [Slip op. at 9-10 (footnote omitted).]

We therefore conclude and hold that in this jurisdiction, the landlord’s common law right of self-help has been abrogated, and the legislatively created remedies for reacquiring possession are exclusive. A tenant has a right not to have his or her possession interfered with except by lawful process, and violation of that right gives rise to a cause of action in tort. Snitman v. Goodman, supra, is hereby overruled.

II.

We now consider an important corollary to the central issue of the legality of landlords’ self-help eviction remedy presented in this case — whether the new rule of law announced herein will be prospective only, or retrospective to some degree, in application.

The question of the retroactive effect to be accorded judicial decisions announcing new rules of law has plagued legal commentators since at least the early eighteenth century. Blackstone emerged in the early nineteenth century as the leading proponent of the “declaratory theory” of the common law which shaped the contours of Anglo-American jurisprudence on this subject until modern times.17 Fundamental to this theory was the notion of the retroactivity as well as prospectivity of all judicial decisions. According to Blackstone, adjudication consisted of a search for the right or true rules of law. In accordance with the thesis that they had always been the law, new legal rules established by decisions overruling precedent, considered discoveries of the true law of the case, were related back to and declared operative at the time the controversies in the overruling cases themselves, and in other cases pending before the court, arose — regardless of the status of the law at their inception. Overruled precedents, denominated failures at discovery of the true law, were deemed, once overturned, to have never existed. Consequently, new rules were not considered really new, but were actually further clarifications of the old — thereby maintaining continuity and consistency in the common law. Thus, as discoverers rather than creators of the law, the function of judges was to “find” the true principles of law and to “declare” them as controlling. As indicated above, the rule of retroactivity of overruling decisions stated by Blackstone became firmly implanted in the common law. See Note, Prospective Overruling and Retroactive Appiication in the Federal Courts, 71 Yale L.J. 907 (1962).

Antithetical to the Blackstonian retroac-tivity doctrine was the prospectivity theory expounded by Austin. This theory conceived of judges not as mere discoverers but as active creators of the law who, through the mechanism of judicial interpretation, imparted meaning to statutory and common law. The overruling of precedent was viewed as integral to the dynamic process of redefinition and reformation that advanced the evolution of the law. The Austrian school advocated recognition of overruled precedents as simply erroneous deci*788sions derived from controlling principles since determined to be fallacious. It rejected the employment of the legal fiction of Blackstone’s “relation back” doctrine, maintaining that prior rules of law, although discredited, continued to control in all intermediate cases decided until the point of their overruling. Because such rules were not “extinguished” by the mere declaration of new governing principles, but rather retained their status as controlling, the establishment of new legal principles did not mandate the disturbance of settled results reached thereunder.18

The Supreme Court, in addressing this question, has accorded the lower federal and state courts wide latitude to employ such approaches as deemed appropriate in formulating responses to the retrospectivity-prospectivity problem. In a landmark decision on the subject, Great Northern Ry. Co. v. Sunburst Oil and Refining Co., 287 U.S. 358, 53 S.Ct. 145, 77 L.Ed. 360 (1932), the Court examined the constitutional dimensions of retroactive as well as prospective application of overruling decisions. The Court declared that retrospectivity is neither required nor prohibited by the Constitution:

We think the federal constitution has no voice upon the subject. A state in defining the limits of adherence to precedent may make a choice for itself between the principle of forward operation and that of relation backward. It may say that decisions of its highest court; though later overruled, are law nonetheless for intermediate transactions. .

On the other hand, it may hold to the ancient dogma that the law declared by its courts had a Platonic or ideal existence before the act of declaration, in which event the discredited declaration will be viewed as if it had never been, and the reconsidered declaration as law from the beginning. [287 U.S. at 364-65, 53 S.Ct. at 148.]

Because it provided an overly simplistic and mechanical solution to a complex problem, adherence to the traditional Blacksto-nian precept of unlimited retroactivity of overruling decisions has been gradually eroded and no longer prevails. Incorporating the basic philosophy of the Austinian theory, contemporary courts have developed a more sophisticated approach to the retro-activity versus prospectivity problem premised on the recognition that no singular definitive formula can automatically dictate the retrospective or prospective effect to be given an overruling decision in any given context.

The abandonment of a rigid, formalistic dogma in favor of the adoption of a more flexible, pragmatic approach to this question has been sanctioned by the Supreme Court. The Court provided guidance on the issue in Linkletter v. Walker, 381 U.S. 618, 85 S.Ct. 1731, 14 L.Ed.2d 601 (1965). In deciding that the rule of Mapp v. Ohio, 367 U.S. 643, 81 S.Ct. 1684, 6 L.Ed.2d 1081 (1961), was not to apply retroactively in collateral attacks on state court convictions which had become final before the issuance of the Mapp decision, the Supreme Court reviewed its pronouncements on the retro-activity problem. Rejecting in dicta the notion that different rules of retroactivity should apply in civil and criminal cases, the Court ruled:

Once the premise is accepted that we are neither required to apply, nor prohibited from applying, a decision retrospectively, we must then weigh the merits and demerits in each case by looking to the prior history of the rule in question, its purpose and effect, and whether retrospective operation will further or retard its operation. [381 U.S. at 629, 85 S.Ct. at 1738.]

Hence, the extent of retroactive application, if any, of an overruling decision should be determined by the courts as a matter of judicial policy, requiring analysis within the general framework of the above-stated factors on an individualized, case-by-*789case basis. See Annot., 10 A.L.R.3d 1371, 1384 (1966).

Beyond setting forth the broad guiding principles of Linkletter, the Supreme Court has continued to leave to other courts resolution of the question of the specific application of a new rule of law in a particular case. An examination of the relevant case law reveals that the courts have developed three general formulae for application of an overruling decision: total retroactive as well as prospective application,19 partial retroactive application (i. e., retroactive application of the new rule to the parties to the case in addition to prospective application),20 and purely prospective application.21 See generally Annot., 10 A.L.R.3d 1371. Several specific criteria have emerged as central in the courts’ determination of which of these techniques of overruling to utilize. These include (1) the extent of the reliance of the parties on the old rule (including the degree of justifiable reliance and the hardship which might result to the litigants as a result of retrospective application); (2) avoidance of altering vested contract or property rights; (3) the desire to reward plaintiffs who seek to initiate just changes in the law; and (4) the fear of burdening the administration of justice by disturbing decisions reached under the overruled precedent.22 Each of these factors will be discussed briefly below.

1. Reliance. The degree of the reliance by the litigants before the court, and in some cases by the public at large, on the legitimacy of the prior rule has played a major role in courts’ analysis of the retroac-tivity issue. Where retroactive application of a new rule would result in substantial disruption of settled transactions and/or injustice to a party because of reliance on the continued validity of the prior legal rule— especially one of long standing — courts are extremely reluctant to accord retroactive effect to overruling decisions. See, e. g., Westbrook v. Mihaly, 2 Cal.3d 765, 471 P.2d 487, 89 Cal.Rptr. 839 (1970) (applying new rule requiring mere majority approval by popular referendum for certification of local bond issues prospectively only because of reliance on the prior two-thirds majority requirement); Lester v. McFaddon, 415 F.2d 1101 (4th Cir. 1969) (applying new rule abolishing federal court jurisdiction in certain wrongful death actions prospectively only since the plaintiff sued in reliance on precedent to the contrary); State v. Martin, *79062 Wash.2d 645, 384 P.2d 833 (1963) (applying new rule that issuance and sale of limited obligation bonds funded by excise taxes on sales and handling of cigarettes contravened state constitutional provision limiting state debt prospectively only due to “massive” reliance on previous rule).23

In addition to the mere fact of reliance by a party on the former rule, courts also consider the reasonableness of any such reliance. Thus, where a party should have known that the old rule was about to be changed, because of either judicial or legislative intimations to that effect, many courts refuse to regard reliance on precedent as a bar to application of the new rule to the parties before the court. For example, the New Jersey court in Wangler v. Harvey, 41 N.J. 277, 196 A.2d 513 (1963), abolished a rule granting immunity to nonresidents from local service of process and applied the new rule to the litigants at bar. In doing so, the court held that indications in prior opinions of judicial dissatisfaction with the prior rule negated a claim by the nonresident party that his reliance on that rule should preclude partial retroactive application of the new rule. See also In re Marriage of Brown, 15 Cal.3d 838, 544 P.2d 561, 126 Cal.Rptr. 633 (1976); Smith v. State, 93 Idaho 795, 473 P.2d 937 (1970); Flemming v. South Carolina Electric & Gas Co., 239 F.2d 277 (4th Cir. 1956).

A further aspect of the factor of reliance on past law which figures largely in courts’ examination of the retroactivity issue is the degree of hardship that the parties before the court, and others in general, may sustain as a result of retroactive application. This is especially true where such application of a new rule may have a significant adverse effect on a sovereign’s potential financial liabilities. See, e. g., Smith v. State, supra (applying new rule abolishing tort immunity of state proprietary activities to parties at bar, but otherwise prospectively). Accord, Carroll v. Kittle, 203 Kan. 841, 457 P.2d 21 (1969); Molitor v. Kaneland Community Unit District No. 302, 18 Ill.2d 11, 163 N.E.2d 89 (1959) (applying new rule abolishing school district tort immunity to parties at bar and otherwise prospectively).

2. Avoidance of altering property or contract rights. In conjunction with the reliance factor, courts also weigh the probable effect of a law change on the vested property or contract rights of the parties. When such rights have accrued by operation of the past law, and would be forfeited by retroactive application of the new rule, courts tend to overrule prospectively only. See, e. g., Kelly Adjustment Co. v. Boyd, D.C.App., 342 A.2d 361 (1975); Lasher v. Commonwealth, 418 S.W.2d 416 (Ky. 1967); Pabon v. Hackensack Auto Sales, Inc., 63 N.J.Super. 476, 164 A.2d 773 (1960); World Fire & Marine Ins. Co. v. Tapp, 279 Ky. 248, 130 S.W.2d 848 (1939); Payne v. City of Covington, 276 Ky. 380, 123 S.W.2d 1045 (1938).

*7913. Rewards to plaintiffs who seek to change the law. In several decisions which apply the new rule of law prospectively, but also apply the rule retroactively to the parties to the overruling case, courts express a desire to benefit the litigant who sought to move the law forward. In Molitor v. Kaneland Community Unit District No. 302, supra, the court stated in this regard:

[T]o refuse to apply the new rule here would deprive appellant of any benefit from his effort and expense in challenging the old rule which we now declare erroneous. Thus there would be no incentive to appeal the upholding of precedent since appellant could not in any event benefit from a reversal invalidating it. [163 N.E.2d at 97.]

Accord, Commonwealth v. Jones, 457 Pa. 563, 319 A.2d 142, cert. denied, 419 U.S. 1000, 95 S.Ct. 316, 42 L.Ed.2d 274 (1974); Holytz v. Milwaukee, 17 Wis.2d 26, 115 N.W.2d 618 (1962); Kojis v. Doctors Hospital, 12 Wis.2d 367, 107 N.W.2d 292 (1961); Parker v. Port Huron Hospital, 361 Mich. 1, 105 N.W.2d 1 (1960); Lasher v. Commonwealth, supra. See also Annot., 10 A.L.R.3d at 1399-1401.

Of particular import in some courts’ analysis of this factor is the status of the plaintiff as an individual or “institutional” litigant. See, e. g., Simpson v. Union Oil Co., 411 F.2d 897 (9th Cir. 1969). If the plaintiff is an individual person, or a small business entity, who, as a “one-time litigant,” will receive no other future benefit from a change in the law if the new rule adopted is not applied to the parties at bar, courts are more inclined to overrule retroactively as to the parties to the overruling case in order to reward the litigant for efforts expended in promoting the progress of the law. However, if the plaintiff is instead an “institutional litigant,” such as an insurance company or a public utility regularly defending damage claims, a manufacturer regularly defending products liability claims, or a major corporation regularly defending antitrust claims, the litigant need not be rewarded for such advocacy since such a party will continue to reap the benefits of a new rule of law in succeeding cases. Id. at 903-04. See also 71 Yale L.J. at 945 n. 192.

4. Fear of burdening the administration of justice with retroactive changes in the law. Operative in criminal as well as civil cases, this factor has assumed great importance in determining the degree of retroac-tivity of overruling decisions. Courts are acutely aware of the potential impact on the judicial system that retrospective application of such decisions, requiring relitigation of cases already decided under prior rules, would have and, consequently, have been reluctant to disturb settled cases. Thus, in the criminal and civil contexts, courts tend to balance the importance of the rights or interests at issue with the burden on the system of administering the new rule retroactively. See Simpson v. Union Oil Co., supra at 902-04. For example, it has been stated that a new rule of criminal law should be applied retroactively where the subject of the rule is a right so fundamental that it implicates the integrity of the fact-finding process and the basic fairness of the trial and thus necessarily casts doubt on results reached under the old rule. E. g., United States v. LaVallee, 330 F.2d 303 (2d Cir.), cert. denied, 377 U.S. 998, 84 S.Ct. 1921, 12 L.Ed.2d 1048 (1964) (applying right to counsel holding in Gideon v. Wainwright, 372 U.S. 335, 83 S.Ct. 792, 9 L.Ed.2d 799 (1963), retroactively). However, where a criminal rule has as its basic purpose deterrence of illegal police action, retroactive application would not serve the objective of the rule. E. g., Linkletter v. Walker, supra, and United States ex rel. Angelet v. Fay, 333 F.2d 12 (2d Cir. 1964), aff’d sub nom. Angelet v. Fay, 381 U.S. 654, 85 S.Ct. 1750, 14 L.Ed.2d 623 (1965) (refusing to apply the Mapp v. Ohio, supra, exclusionary rule retrospectively to collateral attacks on state court convictions finally decided prior to Mapp).

We concur in the foregoing four-point analysis and accordingly adopt the above-enumerated standards as the factors to be considered in determining the retroactive effect of an overruling decision. Ap*792plying these criteria to the case at bar, we hold that the decision we announce today— abolishing the previously recognized right of landlords to use the self-help eviction remedy — will be given partial retroactive effect, i. e., will apply to the instant parties, as well as prospective application.

Under the circumstances of the present case, there is no reason for declining to apply our rule in this manner. We have already stated our reasons for overruling Snitman in Part I, supra. Based on the reasoning set forth therein, we are convinced that the salutary purposes of the statutory remedies that we hold constitute landlords’ exclusive eviction remedy would be served by application of our decision to the parties at bar. We are unpersuaded that such application, while doing justice to appellee, would result in injustice to appellant or otherwise subject him to substantial detriment. Any contention to that effect would have to be based on an argument of reliance on the Snitman precedent. We find any suggestion as to the reasonableness of any reliance on Snitman without merit in this case, in light of the shadow cast on its continuing validity in this jurisdiction in the residential context by Wheeler v. Thompson, supra.24

Furthermore, since appellee-tenant is a one-time individual litigant, we deem it appropriate to extend to her the beneficial effect of the new rule whose establishment she has successfully championed before the court. Finally, because of the inherent nature of partial retroactivity, application of our decision in this manner will not disturb a mass of settled expectations or unduly burden our system’s administration of justice.25

III.

Appellant also challenges the award of punitive damages.

In ruling on this aspect of the claim for damages, the trial court stated that:

[Pjunitive or exemplary damages are allowed under law when a defendant acts with malice, with a wilful and wanton manner, or in utter disregard of the rights of another. Malice may be implied from the acts of the defendant. The decisions by a lawyer fully aware of the law and the procedure for the obtaining of possession of his property, for any one of several reasons which the law recognizes, including nonpayment of rent, requirements of a Government agency, et cetera, and a resort by him to the removal of the possessions of the tenants, in view of this Court, constitutes an utter disregard of the rights of Mrs. Johnson.

Wrongful eviction is a tort for which punitive damages may be allowed. Camaiier & Buckley-Madison, Inc., v. Madison Hotel, Inc., supra, 168 U.S.App.D.C. at 162 n. 93, 513 F.2d at 420 n. 93. Yet the mere commission of a tort is insufficient. “There must be circumstances of aggravation or outrage, such as spite or ‘malice,’ or a fraudulent or evil motive ... or such a conscious and deliberate disregard of the interests of others that his conduct may be called wilful or wanton.” W. Prosser, Law of Torts § 23, at 9-10 (4th ed. 1971). See, e. g., Harris v. Wagshal, D.C.App., 343 A.2d 283, 288 (1975); Franklin Investment Co. v. Homburg, D.C.App., 252 A.2d 95 (1969); Camalier & Buckley-Madison, Inc. *793v. Madison Hotel, Inc., supra; Nader v. Allegheny Airlines, Inc., 167 U.S.App.D.C. 350, 374, 512 F.2d 527, 551 (1975).

It is also true, however, that punitive damages are generally inappropriate when one’s tortious conduct is the outgrowth of an “innocent mistake.” Prosser, supra at 10. Appellant argues that his conduct can be so characterized because he might have assumed, on the basis of prior case law, that his conduct was permissible. We find this argument persuasive. Since we are unable to find in this record sufficient evidence of malice to outweigh the claimed reliance on Snitman v. Goodman, supra, the trial court’s finding that appellant’s conduct was conscious, wilful, and in utter disregard of appellee’s rights, was clearly erroneous.

IV.

For the foregoing reasons, the trial court’s judgment as to the award of punitive damages is reversed; in all other respects, it is affirmed.

So ordered.

with whom HARRIS Associate Judge concurs, dissenting:

The exercise of judicial power to change the law should be undertaken only with the greatest caution and only upon necessity, for in changing the law we necessarily defeat the expectations of those who relied upon its predictability. Because the instant case can properly be decided within the framework of existing law, I am unwilling to join the majority in their use of this case as an occasion for effecting a change in the law.

Snitman v. Goodman, D.C.Mun.App., 118 A.2d 394 (1955), the existing law which the majority have deemed to be in unavoidable conflict with the result desired in the instant case, involved the following circumstances. Snitman was the tenant of a park*794ing lot under Goodman, the landlord. Snit-man had no written lease, but paid a monthly rental. Goodman gave Snitman thirty days notice to quit, then took possession of the premises (the parking lot) peaceably. Snitman sued Goodman for possession of the premises, claiming that Goodman had no right of entry without legal process. We held that the common law right of entry by self-help had not — under the circumstances of that case, the only one before the court — been abrogated by statute.

In the instant case, on the other hand, Johnson is the tenant of a residence. Unlike Snitman, the landlord may not have been entitled to to possession, regardless of the means used to acquire possession.1 Unlike Snitman, Johnson had an opportunity to prove and failed to prove that she suffered any loss as a direct result of dispossession of the leased premises.2 Unlike Snit-man, Johnson claimed — and proved — a loss resulting from damage to her furnishings, i. e., chattels to which the landlord was not even arguably entitled except, possibly, as an' involuntary bailee. Indeed, the trial judge, while expressing disagreement with the Snitman decision, awarded compensatory damages only for damage to the chattels, a result entirely consistent with Snitman.

Each of these differences provides a means for performing our proper role — deciding the case before us without unnecessary disruption of precedent. First, this case is quite distinguishable from Snitman since Snitman involved a commercial lease while Johnson held a residential lease, a distinction acknowledged — and ignored — by the majority. See also Wheeler v. Thompson, 98 D.W.L.R. 41 (D.C.Gen.Sess. 1970).3 The majority, in fact, never consider whether their abrogation of the landlord’s right of entry by self-help might be limited to the residential context for which they evince concern.

Second, there has been no finding in this case that the landlord had any right to possession.4 Without a right to possession, it is irrelevant whether the landlord entered the premises himself or attempted to enlist judicial process, for neither remedy would be legally available to him absent such a right. It hardly requires the overruling of Snitman to hold that a landlord cannot enter his tenant’s residence without a right of possession.

Third, it is a matter of blackletter law that even a rightful repossession of leased premises does not give the landlord the right to damage or destroy furnishings or other chattels on the premises:

If the controlling law [j. e., Snitman ] permits the landlord, or an incoming tenant to use self-help to recover possession of leased property from a tenant improperly holding over after the termination of the lease [a factor not shown], neither the landlord, nor the incoming tenant, is entitled to resort to self-help to recover the possession of the leased property from the tenant, unless recovery by self-help is accomplished:

* sfc * * * *

(3) by using reasonable care to avoid damage to the property of the tenant, or of others on the leased property with the tenant’s permission. [Restatement (Second) of Property § 14.3 (1977).]

*795The distinction between damage to person or property during the course of a self-help eviction and the legal efficacy of self-help eviction to recover the premises (as in Snitman )5 has been recognized at least since Newton v. Harlan, 133 Eng.Rep. 490 (1840) (cited by the majority at 6), and remains a valid one. See, e. g., Allison v. Hodo, 84 Ga.App. 790, 67 S.E.2d 606 (1951) (affirming judgment for actual and punitive damages for injury to tenant’s furniture caused by self-help eviction of the furniture (proper) without taking reasonable precautions to assure its safety (improper)). See also W. Prosser, Law of Torts § 23 at 124 (4th ed. 1971) (“In all cases [whether or not self-help is permitted], he [the landlord] is liable for the use of any force beyond that reasonably necessary.”).

The first of these means of distinguishing Snitman is fully available in the record now before this court. Whether either of the other two means of distinguishing Snitman would require a remand of this most meager of trial records (for findings relating to the landlord’s entitlement to possession or the reasonableness of his care for Johnson’s furniture) would, of course, depend upon the burden of proof on these issues. Suffice it to say that remand either of the record or of the case was available and — I believe — mandated if necessary to assure our proper role as a judicial, not a legislative, body.

With these means properly available for distinguishing Snitman, it is incredible to me that a majority of any division of this court would feel frustrated in their attempts to resolve this case consistent with the principles of M.A.P. v. Ryan, D.C.App., 285 A.2d 310 (1977). See footnote * of the majority opinion. It is even more incredible that this court would then decide, en banc, that the only rationale available for decision is one requiring an en banc decision. En banc decision is unwarranted unless “necessary to secure or maintain uniformity of [court] decisions” or “the proceeding involves a question of exceptional importance.” D.C.App.R. 40(c). Since Snitman is easily and properly distinguishable from the instant case on a number of grounds, the former justification for en banc decision does not exist. I fear, therefore, that today’s decision means that a majority of this court feel free to decide “question[s] of exceptional importance” without the strictest necessity for such decision — a freedom I attribute to our legislative bodies but not to this court.

Because — and only because — the majority have decided this case by wholly overruling Snitman instead of so easily distinguishing it, I concur with my brethren who believe that the decision should have only prospective effect. Had we held for the tenant upon any of the narrower grounds, including the one relied upon by the trial court, the landlord would not have been justified in a reliance upon Snitman. But since the majority, by explicitly overruling Snitman, imply that the landlord’s actions were consistent with existing law, I would insist upon prospective application of an overruling decision.

with whom NEBEKER, Associate Judge, joins in part, and with whom YEAGLEY, Associate Judge, joins, concurring and dissenting:

I agree that in this day and age the use of self-help in this jurisdiction rather than resort to the court to evict a tenant should be relegated to the past.

I think, however, that this new rule of law should be given prospective effect. After all, until today the decision in Snitman v. Goodman, D.C.Mun.App., 118 A.2d 394 (1955), permitted self-help. It is the holding of today’s decision that Snitman v. Goodman is overruled.

I suppose one can rationalize various sorts of academic distinctions on prospective versus retroactive effect that may seem rather impressive but, for me, in the end it gets down to simple equity. Why should this court approve a penalty laid upon a litigant whose act was authorized by an extant decision of this court — a decision which must now be overruled in deciding this case? One might consider this court more nearly responsible than the litigant. In a case quite similar in principle we concluded the decision should be given prospective effect. Kelly Adjustment Co. v. Boyd, D.C.App., 342 A.2d 361 (1975).

It seems to me the court, in the end, should here look predominantly to whether it is equitable to penalize a defendant rather than to whether a plaintiff should be rewarded. That is certainly my concept of our jurisprudence in such circumstances and I would scarcely consider there is anything original about the viewpoint. All things considered, I favor prospective application. To this extent, I dissent.

dissenting in part:

I concur in the decision of the en banc court on the merits, but am of the view that the decision should be given prospective effect only. See Great Northern Ry. v. Sunburst Co., 287 U.S. 358, 365, 53 S.Ct. 145, 149, 77 L.Ed. 360 (1932), to the effect that it is not a constitutional problem and “[t]he choice for any state may be determined by the juristic philosophy of the judges of her courts, their conceptions of law, its origin and nature.”