Ronnie Gene Bass v. Phoenix Seadrill/78, Ltd., Third-Party v. Crown Rig Building Services, Inc., and Branham Industries, Inc., Third-Party
749 F.2d 1154
3rd Cir.
1985
Check Treatment
*1156 RANDALL, Circuit Judge:

This appeal presents important questions with respect to the scope and timeliness of judicial review of partial settlements executed in multi-party lawsuits. Following a bench trial, the district court voided portions of the plaintiff’s settlement agreement with one of three defendants but nonetheless enforced the plaintiff’s release of the settling defendant. 562 F.Supp. 790. Although we uphold the district court’s liability and damage findings, we reverse that portion of its opinion which partially abrogates the settlement agreement.

I. FACTUAL AND PROCEDURAL BACKGROUND.

Ronnie Gene Bass (Bass) sought compensation below from three defendants for personal injuries suffered while performing seaman’s duties aboard a newly commissioned Jones Act vessel. Bass was employed by Phoenix Seadrill/78 Ltd. (Phoenix) as a roughneck on an offshore drilling rig known as Big Foot I. Contractors hired by Phoenix completed construction of Big Foot I in early 1980. Branham Industries, Inc. (Branham) designed and fabricated the rig’s derrick; Crown Rig Building Services, Inc. (Crown Rig) erected the derrick. Upon completion of Big Foot I, a Phoenix crew, of which Bass was a member, commenced drilling operations aboard the rig in the Gulf of Mexico. On May 8, 1980, after only five days of drilling, Bass suffered severe injuries when a forty-pound jack handle fell from a work platform eighty feet above the rig’s floor and struck him on the head. The jack handle, which was used to raise and lower the work platform (known as a “monkey board”), pivoted on a steel pin to which it was supposed to be attached by a cotter key. The cotter key was not located after the accident; the parties devoted much time below to assigning responsibility for its installation and inspection.

Predictably, Bass’ tragic accident spawned much litigation. Initially, Bass sought compensation from Phoenix 1 under the Jones Act and the general maritime law. Phoenix, in turn, brought third-party actions for indemnity against Branham and Crown Rig. Finally, Bass amended his original complaint to assert negligence and strict liability claims directly against Bran-ham and Crown Rig.

Long before trial, Bass compromised his claims against Phoenix in a settlement that the parties and the district court label a “Mary Carter” agreement: 2 for $210,000, Bass forever released Phoenix from all liability arising from the jack handle accident; Bass assigned to Phoenix, however, up to $178,000 of any funds recovered from Branham and Crown Rig. 3 Moreover, Bass agreed not to settle his claims against third parties without the consent of Phoenix.

Bass and Phoenix disclosed their agreement to the court and opposing counsel six months before trial. On the first day of trial, Crown Rig and Branham moved the court to void the settlement agreement on the grounds that it “is against public policy, abuses the judicial form and processes, and fosters champerty and maintenance.” Record Vol. II at 309. The district court, however, did not rule on the motion at that time. Instead, the court tried the case *1157 without a jury. Phoenix participated, notwithstanding Bass’ release, to prosecute its third-party actions for indemnity against Branham and Crown Rig. In a post-trial memorandum opinion, the court awarded Bass $650,000 in damages apportioned, as follows, according to the comparative fault of the defendants: Phoenix — 40 percent; Branham — 40 percent; and Crown Rig — 20 percent. In the same opinion, the court, with the benefit of perfect hindsight, finally ruled on the motion to void the settlement: the court abrogated the rebate and veto provisions of the agreement because of grossly inadequate consideration and a perceived deterrent effect on compromise with the non-settling defendants; the court found, however, that, absent the rebate and veto provisions, the agreement was supported by ample consideration. Accordingly, the court enforced the agreement as a straight eash-for-release settlement and entered judgment that Bass recover $260,000 from Branham and $130,000 from Crown Rig. The release provisions of the settlement agreement, in the court’s view, discharged Phoenix’s liability for the remaining $260,000 of Bass’ damages. The court’s partial abrogation of the agreement relieved Bass of the obligation to rebate $178,000 to Phoenix. In a supplemental order, the court denied Phoenix’s claim to complete indemnity from Branham and Crown Rig for the amount paid to Bass in settlement.

II. CONTENTIONS ON APPEAL.

The parties take various positions with respect to the district court’s partial abrogation of the Bass-Phoenix settlement. Phoenix argues that we have previously sanctioned the use of Mary Carter settlement techniques and that the district court lacked the authority to void the rebate provision of this agreement. Seizing on a chance to save $178,000, Bass argues, for the first time on appeal, that he in fact received grossly inadequate consideration for the rights he relinquished in the settlement agreement. Branham and Crown Rig also claim that the agreement should have been voided, but take the argument one step further: the court deprived them of a fair trial by failing to void the agreement before trial. Knowledge of the agreement, according to Branham, undoubtedly colored the court's perception of the evidence and influenced the court to assign a large percentage of fault to parties other than Phoenix and to inflate the damage award.

The parties also challenge the district court’s allocation of fault for the jack handle accident. Phoenix asserts that, notwithstanding any negligence on its own part, it is entitled to Ryan 4 indemnity from Branham and Crown Rig because they both breached a warranty of workmanlike performance implicit in their respective contracts with Phoenix. Branham and Crown Rig, on the other hand, argue that the district court’s liability findings are clearly erroneous. They both claim as an initial matter that the evidence conclusively demonstrates that Phoenix’s negligence was the sole proximate cause of the accident; alternatively, Branham and Crown Rig both argue that, as between themselves, the other party is responsible for any fault not assigned to Phoenix. Crown Rig argues that, at any rate, it is entitled to complete indemnity from Branham because of a difference in the degree of fault for which Branham and Crown Rig are liable: Branham is strictly liable to Bass on a products liability theory; Crown Rig’s liability, if any, is premised on passive negligence.

Finally, Crown Rig claims that the district court’s damage finding is excessive. Bass, naturally, argues that the court’s damage award is amply supported by the evidence.

III. THE SETTLEMENT AGREEMENT.

The district court analyzed four deleterious effects that allegedly flow from the *1158 Bass-Phoenix settlement agreement: (1) the agreement fosters champerty and maintenance; (2) it deters Bass from settling with the other defendants; (3) the agreement falsely aligns the parties before the trier of fact; and (4) it overreaches Bass who, as a seaman, is a ward of admiralty. 5

The court found that, because the settlement was fully disclosed and the case was not tried to a jury, the agreement’s economic realignment of the parties did not impugn the fact-finding process. 6 The court concluded, however, that the rebate and veto provisions of the settlement agreement are champertous and that they effectively deprived Branham and Crown Rig, in degradation of the strong policy favoring settlements, of a chance themselves to compromise Bass’ claims against them. Moreover, the court found, despite Bass’ professed satisfaction with the deal, that the settlement agreement is grossly unfair to Bass: $210,000 is manifestly insufficient consideration, in the court’s view, for a release, an assignment of $178,000 of potential recovery, and a veto power over settlements with other defendants. The court determined, however, that $210,000 constitutes fair consideration, on these facts, for a release of liability. Therefore, the district court voided the rebate and veto provisions of the agreement, assigned all of the consideration paid by Phoenix to support the agreement’s release, and simply enforced the agreement as if the parties had executed a straight settlement.

The propriety of Mary Carter settlement techniques has received much attention in recent cases and commentaries. See, e.g., Note, Mary Carter Agreements in Maritime Personal Injury Suits, 22 S.Tex.L.J. 545 (1983). We note at the outset that, while we have yet to conduct a plenary analysis of the subject, we have implicitly approved of the use of Mary Carter settlement agreements. See, e.g., Wilkins v. P.M.B. Systems Eng’g, 741 F.2d at 798 n. 2; Leger v. Drilling Well Control, Inc., 592 F.2d 1246 (5th Cir.1979) (affirming decision by district court, after it ordered a new trial because of undisclosed Mary Carter settlement, to grant non-settlor proportionate, not dollar-for-dollar credit, for amount paid to plaintiff by settlor); Reichenbach v. Smith, 528 F.2d at 1072 (finding no prejudicial error in failure to disclose Mary Carter settlement to jury). Despite some broad language to the contrary, 7 we do not read the district court’s *1159 opinion to adopt a per se ban on Mary Carter settlements in maritime personal injury cases. Rather, we think that the district court simply held that this agreement is unenforceable as written because of the deleterious effect of certain of its provisions on specific parties before the court: (1) the rebate provision renders the consideration to Bass inadequate and (2) the veto provision impermissibly deprives the non-settling defendants of a fair chance to reach their own compromise with Bass.

Whatever the validity of its decision on the merits, we are particularly troubled by the timing of the court’s ruling on the Bass-Phoenix settlement. We have grave doubts about the propriety of reviewing the adequacy of consideration for a settlement, the very purpose of which is, at least in part, to avoid the uncertainties of litigation, after the case has been fully tried and the court knows, to the degree of certainty required to make findings of fact, the full extent of the settling defendant’s liability to the plaintiff. Cf. Leger v. Drilling Well Control, Inc., 592 F.2d at 1250-51 (“Whether the plaintiff or any of the defendants are ultimately found to have made a favorable settlement, we will ‘respect the aleatory nature of the settlement process....’”) (quoting Doyle v. United States, 441 F.Supp. 701, 711 n. 5 (D.S.C.1977)). Although compromise at the appellate level is not unheard of, we also question the efficacy of a post -trial ruling designed to remove a settlement deterrent. At any rate, we need not decide whether the untimeliness of the district court’s ruling alone constitutes reversible error. We reverse the partial abrogation of the settlement agreement, and would do so even if the court had made its ruling before trial, for two reasons: (1) the district court exceeded the scope of its authority to protect seamen from unfair settlement agreements and (2) from the record before us, we discern no plain prejudice to the rights of the non-settling defendants.

In Wilkins v. P.M.B. Systems Eng’g, 741 F.2d at 798 n. 2, a case involving a similar abrogation, by the same court that tried this case, of the Mary Carter provisions of a settlement agreement, we noted that district courts have discretion to review settlement agreements, particularly Mary Carter settlements, that are reached in pending cases:

However, because of the adverse prejudicial effect to non-settling defendants, the plaintiff, or both, resulting from some of the manipulations possible under the Mary Carter device, and because of the inherent power of a trial court to enforce settlement agreements reached in cases pending before it, and to determine the validity of such agreements it is important that the trial court retain a significant degree of discretion in approving and enforcing Mary Carter agreements, as well as in disclosing their terms to the jury. Where, as in this case, the settling plaintiff is a seaman, and thus a traditional ward of admiralty, that discretion of the court to scrutinize and determine the validity of Mary Carter agreements is magnified. 8

We do not think that the Wilkins panel intended to suggest, however, that the district court can substitute its own judgment for that of the settling parties. Cf. United States v. Texas Educ. Agency, 679 F.2d 1104, 1108 (5th Cir.1982) (court should not “substitute [its] ‘own judgment as to op *1160 tional settlement terms for the judgment of the litigants and their counsel' ”) (quoting Armstrong v. Board of School Directors, 616 F.2d 305, 315 (7th Cir.1980)). Rather, we think a district court’s discretion is necessarily limited: the court may protect the legal rights of the parties before it and guard against abuse of its own processes; the court should not, however, directly enter the settlement process on the side of either party. 9 Accordingly, we must carefully review each of the grounds relied upon by the district court to void the rebate and veto provisions of the settlement.

A. Unfairness to Bass. 10

Seamen, of course, are wards of admiralty whose rights federal courts are *1161 duty-bound to jealously protect. See, e.g., Garrett v. Moore-McCormack Co., 317 U.S. 239, 63 S.Ct. 246, 87 L.Ed. 239 (1942). We have long recognized that courts must be particularly vigilant to guard against overreaching when a seaman purports to release his right to compensation for personal injuries. See, e.g., Wink v. Rowan Drilling Co., 611 F.2d 98, 100 (5th Cir.) (“releases or settlements involving seaman’s rights are subject to careful scrutiny”), ce rt. denied, 449 U.S. 823, 101 S.Ct. 84, 66 L.Ed.2d 26 (1980).

It does not follow, however, that a seaman may not compromise his rights without prior judicial approval of the fairness of the agreement. In other contexts, such as class actions, in which we closely scrutinize settlement agreements, a pending case may not be settled until the court determines that the settlement is fair. 11 See Fed.R.Civ.P. 23(e) (“A class action shall not be ... compromised without the approval of the court....”). Despite the solicitude with which we view the abdication of seamen’s rights, we have never adopted a similar requirement in seamen’s personal injury cases that district courts independently review the terms of settlement for fairness. 12

Our ultimate concern in these cases is not whether the seaman has received what the court considers to be adequate consideration for the rights he has relinquished; rather, we inquire whether the seaman relinquished those rights with “ ‘an informed understanding of his rights and a full appreciation of the consequences’ when he executed the release.” Charpentier v. Fluor Ocean Serv., 613 F.2d 81, 84 (5th Cir.1980) (quoting Cates v. United States, 451 F.2d 411, 414 (5th Cir.1971)). The adequacy of consideration is simply a factor to be considered in determining whether the seaman had a full understanding of his rights and of the consequences of the settlement agreement. See, e.g., Garrett v. Moore-McCormack Co., 317 U.S. at 248, 63 S.Ct. at 252 (“The adequacy of the consideration ... [is] relevant to an appraisal of this understanding [of a seaman’s rights].”); Wink v. Rowan Drilling Co., 611 F.2d at 100 (same); Strange v. Gulf & South American S.S. Co., 495 F.2d 1235, 1237 (5th Cir.1974) (seaman cannot avoid settlement “merely because of the fact that plaintiff may have received considerably less for his injuries than at the end of a controversy he might reasonably have hoped to obtain”) (quoting Harmon v. United States, 59 F.2d 372 (5th Cir.1932)). Rather than demanding prior judicial approval as a matter of course, we protect seamen’s rights in the settlement context by requiring the proponent of a release that has been attacked to show that it was freely and knowingly executed. Garrett v. Moore-McCormack, 317 U.S. at 248, 63 S.Ct. at 252.

Applying these principles, we conclude that the district court exceeded its authority in voiding the rebate and veto provisions of this agreement because of perceived unfairness to Bass. A review of *1162 the district court’s opinion reveals that the only finding with respect to the agreement’s impact on Bass is that Phoenix did not pay adequate consideration for the rights that Bass relinquished. The opinion does not expressly reach what, in our view, is the ultimate question: did Bass fully understand his rights and the effect of the settlement on those rights when he executed the agreement? See Cates v. United States, 451 F.2d 411, 416 n. 14 (5th Cir.1971) (“The controlling inquiry ... should be: Did the [proponent of the settlement] demonstrate that [the seaman] fully understood the consequences of his release?”). 13

We do not intend to denigrate the court’s role in protecting seamen from overreaching. Moreover, we recognize that the proponent of a seaman’s settlement must demonstrate that it was freely and knowingly executed and that the burden of doing so is often more difficult when the agreement is based upon inadequate consideration. We simply hold that adequacy of consideration is not the touchstone of a valid seaman’s release; absent a finding that the settlement was not executed with a full understanding of the seaman’s rights and the effect of the agreement thereon, the district court lacks authority, especially where the seaman testifies to complete satisfaction, to void the agreement because the court thinks the seaman could have negotiated a better deal.

The district court’s opinion does not discuss Bass’ understanding of his rights or the effect of the agreement on those rights. The opinion does conclude, however, that Phoenix did not sustain its burden with respect to the agreement. Although the nature of the burden is not articulated, presumably the court was aware that the settlement proponent’s burden is to demonstrate that the seaman acted with a complete understanding. At any rate, whether the court acted on the belief that inadequate consideration alone justified abrogation of the settlement 14 or from a conviction that Bass did not fully comprehend the consequences of the agreement, we cannot sustain the court’s action. The former conclusion is an erroneous view of the law; the latter conclusion is an erroneous fact determination. 15

The record clearly demonstrates that Bass executed the agreement freely and knowingly, with a full understanding of his rights and the consequences of settlement; any implicit finding to the contrary in the district court’s opinion is clearly erroneous. 16 The agreement itself is replete with *1163 warnings that the release of Phoenix is indeed a complete one. Moreover, the rebate and veto provisions of the agreement are set forth in plain and intelligible language. Bass was represented by counsel during negotiation and execution of the agreement. During his opening statement at trial, Bass’ counsel extolled the benefits that the settlement agreement bestowed on Bass: “The partial settlement was entered into by the Plaintiff freely and voluntarily and for reasons — financial reasons and other reasons, Your Honor. Since the settlement has been entered into the Plaintiff has bought him a mobile home and moved out to [live by] himself. He had lived with his sister and brother-in-law most of the time.” Record Vol. III at 9. Moreover, Bass himself testified that the rebate provisions of the settlement were fully explained to him and that he willingly executed the agreement without reservation. 17 We note that, even though Bass has finally decided to attack the agreement for the first time on appeal, he still does not allege that he executed the agreement with an imperfect understanding of his rights or the consequences of settlement: his brief is limited to an attack on the adequacy of consideration; his counsel at oral argument again candidly admitted that both he and Bass completely understood the agreement at the time it was executed. While we acknowledge that representation by counsel does not perforce render a seaman’s release enforceable, see, e.g., Blanco v. Moran Shipping Co., 483 F.2d 63 (5th Cir.1973), ce rt. denied, 416 U.S. 904, 94 S.Ct. 1608, 40 L.Ed.2d 108 (1974), the record is absolutely clear that Bass knowingly executed the release. We hold, therefore, that the district court erred in voiding portions of the agreement because they were, in the view of the court and the non-settling defendants, but not in the view of Bass or his counsel, unfair to Bass.

B. Settlement Deterrent.

We turn now to an analysis of the district court’s additional ground for voiding the settlement: its deterrent effect on settlements with Crown Rig and Branham. The district court noted two aspects of the agreement that purportedly deter settlements: (1) because of the rebate provision, Bass does not receive the full value of *1164 settlement dollars paid by other defendants until $251,000 has been recouped; thus, Bass is likely to demand at least $251,000 to compromise his remaining claims against the non-settling defendants, which may artificially inflate the settlement value of the case and (2) because of the veto provision, Branham and Crown Rig are faced with more than Bass’ economic disincentive to settle fo? less than $251,000: Phoenix will likely prevent Bass from settling for any amount less than that needed for Phoenix to obtain a full rebate of $178,000. Although its holding is less than clear, the court implies that the economic settlement deterrent built into the agreement would not itself justify abrogation of the settlement; it is the veto provision that, in the district court’s opinion, renders the agreement unenforceable on settlement deterrent grounds. 18

We certainly agree with the district court’s initial premise: public policy favors voluntary settlements which obviate the need for expensive and time-consuming litigation. See, e.g., Miller v. Republic National Life Ins. Co., 559 F.2d 426, 428 (5th Cir.1977). We also recognize, however, that, in multi-party law suits, complete settlement is often not possible: “Likewise, an omnipresent characteristic of modern maritime litigation is its multiparty nature. Settlements with some, but not all, are frequently made — and should be even more encouraged.” Cates v. United States, 451 F.2d at 415-16 (footnote omitted) (rejecting unity of release rule in maritime personal injury cases).

While we do not doubt that, in proper circumstances, a district court may refuse to enforce provisions of a settlement agreement that burden the rights of non-settling parties who remain in the lawsuit, see Dunn v. Sears, Roebuck & Co., 639 F.2d at 1173, we are not convinced that this is such a case. In Dunn, plaintiff settled with some, but not all, of the defendants before trial. As part of the agreement, the settling defendants agreed not to make their employees, whose deposition testimony was damaging to plaintiff, available to the non-settling defendants. At trial, plaintiff argued that the terms of the settlement agreement rendered the depositions inadmissible. The trial court disagreed, and we affirmed its decision to void the portions of the settlement agreement which purported to deprive the non-settling defendants of relevant, otherwise admissible evidence.

We think this case is fundamentally different from Dunn. In Dunn, the settlement purported to deprive the non-settling defendants of their legal right to present admissible evidence. In this case, the settlement agreement does not prejudice the rights of Branham or Crown Rig. Bass is under no obligation to settle his claims; in fact, he has an absolute right to refuse to settle and to submit his case to a trier of fact. The fact that this agreement makes it factually less likely that Bass will settle with Crown Rig and Branham, in our view, does not justify setting it aside.

The Seventh Circuit has recently considered the circumstances under which a court is justified in abrogating a partial settlement because of prejudice to non-settling parties. In Quad/Graphics, Inc. v. Fass, 724 F.2d 1230 (7th Cir.1983), the court held:

Therefore, we do not believe that a court should inquire into the propriety of a partial settlement merely upon á showing of factual injury to a non-settling party. Some disadvantage to the remaining defendants is bound to occur and may, in fact, be the motivation behind the settlement. But just as a court has *1165 no justification for interfering in the plaintiffs initial choice of the parties it will sue (absent considerations of necessary parties), the court should not intercede in the plaintiffs decision to settle with certain parties, unless a remaining party can demonstrate plain legal prejudice. This standard should strike the proper balance between the policy consideration of encouraging voluntary resolutions of litigation and the court’s duty to protect the rights of the parties before it.

Id. at 1233.

Although we have not expressly adopted such a standard, we think that it has much to recommend it and, moreover, that it comports with the jurisprudence of this circuit. See, e.g., In re Beef Industry Antitrust Litigation, 607 F.2d at 172 (non-settlor generally lacks standing to attack settlement agreement; non-settlor may attack settlement that purports to deprive him of right to seek indemnification from settlor); Dunn v. Sears, Roebuck & Co., 639 F.2d at 1173 (non-settlor successfully attacks settlement that attempts to deprive him of right to present relevant evidence at trial); Holiday Queen Land Corp. v. Baker, 489 F.2d 1031 (5th Cir.1974) (applying “plain legal prejudice” standard in Rule 41 voluntary dismissal case; tactical disadvantage alone will not suffice). A defendant who loses the race to settle with the plaintiff, or who simply decides to take his chances at trial, should not suffer any substantive or procedural disadvantage when other defendants do settle. We fail to see, however, why a plaintiff should be foreclosed from voluntarily settling with one defendant on mutually agreeable terms simply because those terms remove plaintiff’s economic incentive to settle with the other defendants. That, it seems to us, is a consequence that may well flow from any settlement with less than all defendants. The fact that such a consequence is assured when the settlement agreement includes a Mary Carter provision, assuming no additional terms which purport to deprive the non-settlors of substantive or procedural rights to which they are entitled, does not disturb us. Cf. Leger v. Drilling Well Control, 592 F.2d at 1251 (“If a party decides to try a case, it must be prepared to accept whatever benefits or burdens flow from its decision.”).

In sum, we find no basis for setting aside the veto and rebate provisions of the agreement. Accordingly, we reverse that portion of the district court’s judgment which voids those provisions of the agreement. 19

IV. ALLOCATION OF FAULT.

All parties except Bass attack the district court’s allocation of responsibility for the jack handle accident. The district court concluded that the fall of the jack handle was precipitated by (1) vibrations caused by the impact of the crown block against the monkey board 20 and (2) the absence or failure of the cotter key which was supposed to secure the jack handle to the monkey board. Despite allegations in the pleadings of negligence, strict liability in tort, unseaworthiness and breaches of express and implied warranties, the court considered only whether the defendants’ conduct with respect to the monkey board and crown block constituted negligence. The court *1166 concluded that all defendants were negligent. Branham, to whom forty percent of the responsibility for the accident was assigned, negligently designed and manufactured the monkey board by supplying a cotter key of inadequate size to secure the jack handle and by failing to provide a safety chain to prevent a fall in the event of cotter key failure. Crown Rig, on the other hand, who, under the court’s findings, bears twenty percent of the responsibility, negligently failed to inspect the jack handle assembly supplied by Branham during erection of the Branham components. Phoenix, who also bears forty percent of the blame, negligently continued to operate the rig with knowledge that the crown block was subjecting the monkey board to extreme vibration.

Branham and Crown Rig attack this allocation of fault as clearly erroneous, which, of course, is the standard that governs our review of a district court’s findings of fact. See Fed.R.Civ.P. 52(a). Specifically, they attack certain subsidiary fact-findings announced in the district court’s opinion. Both Crown Rig and Branham argue that the district court erroneously determined that Branham specifications for the monkey board assembly called for a cotter key “one third the diameter of its hole.” Record Vol. II at 443-44. They claim that the pin was actually two-thirds the size of its hole. This error is significant to Bran-ham because it allegedly undermines the district court’s conclusion that Branham negligently supplied a cotter key of inadequate size; it is significant to Crown Rig because it purportedly vitiates the district court’s finding that a reasonable inspection by Crown Rig would have revealed the inadequacy of the cotter key supplied by Branham.

Branham also argues that the district court erroneously concluded that Branham shipped the monkey board to Crown Rig with the jack handle already attached; rather, Branham claims, the monkey board was shipped unassembled and any failure to properly install the cotter key is the fault of Crown Rig.

Finally, Crown Rig argues that the district court erroneously determined that Crown Rig finished erecting Big Foot I’s derrick “at the beginning of May, 1980.” Rather, Crown Rig argues, its work was completed by February 29, 1980. This error is significant, Crown Rig claims, because the evidence establishes that between February and May Branham modified the monkey board, after Crown Rig’s inspection, in ways that may have contributed to the accident.

Our review of the record indicates that, even if these subsidiary fact-findings are clearly erroneous, the district court’s ultimate fault allocation is supported by the evidence. The district court was indeed faced with a difficult task in assigning responsibility for this accident. Since the cotter key was not located following the accident, it is impossible to know whether the handle fell because (1) a properly installed key failed; (2) an improperly installed key failed; or (3) the key was never installed in the first instance. We cannot say, however, that the district court’s findings are clearly erroneous.

Assuming Crown Rig assembled the monkey board and failed to insert the cotter key or did so improperly, there is evidence to support a negligence finding against Branham for failing to provide a backup safety device. Assuming Branham assembled the monkey board, and either properly or improperly installed the cotter key, there is evidence to support a negligence finding against Crown Rig for failing to notice the inadequate or improperly installed cotter key. Moreover, we think these alternative conclusions are supported by the evidence whether the cotter key was one-third or two-thirds the size of the hole in which it was inserted. Finally, even if Crown Rig did complete its work in late February, the evidence nevertheless supports the conclusion that its failure adequately to inspect contributed to the fall of the jack handle.

Y. INDEMNITY.

Both Crown Rig and Phoenix seek complete indemnity. Crown Rig argues *1167 that it is entitled to indemnity from Branham because Branham’s liability is bottomed on strict liability in tort, see Restatement of Torts (Second) § 402A; Crown Rig, on the other hand, is liable for negligence. This argument is flawed for two reasons: (1) the district court did not conclude that Branham is strictly liable for supplying a defective product; rather, the court premised Branham’s liability, like Crown Rig’s, on simple negligence and (2) the only arguable difference in the degree of fault attributed to Branham and Crown Rig by the district court is that Branham may have been actively negligent while Crown Rig was passively negligent; we have, however, abandoned the right to indemnity based on this distinction, in favor of a system of comparative fault, see Loose v. Offshore Navigation, Inc., 670 F.2d 493 (5th Cir.1982). On a more fundamental note, our review of the record reveals that Crown Rig did not file a cross action seeking indemnity from Branham. We discern no error, therefore, in the failure to grant Crown Rig indemnity from Branham.

Relying principally on Stevens v. East-West Towing Co., 649 F.2d 1104 (5th Cir.1981), cert. denied, 454 U.S. 1145, 102 S.Ct. 1007, 71 L.Ed.2d 298 (1982), Phoenix seeks indemnity from Crown Rig and Branham for the funds paid to Bass in settlement. Phoenix alleges that Branham and Crown Rig breached a warranty of workmanlike performance (WWLP) implicit in their contracts with Phoenix. Although the district court’s explanation that “having found Phoenix negligent ... it would be inconsistent to allow Phoenix indemnity,” Record Vol. II at 490, is unsatisfying, we ultimately agree that this is an inappropriate case for indemnification based upon breach of the WWLP.

In Stevens, we applied an indemnity rule first announced in Ryan Stevedoring Co. v. Pan-Atlantic S.S. Corp., 350 U.S. 124, 76 S.Ct. 232, 100 L.Ed. 133 (1956), which, because of statutory changes in the law of longshoremen’s compensation and subsequent decisions limiting the rule’s scope, we have since labelled a “withered” doctrine. See Gator Marine Service v. J. Ray McDermott & Co., 651 F.2d 1096 (5th Cir.1981). In Ryan itself, a longshoreman was injured by shifting cargo improperly stowed by a stevedore. The longshoreman sued the vessel owner for breach of the vessel’s, absolute and non-delegable duty to provide a seaworthy vessel. The Supreme Court held that the vessel was entitled to indemnity from the stevedore because the stevedore breached a duty of workmanlike performance implicit in its contract with the vessel which thereby subjected the vessel to absolute liability under the seaworthiness doctrine. Indemnity was felt necessary to relieve the vessel of onerous liability for an unseaworthy condition that arose “when the shipowner ... relinquished control of his vessel ... [to] another party ... [who was] better situated to prevent losses caused by shipboard injuries.” Hobart v. Sohio Petroleum Co., 445 F.2d 435, 438 (5th Cir.), cert. denied, 404 U.S. 942, 92 S.Ct. 288, 30 L.Ed.2d 256 (1971). In a subsequent decision, the Court made clear that, where Ryan indemnity applies, the indemnitee is not necessarily precluded from recovery by its own negligence; the right to indemnity is defeated in the traditional Ryan case by conduct on the vessel owner’s part which prevents the stevedore from performing in a workmanlike manner. See Weyerhaeuser S.S. Co. v. Nacirema Operating Co., 355 U.S. 563, 567, 78 S.Ct. 438, 440, 2 L.Ed.2d 491 (1958).

We have been reluctant to expand the Ryan doctrine beyond its facts. See, e.g., Hercules v. Stevens Shipping Co., 698 F.2d 726, 738 n. 25 (5th Cir.1983) (en banc) (“This circuit has reluctantly expanded the application of the Ryan doctrine.”); Gator Marine Service v. J. Ray McDermott & Co., 651 F.2d at 1100 (refusing to extend Ryan doctrine to disputes between vessels and stevedores over damaged cargo); Verrett v. McDonough Marine Serv., 705 F.2d 1437 (5th Cir.1983) (finding no basis for application of WWLP indemnity in towage context). We have steadfastly held, for example, that Ryan indemnity applies only when the indemnitor’s breach of the WWLP “renders the vessel unseaworthy *1168 and imposes absolute liability on the owner for the consequences of that unseaworthiness.” Sandoval v. Mitsui Sempaku, 460 F.2d 1163, 1168 (5th Cir.1972). See also Hobart v. Sohio Petroleum Co., 445 F.2d at 435 (refusing to extend Ryan indemnity to benefit shipper who owed duty of ordinary care, not absolute duty to provide seaworthy vessel). Still, we have applied the doctrine outside of the stevedoring context in which it arose. In fact, our cases have at least suggested that Ryan indemnity is potentially available in the context now before us: a third-party action by a vessel owner against the vessel’s builder and the supplier of certain of its component parts. See, e.g., Haupt v. Atwood Oceanics, Inc., 681 F.2d 1058, 1061 (5th Cir.1982) (indemnity action by vessel owner against supplier of component parts for vessel’s tensioner system); Leckelt v. Superior Oil Co., 608 F.2d 592 (5th Cir.1979) (indemnity action by vessel owner against manufacturer of vessel); Williams v. Brasea, Inc., 497 F.2d 67 (5th Cir.1974) (indemnity action by vessel owner against shipbuilder and designer-manufacturer of ship’s winch assembly), cert. denied, 423 U.S. 906, 96 S.Ct. 207, 46 L.Ed.2d 136 (1975); Parfait v. Jahncke Service, Inc., 484 F.2d 296 (5th Cir.1973) (indemnity action by vessel owner against welder who came aboard to repair pump engine), cert. denied, 415 U.S. 957, 94 S.Ct. 1485, 39 L.Ed.2d 572 (1974); Grigsby v. Coastal Marine Service, 412 F.2d 1011 (5th Cir. 1969) (indemnity action by vessel charterer against ship repairer), cert. denied, 396 U.S. 1033, 90 S.Ct. 612, 24 L.Ed.2d 531 (1970).

The Ryan doctrine, which is premised on an “equitable spirit,” Whisenant v. Brewster-Bartle Offshore Co., 446 F.2d 394, 400 (5th Cir.1971), to place “the burden ultimately on the company whose default caused the injury,” Italia Societa v. Oregon Stevedoring Co., 376 U.S. 315, 324, 84 S.Ct. 748, 754, 11 L.Ed.2d 732 (1964), requires the district court to weigh the conduct of the putative indemnitor and indem-nitee to determine “(1) whether the WWLP ... was breached; (2) whether that breach proximately caused the injury;” and (3) whether the indemnitee’s conduct is sufficiently blameworthy to preclude indemnity. Parfait v. Jahncke Serv., 484 F.2d at 296.

Assuming that Ryan indemnity is applicable in a seaman’s case like this one, see Leckelt v. Superior Oil, 608 F.2d at 593 n. 1, and further assuming that implied contractual indemnity principles survive our decision in Loose, 670 F.2d at 502 n. 18 (“Both contractual indemnity and implied indemnity ... continue to supplement comparative negligence allocation.”), we nevertheless agree with the district court that Phoenix is not entitled to Ryan indemnification. The district court did not specifically address each of the elements of WWLP indemnity. The court apparently assumed that Branham and Crown Rig breached a WWLP, which proximately caused the accident, and focused solely on whether Phoenix’s conduct precludes indemnity. In the classic Ryan case, negligence alone does not bar indemnity; “only conduct on the part of the shipowner which prevents the stevedore’s workmanlike performance” will preclude recovery. Agrico Chemical Co. v. M/V Ben W. Martin, 664 F.2d 85, 93 n. 3 (5th Cir.1981). There is no evidence here that Phoenix prevented Branham from supplying an adequate cotter key or prevented Crown Rig from adequately inspecting the monkey board assembly. We think, however, that Phoenix’s conduct nevertheless precludes Ryan indemnity.

This is not a classic Ryan case in which a vessel owner, by entrusting his vessel to a contractor who rendered the vessel unsea-worthy, has been subject to absolute liability without directly contributing to the un-seaworthy condition that caused the accident. Instead, we are faced with an accident that was proximately caused by the concurrent negligence of a designer who supplied an inadequate part, an erector who did not adequately inspect that part, and a vessel owner who operated the vessel in a negligent manner that accelerated the failure of that part. The district court found that the jarring of the monkey board, which Phoenix was negligent to allow to continue, “ultimately resulted” in a *1169 fall of the jack handle; rendered the accident far more likely, and proximately caused Bass’ injuries. The jack handle would not have fallen on May 8, 1980, absent the concurrence of the negligence of Branham, Crown Rig and Phoenix: Phoenix’s conduct is clearly sufficient to preclude Ryan indemnity. See Coffman v. Hawkins & Hawkins Drilling Co., 594 F.2d 152, 156 (5th Cir.1979) (denying indemnity to shipowner who “maintained joint control over injury-producing equipment with the contractor and thus shared the power to minimize the risk”); Agrico Chemical Co. v. Martin, 664 F.2d 85 (5th Cir.1981) (denying indemnity where concurrent negligence caused injury); Gator Marine Serv. v. J. Ray McDermott, 651 F.2d at 1096 (denying indemnity where concurrent negligence caused injury).

VI. DAMAGES.

Crown Rig attacks the district court’s award to Bass of $650,000 in damages. Our review of this claim, of course, is limited by the clearly erroneous standard. We review a trial court’s damage award, like a jury’s verdict on damages, according to the maximum recovery rule: “A verdict [or finding] is excessive if it is greater than the maximum amount a trier of fact could properly have awarded.” Sosa v. M/V Lago Izabal, 736 F.2d 1028, 1035 (5th Cir.1984). In other words, a damage award is not excessive simply because we, as triers of fact, might have awarded a lesser sum. Id. We cannot say on these facts that Bass received more than a trier of fact could properly have awarded.

VII. CONCLUSION.

We affirm the district court’s liability and damage findings. For the reasons set forth above, however, we reverse the district court’s partial abrogation of the Bass-Phoenix settlement and remand for further proceedings consistent with this opinion.

AFFIRMED IN PART, REVERSED IN PART, AND REMANDED.

Notes

1

. Bass sued Phoenix Seadrill/78 Ltd., his employer, and Phoenix Management Corporation, its managing general partner. The two entities are collectively referred to as "Phoenix” throughout this opinion.

2

. As we explained in Wilkins v. P.M.B. Systems Eng'g, Inc., 741 F.2d 795, 798 n. 2 (5th Cir.1984), Mary Carter agreements, which owe their name to Booth v. Mary Carter Paint Co., 202 So.2d 8 (Fla.Dist.Ct.App.1967), generally include "a release of the plaintiffs cause of action in return for a settlement payment, along with a provision providing that the settling defendant would be reimbursed to some specified degree from any recovery the plaintiff received in a suit against another non-settling defendant.”

3

. The agreement assigns to Phoenix one hundred percent of the first $105,000 that Bass recovers from third parties. The next $146,000 of recovery is to be split evenly between Bass and Phoenix. Any sums recovered over $251,000 shall be retained by Bass. Thus, if Bass recovers at least $251,000, Phoenix receives a rebate of $178,000 and has effectively settled the case for $32,000.

4

. See Ryan Stevedoring Co. v. Pan-Atlantic S.S. Corp., 350 U.S. 124, 76 S.Ct. 232, 100 L.Ed. 133 (1956).

5

. Arguably, the last of these grounds — the overreaching of Bass — was raised sua sponte by the court. The written motion to void the settlement, Record Vol. II at 309, avers generally that the agreement contravenes public policy. Although it also describes generally the economics of the agreement, the motion does not specifically allege that Bass was overreached. On appeal, Branham and Crown Rig expressly rely on the agreement’s alleged unfairness to Bass.

6

. Our conclusion that the court’s liability and damage findings are not clearly erroneous, see parts IV & VI, infra, disposes of the claim that knowledge of the agreement skewed the court’s fact-findings. There is no reason, therefore, to disturb the court’s conclusion that "the effective realignment of the parties by virtue of the Agreement did not affect the ability of the court to make accurate findings of fact.” Record Vol. II at 450. We are careful to note, however, that nothing we say in this opinion is intended to lessen a district court’s discretion, through disclosure of Mary Carter agreements to the jury or otherwise, to ensure that the trier of fact is not deceived by the effects of Mary Carter settlement techniques. Compare Reichenbach v. Smith, 528 F.2d 1072 (5th Cir.1976) (finding no prejudicial error in trial court’s refusal to disclose Mary Carter agreement to jury) with Daniel v. Penrod Drilling Co., 393 F.Supp. 1056 (E.D.La.1975) (granting new trial because of prejudicial effect of undisclosed Mary Carter agreement). We share the concern that some Mary Carter agreements may in fact skew the fact-finding process; that concern is simply not implicated in this case.

7

. The section of the court’s opinion which characterizes the agreement as champertous does purport to fashion a federal common law of champerty for maritime cases under which any Mary Carter agreement which assigns the settling defendant a portion of plaintiff's recovery from other defendants would presumably fail. In a supplemental order, the district court granted Phoenix's motion to "expunge the words ‘champerty’ [and] ‘officious intermeddling’ ” from its opinion; the court noted, however, that "the effect of such deletion is not to change the content and effectiveness" of the opinion. 573 F.Supp. 866. It seems to us that a conclusion that the word "champerty” is an inappropriate *1159 label on these facts necessarily means that the relationship which the word describes does not exist here. We think, therefore, that, notwithstanding its caveat, the effect of the supplemental order is to remove the champerty issue from the case. Moreover, even if the district court intended to do so, we are unwilling to pronounce a general ban on Mary Carter settlement techniques. Despite the potential for abuse, we think that properly disclosed Mary Carter agreements serve a legitimate function of providing litigants with capital with which to continue prosecution of their claims. Abuse of the device, we think, is more properly dealt with on a case-by-case basis.

8

. It is not clear from the panel opinion in Wilkins, or from the district court's opinion, 553 F.Supp. 201 (E.D.Tex.1982), whether the validity of the agreement was raised by the parties or decided on the court's own motion. See note 10, supra.

9

. Although the panel opinion in United States v. City of Miami, 614 F.2d 1322 (5th Cir.1980), was vacated by later en banc consideration, 664 F.2d 435 (5th Cir.1981) (en banc), we think the following passage is instructive:

In what can be termed "ordinary litigation,” that is, lawsuits brought by one private party against another private party that will not affect the rights of any other persons, settlement of the dispute is solely in the hands of the parties. If the parties can agree to terms, they are free to settle the litigation at any time, and the court need not and should not get involved. As Judge Wyzanski has described this situation: "the traditional view is that the judge merely resolves issues submitted to him by the parties ... and stands indifferent when the parties, for whatever reason commends itself to them, choose to settle a litigation." Heddendorf v. Goldfine, 167 F.Supp. 915, 926 (D.Mass.1958).
Moreover, procedurally it would seem to be impossible for the judge to become involved in overseeing a settlement, because the parties are free at any time to agree to a resolution of the dispute by private contractual agreement, and to dismiss the lawsuit by stipulation. In this situation, then, the trial court plays no role in overseeing or approving any settlement proposals.

Id. at 1330 (footnote omitted).

10

. We note at the outset that this case raises difficult questions of standing. The general rule, of course, is that "a non-settling defendant ... [who] is not prejudiced by the settlement ... has no standing to complain about the settlement.” In re Beef Industry Antitrust Litigation, 607 F.2d 167, 172 (5th Cir.1979), cert. denied, 452 U.S. 905, 101 S.Ct. 3029, 69 L.Ed.2d 405 (1981). See also In re Nissan Motor Corp. Antitrust Litigation, 552 F.2d 1088, 1103 n. 17 (5th Cir.1977) (non-settling defendants "were not entitled to object to the merits of the proposed settlement”); Liddell v. State of Missouri, 731 F.2d 1294, 1315 (8th Cir.) (same), cert. denied, — U.S. —, 105 S.Ct. 82, 83 L.Ed.2d 30 (1984). We doubt, therefore, whether Branham and Crown Rig have standing to vicariously assert the agreement’s unfairness to Bass and thereby subject the agreement to the strict scrutiny owed a seaman’s release. Moreover, Bass himself did not attack the settlement agreement until he filed his appellate brief. Under settled principles of appellate review, we generally do not consider claims raised for the first time on appeal. E.g., Masat v. United States, 745 F.2d 985 (1984 5th Cir.). Although there are exceptions to this rule, it seems to us manifestly unjust, absent allegations of fraud, duress or other circumstances vitiating assent, to allow a settling plaintiff to postpone his attack on the adequacy of consideration for a partial release until after a trier of fact has passed on both liability and damages. Cf. Leger v. Drilling Well Control, 592 F.2d at 1251 (noting “aleatory nature” of settlement process); Goehring v. Diamond Milling Co., 461 F.2d 77 (3d Cir.1972) (plaintiff cannot set aside a joint tortfeasor's release after waiting until all the evidence is in).

We think, therefore, that the district court’s analysis of the fairness of this agreement to Bass was proper only if a district court’s discretion to review and enforce settlement agreements allows it to sua sponte reject a settlement which has not been attacked by one with standing to object. The subject of a district court’s discretion to review and enforce settlements has arisen in our cases in the following circumstances: (1) judicial approval of an agreement, such as a class action settlement, which by statute or rule must be independently evaluated by the court, see, e.g.. In re Corrugated Container Antitrust Litigation, 643 F.2d 195 (5th Cir.1981); (2) an attempt to enforce a settlement where one party has breached or repudiated the agreement, see, e.g., Mid-South Towing Co. v. HarWin, Inc., 733 F.2d 386 (5th Cir.1984); (3) an attempt to avoid a prior release set up as an affirmative defense in a pending action, see, e.g., Halliburton v. Ocean Drilling & Exploration, 620 F.2d 444 (5th Cir.1980); and (4) an attempt by a non-party to the settlement to void portions of the agreement that purport to affect the rights of non-settlors, see, e.g., Dunn v. Sears & Roebuck & Co., 639 F.2d 1171 (5th Cir.), modified, 645 F.2d 511 (5th Cir.1981). Despite the solicitude with which we view seamen's releases, they do not fall within the category of settlements requiring independent judicial review for fairness. See note 12, supra and accompanying text. Our review of this precedent reveals no case involving "ordinary litigation,” see note 9, supra, or a seaman’s release, in which a district court has sua sponte raised the validity of the agreement. In light of our disposition of this *1161 case, we assume that a district court has the inherent authority to do so. We thereby pretermit difficult questions of standing.

11

. These cases generally involve concerns not present in a seaman’s suit. A class action settlement, for example, may bind absent class members. See, e.g., In re Corrugated Container Antitrust Litigation, 643 F.2d at 195.

12

. In fact, the local rules in at least one of the districts of this circuit expressly prohibit a trial judge from independently reviewing a seaman’s settlement in a pending suit. Rule 9.9 of the Local Rules of the Eastern District of Louisiana provides, in pertinent part:

(b) As to those [seaman's] cases which constitute a legitimate and bona fide case at the time of filing and in which the parties have agreed to a compromise at some stage prior to trial, the Court if requested, but only if requested, will consider the matter upon filing with the Court a joint motion for approval of the compromise.
******
(d) Although court intervention is not necessary in order for parties to effect a compromise and settlement of their claim, if it is their desire to obtain this Court’s approval of such, they must follow the above procedure,

(emphasis in original). It is not uncommon, of course, for the parties to a seaman’s settlement to seek judicial approval of a release. See, e.g., Wink v. Rowan Drilling Co., 611 F.2d at 98 (friendly Jones Act suit filed for purposes of obtaining judicial approval of a settlement).

13

. Our research reveals only one modern case voiding a seaman’s settlement for unfair consideration alone. See Wilkins v. P.M.B. Systems, 553 F.Supp. at 201. On appeal, we did not reach the validity of the court’s action with respect to the settlement. 741 F.2d at 798.

14

. We are not altogether convinced that Bass in fact received inadequate consideration for the rights he relinquished in the agreement. The district court's analysis, undertaken at a time when liability and damages were fixed, focuses solely on the dollars involved. While the agreement rendered it possible that Phoenix’s total settlement cost would be reduced to $32,000, it also guaranteed that Bass would receive at least $210,000. Moreover, it provided Bass with both funds and additional expertise with which to continue prosecution of his claims against Crown Rig and Branham. At any rate, in light of our disposition of the case, we need not decide if the consideration received was in fact inadequate.

15

. The standard of review, of course, depends upon which of these conclusions the district court has bottomed its decision. To the extent the district court voided the agreement for inadequate consideration alone, it abused its discretion, see, e.g., Matter of Aweco, Inc., 725 F.2d 293 (5th Cir.) (abuse of discretion standard governs review of approval or rejection of settlement), ce rt. denied, — U.S. —, 105 S.Ct. 244, 83 L.Ed.2d 182 (1984). To the extent the district court voided the agreement because Bass did not understand his rights, it was clearly erroneous, see Fed.R.Civ.P. 52(a).

16

. We again note that the timing and nature of the district court’s ruling makes this a difficult issue to review. It is manifest to us that, when a district court becomes concerned with the validity of a seaman's settlement agreement, the better practice would be to hold a hearing at that time at which the proponent of the agreement can demonstrate, as is its burden, that the seaman executed the agreement with a full understanding of his rights. The court’s failure to do so here, together with its apparent failure to consider the issue based on the trial transcript, leaves us with the unattractive option of *1163 remanding the case for that purpose at a time when liability and damages have been fixed by a full trial on the merits and the seaman has been provided with an economic incentive to repudiate the agreement. Thankfully, because of the record in this case, such a remand is not necessary.

17

. Bass testified as follows:

Q. Now, did you settle a portion of this case, Ronnie, after a period of time?
A. Yes, sir, I did.
******
Q. I showed you a copy of this, in fact, last night didn’t I, Mr. Bass?
A. Yes, sir, you did.
Q. And you looked it over at that time?
A. Yes, sir, I did.
Q. Now you signed it?
A. Yes, sir, I did.
Q. Prior to the execution by you of this instrument, did I go over it with you line by line?
A. Yes, sir, you did.
Q. Did I explain to you exactly what you were doing?
A. Yes, sir, you did.
Q. Did you voluntarily execute it?
A. Yes, sir, I did.
Q. Do you understand now and did you understand then that the essence of this settlement agreement was that you received $210,000.00?
A. Yes, sir.
Q. And that in the event, I explained to you, the Court that tried the case awarded other damages, that out of the first $157,-000.00 that Phoenix Seadrill would be reimbursed, they would get the first $157,-000.00?
A. Yes, sir, you told me that.
Q. And you executed it voluntarily, knowing that?
A. Yes, sir, I did.
Q. Do have any regret about that?
A. No, sir, I do not.
Q. Now, after you received your settlement, Ronnie, did you buy an automobile?
A. Yes, sir, I did.
Q. Did you try out several types of cars before you bought one? Or did you just go buy one?
A. No, sir, I tried several out.
Q. And what kind of automobile did you buy?
A. I bought a 1981 Corvette.

Record Vol. III at 58-59.

18

. We are not altogether convinced that the possibility of settlement with Branham and Crown Rig is rendered less likely, because of the veto provision, than it would have been if Phoenix lacked the ability to reject settlements. We note first that the veto provision does not give Phoenix the power to require that Bass settle on terms offered by Branham or Crown Rig; it simply states that Bass cannot accept a settlement unless Phoenix consents; “No settlement ... [may] be effected without the joint consent of the parties to this agreement.” Record Vol. I at 189. It seems to us that Bass’ economic self-interest would dictate rejection of the very same settlements that Phoenix would be economically motivated to reject.

19

. Like the panel in Wilkins, 741 F.2d at 798 n. 2, we do not wish to imply that, if we had affirmed its abrogation of the rebate and veto provisions, we would have also approved of the district court’s enforcement of the release. We have grave doubts whether a district court faced with an unenforceable settlement provision can simply write the offensive provision out of the agreement and effectively bind the parties to a deal they never contemplated. While the court may have the authority to refuse enforcement of a severable, ancillary provision, while nonetheless enforcing the main agreement, see Dunn v. Sears, Roebuck & Co., 639 F.2d at 1173, we do not imply that a court can gut the agreement and, in effect, force the parties to settle on the court’s terms, rather than their own terms.

20

. The monkey board and crown block are components of the derrick used in raising and lowering pipe. The monkey board is a small platform upon which a worker stands. The crown block is a 12,000 pound moveable structure which travels to pick up pipe and stabilizes the pipe before it is lowered into the hole that is being drilled. The crown block on Big Foot I rammed into the monkey board during drilling operations.