John M. Walker, JR., Circuit Judge:
Plaintiff-Appellant Marcel Fashions Group, Inc. ("Marcel") and Defendants-Appellees Lucky Brand Dungarees, Inc. and affiliates ("Lucky Brand"), competitors in the apparel industry, have been hotly contesting their respective rights as to certain trademarks for nearly two decades. In this latest round, Marcel sues under the Lanham Act, alleging that Lucky Brand is infringing on Marcel's "Get Lucky" trademark through its use of "Lucky" on its merchandise, and that Lucky Brand does so in violation of an injunction entered in an earlier action between the parties. The district court dismissed the action, concluding that Marcel released its claims through a 2003 settlement agreement that resolved an earlier substantially similar litigation between the parties. We conclude that the district court did so in error because res judicata precludes Lucky Brand from raising its release defense in this action. To arrive at that result, we determine that under certain conditions parties may be barred by claim preclusion from litigating defenses that they could have asserted in an earlier action, and that the conditions in this case warrant application of that defense preclusion1 principle. Consequently, we vacate the judgment entered by the district court and remand for further proceedings.
I.
In a previous opinion vacating the entry of summary judgment dismissing the claims in Marcel's initial complaint, we discussed in detail the claims at issue in this case, as well as the parties' relevant history *234of litigation. Marcel Fashions Grp., Inc. v. Lucky Brand Dungarees, Inc. , 779 F.3d 102, 105-07 (2d Cir. 2015) (" Marcel I "). We reiterate that discussion here to the extent necessary to frame the issues relevant to an assessment of Lucky Brand's release defense.2
The 2001 Action . The settlement agreement through which Lucky Brand in this action asserts Marcel released its claims resolved a 2001 suit in which Marcel sued Lucky Brand for its alleged infringement of Marcel's "Get Lucky" mark (the "2001 Action"). See Marcel I , 779 F.3d at 105. The agreement provided that, inter alia , Lucky Brand would "desist henceforth from use of 'Get Lucky,' " and, pertinent to this appeal, Marcel agreed, through Section 8(e) of the agreement, to release certain claims it might have in the future arising out of its trademarks:
Marcel hereby forever and fully remises, releases, acquits, and discharges Defendants [Lucky Brand] ... from any and all actions, causes of action, suits ... or relief of any nature whatsoever, whether known or unknown, foreseen or unforeseen ... that Marcel ever had, now has or hereafter can, shall or may have, by reason of or arising out of any matter, cause or event occurring on or prior to the date hereof, including, but not limited to ... any and all claims arising out of or in any way relating to Lucky Brand's rights to use, license and/or register the trademark LUCKY BRAND and/or any other trademarks, trade names, brands, advertising slogans or tag lines owned, registered and/or used by Lucky Brand ... as of the date of this Agreement. No claims of any kind are reserved.
App'x 85-86 (the "Release"). Marcel and Lucky Brand's views have been consistently at odds on the scope of the Release. Marcel contends that it only released claims as to infringement that occurred prior to the 2003 execution of the agreement. Lucky Brand, for its part, contends that the Release is far broader, releasing any claim Marcel may have in the future in relation to any trademark registered prior to the execution of the agreement. The distinction is vital. Because each of Marcel's claims in this action ultimately allege a misappropriation of a mark registered before 2003, the latter interpretation would bar Marcel's claims, but the former would not.
The 2005 Action. Due at least in part to the parties' conflicting views on the breadth of the Release, further litigation followed the parties' settlement of the 2001 Action. Specifically, in 2005, Lucky Brand sued Marcel over Marcel's issuance of a license for use of the "Get Lucky" mark and Marcel counterclaimed with infringement claims of its own while also contending that Lucky Brand's use of "Get Lucky" violated the 2003 settlement agreement (the "2005 Action"). See Marcel I , 779 F.3d at 105-06.
Early in the 2005 Action, Lucky Brand tested its theory that the Release broadly barred Marcel's infringement claims. Specifically, Lucky Brand argued that, because the marks at issue were registered prior to the settlement agreement, Marcel released any claim alleging infringement of those marks. Lucky Brand moved to dismiss on this theory, arguing that Marcel's infringement counterclaims were barred *235by the terms of the Release. App'x 225-27. Marcel opposed the motion, arguing, as it does here, that the Release does not bar any claims as to infringing uses occurring after execution of the settlement agreement. See No. 05-cv-06757, Dkt. No. 49 at 8-9 (S.D.N.Y. Oct. 26, 2005). The district court denied the motion in relevant part, effectively concluding that it was premature to determine which claims in that action, if any, were subject to the Release. See App'x 257-58. The district court stated, however, that Lucky Brand was "free to raise the issue ... again after the record is more fully developed, including further development of the nature and use of the post-2003 marks." App'x 257-58. Heeding the district court's instruction, Lucky Brand again raised the Release in its answer, asserting as an affirmative defense that the "Settlement Agreement bars [Marcel's] Counterclaims." No. 05-cv-06757, Dkt. No. 67 ¶ 100 (S.D.N.Y. Jan. 2, 2007).
Despite relying on the release argument as part of its motion to dismiss and in its answer, however, Lucky Brand never again asserted a release defense in the 2005 Action. It was not for want of opportunity. The parties engaged in extensive summary judgment proceedings, substantial pre-trial motion practice, and a lengthy jury trial during which the release defense could have been asserted, but was not. And, without any argument to the jury by Lucky Brand that the Release barred Marcel's counterclaims as to infringement of its "Get Lucky" mark, the jury found in favor of Marcel on its counterclaim that Lucky Brand infringed that mark. See Marcel I , 779 F.3d at 106. The district court thereafter entered an injunction prohibiting Lucky Brand's use of the "Get Lucky" mark, and entered a declaration that the "Lucky Brand Parties infringed Marcel Fashion's GET LUCKY trademark ... by using GET LUCKY, the LUCKY BRAND trademarks, and any other trademarks including the word 'Lucky' after May 2003." Id.
The Instant Action. Lucky Brand did not appeal the judgment in the 2005 Action, but the parties' dispute over use of the relevant marks continued. In 2011, relying on the broad language of the court's injunction in the 2005 Action, Marcel filed the instant suit, seeking relief for Lucky Brand's alleged continued use of the "Lucky Brand" mark following that injunction. See id. at 106-07. Consistent with its non-assertion of the Release before the jury in the 2005 Action, however, Lucky Brand did not assert a release defense in the early stages of the instant action. Lucky Brand did not plead the Release as an affirmative defense in its answer to the initial complaint. No. 11-cv-05523, Dkt. No. 38 (S.D.N.Y. June 29, 2011). Nor did Lucky Brand raise the Release as a basis for dismissal when it moved for summary judgment. No. 11-cv-05523, Dkt. No. 65 (S.D.N.Y. Jan. 23, 2012).
Rather, Lucky Brand moved for summary judgment on the basis that Marcel's claims were precluded by res judicata (specifically, claim preclusion) in light of the final disposition in the 2005 Action. The district court agreed, but, in Marcel I , we reversed. Relying on TechnoMarine SA v. Giftports, Inc. , 758 F.3d 493 (2d Cir. 2014), we concluded that Marcel's claims were not barred by res judicata because Marcel alleged infringements that occurred subsequent to the judgment in the 2005 Action, claims which "could not possibly have been sued upon in the previous case." Marcel I , 779 F.3d at 108 (internal quotation marks omitted).
On remand, Marcel filed a second amended complaint. Armed with new counsel, Lucky Brand moved to dismiss, this time raising as its only argument that the *236Release bars Marcel's claims. The district court agreed and granted the motion, concluding that Marcel's claims "are plainly foreclosed by the Settlement Agreement" because each of the claims relates to trademarks that "were registered prior to the 2003 Settlement Agreement." App'x 247. Specifically, the district court concluded that "Section 8(e) ... rests on the nature, not the timing, of the claim. ... The release provision on which Lucky Brand's defense turns thus is not bound by temporal parameters." App'x 248. Consequently, the district court concluded that, because the settlement agreement released claims "in any way relating" to trademarks registered prior to the agreement, and the "Get Lucky" mark was registered prior to the agreement, Marcel had released its claims as to the "Lucky Brand" mark. App'x 248-49.
In arriving at its conclusion, the district court rejected Marcel's argument that Lucky Brand was precluded by res judicata from invoking the Release as a defense. Acknowledging that res judicata encompasses both issue and claim preclusion, the district court concluded that "[i]ssue preclusion does not apply, because the applicability of the Settlement Agreement's release provision was not actually litigated and resolved in the 2005 Action," and "[c]laim preclusion does not apply because Lucky Brand is not asserting a claim against Marcel." App'x 249. Marcel appealed.
II.
"We review de novo a grant of a motion to dismiss pursuant to Rule 12(b)(6), accepting the complaint's factual allegations as true and drawing all reasonable inferences in the plaintiff's favor." Brown Media Corp. v. K&L Gates, LLP , 854 F.3d 150, 156-57 (2d Cir. 2017) (internal quotation marks omitted); see also Ashcroft v. Iqbal , 556 U.S. 662, 678, 129 S.Ct. 1937, 173 L.Ed.2d 868 (2009) ; Twombly , 550 U.S. at 570, 127 S.Ct. 1955.
Marcel argues that the district court erred in concluding that the Release bars its claims, raising three principal contentions: (i) Lucky Brand is precluded by res judicata from asserting its release defense; (ii) Lucky Brand waived its release defense; and (iii) Marcel's claims are not barred by the terms of the Release. We agree with Marcel that, on this record, res judicata bars Lucky Brand's release defense, and we need not and do not address Marcel's other contentions. Specifically, we conclude that the doctrine of claim preclusion (or, more precisely, defense preclusion) may be applied in contexts such as this to bar the litigation of a party's defense and that the district court erred in holding to the contrary. We then conclude that this defense preclusion doctrine bars Lucky Brand from invoking its release defense in this action and thus vacate the dismissal of Marcel's claims and remand for further proceedings.
A.
In Marcel I , we identified the contours of res judicata (albeit in the context of the assertion of claims and not defenses), and discussed in detail the two doctrines that it encompasses: claim preclusion and issue preclusion. 779 F.3d at 107-08. At issue here is claim preclusion,3 a doctrine which, in the usual situation, bars a plaintiff from relitigating claims against a defendant that it lost in a previous action *237against the same defendant and claims that the plaintiff could have brought in that earlier action but did not. See id. Claim preclusion, applied in this manner, "serves the interest of society and litigants in assuring the finality of judgments, [and] also fosters judicial economy and protects the parties from vexatious and expensive litigation." Curtis v. Citibank, N.A. , 226 F.3d 133, 138 (2d Cir. 2000). The doctrine ensures these efficiency aims by "achieving finality and preventing piecemeal and wasteful litigation." N. Assurance Co. of Am. v. Square D Co. , 201 F.3d 84, 89 (2d Cir. 2000). Stated more broadly, claim preclusion ensures that "[w]hen a party is victorious, it [does] not have to defend that victory again." Id.
A party seeking to invoke claim preclusion must generally make three showings: (i) an earlier action resulted in an adjudication on the merits; (ii) that earlier action involved the same counterparty or those in privity with them; and (iii) the claim sought to be precluded was raised, or could have been raised, in that earlier action. See Monahan v. N.Y.C. Dep't of Corr. , 214 F.3d 275, 285 (2d Cir. 2000).
Relying on these elements and the policies supporting claim preclusion, Marcel argued to the district court that the principle of claim preclusion bars Lucky Brand from asserting its release defense in the instant action. The district court gave the argument short shrift. Rather than addressing the elements of claim preclusion, the district court dismissed Marcel's argument for a threshold reason: the doctrine is simply not available here where Marcel seeks to preclude a defense, and not a "claim." App'x 249. Although an understandable conclusion, given the predominant use of res judicata to preclude claims, we disagree with it, and conclude that the district court overlooked the principle that defenses are also subject to preclusion under res judicata .
B.
For the past quarter of a century, we have assumed that claim preclusion may bar a litigation defense but we have not had a case in which we have found a defense to be so precluded. Our most expansive decision on the issue is Clarke v. Frank , 960 F.2d 1146 (2d Cir. 1992), which addressed an action to enforce a judgment entered against the U.S. Postal Service in a proceeding at the Equal Employment Opportunity Commission (EEOC). The issue there was whether claim preclusion barred the Postal Service from raising a damages mitigation defense that it did not raise at the EEOC. We identified the governing principle as follows: "[c]laim preclusion prevents a party from litigating any issue or defense that could have been raised or decided in a previous suit, even if the issue or defense was not actually raised or decided." Clarke , 960 F.2d at 1150 (emphases added); see also Woods v. Dunlop Tire Corp. , 972 F.2d 36, 38 (2d Cir. 1992). Although we ultimately concluded that the Post Office was not precluded from raising the mitigation defense, it was not because claim preclusion was not generally available to defenses, but because one of the elements of the doctrine went unsatisfied: the Postal Service could not have raised the mitigation defense during the EEOC proceedings. See Clarke , 960 F.2d at 1151.
Other courts in our circuit have relied on Clarke to subject litigation defenses to claim preclusion, see, e.g. , Beckford v. Citibank, N.A. , 2000 WL 1585684, at *3-4 (S.D.N.Y. Oct. 24, 2000) ; see also Atateks Foreign Trade Ltd. v. Dente , 2017 WL 4221085, at *3-4 (S.D.N.Y. Sept. 22, 2017), but we have yet to further discuss the defense preclusion doctrine or frame its *238parameters. At the same time, we have never cast doubt on it.4
We are therefore called on for the first time to address whether a party can be barred by claim preclusion doctrine from prosecuting a litigation defense in a situation in which it is outcome-determinative. Although we are largely unbound by precedential authority, we are not without assistance. Specifically, we have guidance on this question from claim preclusion's sister res judicata component, issue preclusion.5
Issue preclusion was historically an exceedingly narrow doctrine, with two major limits to its reach: (i) it could only be applied by a defendant against a plaintiff to bar relitigation of an issue that had been actually litigated and lost by the plaintiff; and (ii) it could only be applied where the parties in the two actions were the same. See Triplett v. Lowell , 297 U.S. 638, 642-44, 56 S.Ct. 645, 80 L.Ed. 949 (1936) ; Bigelow v. Old Dominion Copper Min. & Smelting Co. , 225 U.S. 111, 127-28, 32 S.Ct. 641, 56 L.Ed. 1009 (1912). The Supreme Court abandoned the latter limitation, known as the "mutuality" principle, in Blonder-Tongue Laboratories, Inc. v. University of Illinois Foundation , 402 U.S. 313, 328-30, 91 S.Ct. 1434, 28 L.Ed.2d 788 (1971), but Blonder-Tongue expressly left open whether a plaintiff may invoke issue preclusion, id. at 329-30, 91 S.Ct. 1434.
In Parklane Hosiery Co. v. Shore , 439 U.S. 322, 99 S.Ct. 645, 58 L.Ed.2d 552 (1979), the Court resolved that open question when it rejected any per se prohibition on a plaintiff's use of issue preclusion. There, a class of shareholders of the Parklane Hosiery Company alleged that the company issued a materially misleading proxy statement. Trial on the shareholders' claims followed a suit brought by the Securities and Exchange Commission against Parklane Hosiery, which had resulted in a ruling following a bench trial that the relevant proxy statement was in fact false and misleading. The shareholders thereafter sought a ruling that Parklane Hosiery was collaterally estopped by issue preclusion from arguing that the relevant proxy statement was not false or misleading. The district court rejected the argument, purportedly on the basis that allowing for issue preclusion in this instance-where the earlier action was a bench trial-would violate the company's Seventh Amendment right to a jury trial. We reversed, 565 F.2d 815 (2d Cir. 1977), and the Supreme Court affirmed. The Court split its analysis in two parts, first addressing whether issue preclusion can be used offensively, and second, whether issue preclusion's offensive use by the *239shareholders would run afoul of Parklane Hosiery's Seventh Amendment rights. Its conclusion on whether issue preclusion can be used offensively is instructive here.
The Court noted at the outset that there generally "is no intrinsic difference between 'offensive' as distinct from 'defensive' issue preclusion" that would allow for preclusion in the latter but entirely forbid it in the former. Parklane , 439 U.S. at 331 n.16, 99 S.Ct. 645. It concluded that, at least under certain conditions, the doctrine of issue preclusion may be invoked by a plaintiff to estop a defendant from raising issues it lost in a previous proceeding. See id. at 329-33, 99 S.Ct. 645.
Although issue preclusion is different from claim preclusion in several important respects, we believe that an analysis similar to that of the Supreme Court's in Parklane informs the assessment of defense preclusion. We do not think the principles animating the claim preclusion doctrine disappear when that which is sought to be precluded is a defense. Rather, we view the efficiency concerns as equally pressing when the matter subject to preclusion is a defense rather than a claim. The following efficiencies are readily apparent. First , defense preclusion incentivizes defendants to litigate all their relevant defenses in an initial action, thereby promoting judicial efficiency at least to the same extent as does precluding claims. Second , absent defense preclusion, plaintiffs might be hesitant to rely on judicial victories for fear that a hidden defense will later emerge to alter their judicially established rights. Third , and relatedly, defense preclusion prevents wasteful follow-on actions that would not have been filed had the defense been asserted (and maintained) at the first opportunity. As with the preclusion of claims, therefore, defense preclusion ensures that "[w]hen a party is victorious, it [does] not have to defend that victory again." N. Assurance , 201 F.3d at 89.
The instant case is an apt example of the inefficiencies in prohibiting defense preclusion. It is Lucky Brand's contention that the Release bars any claim that Lucky Brand infringed on Marcel's "Get Lucky" mark, including those infringements that occurred after the parties executed their settlement agreement. Consequently, under Lucky Brand's theory, the Release bars all of Marcel's claims in the instant action. It follows, then, that had Lucky Brand litigated and prevailed on its release defense in the 2005 Action, in which Marcel sought relief for Lucky Brand's post-settlement agreement infringement of the "Get Lucky" mark, the instant action would have been avoided. This case therefore plainly demonstrates the inefficiencies that would have to be tolerated were we to prohibit defense preclusion. Assuming arguendo that Lucky Brand is correct in its interpretation of the Release (a question we do not reach), the court system will have been unnecessarily burdened with seven-plus years of litigation, involving 179 district court docket entries and two appeals to this Court. The entire endeavor would have been avoided, however, had Lucky Brand successfully litigated and not cast aside its release defense in the 2005 Action.
We acknowledge, however, that there exist distinctions between preclusion as a shield by defendants and as a sword against defendants. Referencing issue preclusion, the Supreme Court acknowledged as much in Parklane , leading it to conclude that "the[se] two situations should be treated differently." 439 U.S. at 329, 99 S.Ct. 645. The Court identified two reasons that offensive use of issue preclusion should be more circumscribed than its defensive counterpart. See id. at 329-31, 99 S.Ct. 645. First , offensive issue preclusion *240does not promote judicial efficiency to the same extent as does defensive issue preclusion.6 ibr.US_Case_Law.Schema.Case_Body:v1">See id. at 329-30, 99 S.Ct. 645. Second , in certain circumstances, the application of offensive preclusion might be unfair to defendants. See id. at 330-31 & n.14-15, 99 S.Ct. 645 (providing examples). Acknowledging these concerns, the Parklane Court resolved as follows:
We have concluded that the preferable approach for dealing with these problems in the federal courts is not to preclude the use of offensive collateral estoppel, but to grant trial courts broad discretion to determine when it should be applied. The general rule should be that in cases where a plaintiff could easily have joined in the earlier action or where, either for the reasons discussed above or for other reasons, the application of offensive estoppel would be unfair to a defendant, a trial judge should not allow the use of offensive collateral estoppel.
Id. at 331, 99 S.Ct. 645 (footnote omitted).
Parklane 's assessment of offensive issue preclusion provides a helpful framework for resolving whether, and how, we should permit the use of claim preclusion principles to bar the litigation of defenses. We acknowledge that, as with offensive issue preclusion, certain applications of defense preclusion could be unfair to defendants. The Court's observations on this point in Parklane are enlightening:
If, for example, the defendant in the first action was forced to defend in an inconvenient forum and therefore was unable to engage in full scale discovery or call witnesses, application of offensive collateral estoppel may be unwarranted. Indeed, differences in available procedures may sometimes justify not allowing a prior judgment to have estoppel effect in a subsequent action even between the same parties, or where defensive estoppel is asserted against a plaintiff who has litigated and lost. The problem of unfairness is particularly acute in cases of offensive estoppel, however, because the defendant against whom estoppel is asserted typically will not have chosen the forum in the first action.
439 U.S. at 331 n.15, 99 S.Ct. 645.
Defense preclusion raises a number of similar concerns. It might be unfair to bar a defendant from raising a defense that it elected not to bring in an earlier action because that action was of a significantly smaller scope, or the defense was somehow tangential to the matter. Put differently, it would be unfair to preclude a defense that the defendant had little to no incentive to raise in the earlier action. Relatedly, because it is generally not a defendant's prerogative to be hauled into court, they should be given some room to make tactical choices to attempt to end the suit against them with as little cost as possible without facing the unforeseen consequences of forever abandoning a defense. We also acknowledge that what constitutes a "defense" may not always be as clear as what constitutes a "claim," and that a broad understanding of "defense" in this context risks eliding the distinction between claim and issue preclusion. Finally, the fairness of defense preclusion may depend on the nature of the action. For example, there will hardly ever be unfairness in applying defense preclusion to bar a defendant from invoking defenses that could have been asserted in a previous *241action in a subsequent action to enforce a judgment previously entered against it. See Wright & Miller, supra note 4, § 4414 ; see also Clarke , 960 F.2d at 1150-51. This is especially true here, where sophisticated parties, armed with able counsel, litigate claims and counterclaims for nearly two decades. In contrast, pro se civil defendants might not initially mount their best defense and we should be wary of compounding that misfortune in subsequent litigation on nearly identical issues as to which they manage to muster a superior defense.7
The above examples, which are illustrative, and not exhaustive, demonstrate the wide array of fairness considerations potentially implicated by defense preclusion. Consistent with Parklane , however, we "conclud[e] that the preferable approach for dealing with these problems in the federal courts is not to preclude the use of [defense preclusion], but to grant trial courts broad discretion to determine when it should be applied." 439 U.S. at 331, 99 S.Ct. 645. District court discretion should be bound by the twin concerns discussed above: judicial efficiency and fairness.
In sum, we conclude that defense preclusion bars a party from raising a defense where: (i) a previous action involved an adjudication on the merits; (ii) the previous action involved the same parties or those in privity with them; (iii) the defense was either asserted or could have been asserted, in the prior action; and (iv) the district court, in its discretion, concludes that preclusion of the defense is appropriate because efficiency concerns outweigh any unfairness to the party whose defense would be precluded.8 Cf. Ward v. Harte , 794 F.Supp. 109, 118 (S.D.N.Y. 1992) (concluding "that the use of offensive collateral estoppel would not be unfair" where "the Court ha[d] before it the same parties who appeared" in a prior action, and the defendant "was represented at all times by counsel, which significantly mitigates, in the Court's view, any possible unfairness towards him" and observing that, under these circumstances, Parklane 's fairness "concerns are somewhat lessened").
C.
The above stated factors are easily met here. Lucky Brand does not (and cannot) dispute the first three, and it would have been an abuse of discretion for the district to have concluded anything other than that any unfairness to Lucky Brand is substantially outweighed by the efficiency concerns identified above. Our review of the record evinces no conceivable justification for Lucky Brand, a sophisticated *242party engaged in litigation pertaining to its ability to use some of its core trademarks, not to have fully litigated the release defense in the 2005 Action and Lucky Brand has not suggested one.9 It should be the rare case that application of defense preclusion will be unfair where not even a theoretical explanation for the omission of the defense in the earlier action is apparent. Lucky Brand cannot seriously contend that it viewed the release defense as a minor or tangential issue. To the contrary, despite its failure to reintroduce the Release as a defense in the 2005 Action, it initially viewed the Release as a complete defense when it moved to dismiss the counterclaims in that action, see No. 05-cv-06757, Dkt. No. 44 at 13-15 (S.D.N.Y. Oct. 6, 2005), before it decided to forego the defense at summary judgment. This is the same position it takes as to the scope of the Release in this action. Relatedly, a release defense is a specifically identified affirmative defense in the federal rules, see Fed. R. Civ. P. 8(c)(1), obviating any concern that application of claim preclusion here elides the line between claim and issue preclusion. Finally, the nature of the instant suit makes this case particularly suitable for defense preclusion. Marcel styled its complaint as one that effectively sought to enforce the judgment entered in the 2005 Action. And, the judgment-enforcement context is especially deserving of a defense preclusion rule.10 Wright & Miller, supra note 3, § 4414 (the "simplest rule of all is that direct enforcement of a judgment cannot be resisted merely by raising defenses that might have been raised before the judgment was entered"). To conclude otherwise would allow judgment-debtors wide leeway to forgo payment while they assert previously unasserted defenses to successive judgment enforcement actions that they force their creditors to bring.
Though it will be the infrequent case that a defense will be precluded by the rule we describe, it is the proper resolution of the case before us. On remand, Lucky Brand is barred from asserting the Release as a defense to Marcel's infringement claims as set forth in the operative complaint.
III.
For the reasons stated above, we VACATE the judgment of the district court and REMAND for proceedings consistent with this opinion.