SELYA, Circuit Judge.
Ralph W. Moores, Jr., plaintiff, was injured while laboring as a longshoreman in Maine. After collecting compensation benefits from the stevedoring firm for which he worked -- benefits actually paid by that firm's insurer, Liberty Mutual Insurance Company (LMIC) -- he brought a third-party liability suit against the shipowners in Maine's federal district court. Nathan Greenberg was his attorney. They agreed that the lawyer's compensation would be contingent: the standard one-third of any judgment or settlement. But, the case was lost.
Moores wasted little time in turning upon his erstwhile champion. He sued for malpractice in a Massachusetts state court. Greenberg removed the case to the United States District Court for the District of Massachusetts.1 Following a jury trial, Moores was awarded $12,000. Although both parties assign error, we find no reason to forsake the verdict.
I. THE LAW OF THE CASE.
We pause briefly to reflect on the source of the applicable law. The district court, sitting in diversity jurisdiction, was duty bound to use state substantive law. Erie R. Co. v. Tompkins, 304 U.S. 64, 78, 82 L. Ed. 1188, 58 S. Ct. 817 (1938). The litigants acknowledge that Maine law pertains.2 And, since no Maine court of record has spoken to certain of the issues before us, it becomes our duty to vaticinate how the state's highest tribunal would resolve matters. Nature Conservancy v. Machipongo Club, Inc., 579 F.2d 873, 875 (4th Cir.), cert. denied, 439 U.S. 1047, 58 L. Ed. 2d 706, 99 S. Ct. 724 (1978). See Moosehead Sanitary Dist. v. S.G. Phillips Co., 610 F.2d 49, 53 (1st Cir. 1979). "In undertaking this forecast, the court must look to relevant, i.e., analogous, state court decisions, and may assay sister state adjudications of the issue." Plummer v. Abbott Laboratories, 568 F. Supp. 920, 922 (D.R.I. 1983) (citations omitted). In the process, we "may reasonably assume that [the state court] will follow the rule that appears best to effectuate" relevant policies. Bowen v. United States, 570 F.2d 1311, 1322 (7th Cir. 1978).3
II. THE LAWYER'S APPEAL.
Greenberg's appeal is a divaricated one. We travel each fork separately.
A. The Directed Verdict Motion. Greenberg contests the district court's denial of his motion for a directed verdict. He asserts that the evidence presented was insufficient as a matter of law to support liability. Given the operative facts of the case, however, this asseveration need not occupy us for long.
We recently have had occasion to summarize the principles which steer appellate oversight of rulings on directed verdict motions:
Wagenmann v. Adams, 829 F.2d 196, 200 (1st Cir. 1987) (citations omitted). Greenberg has not come close to meeting this rigorous standard.
At a very minimum, there was evidence before the jury which, if believed, proved that while the third-party suit was in progress, the shipowners offered to settle first for $70,000 and later for $90,000. There was also evidence that Greenberg failed to relay either offer to plaintiff. In the subsequent malpractice suit, Moores claimed that he would have accepted the $90,000 offer had he been informed of it. Instead of a fat settlement, he received nothing but a rebuff from the jury.
This evidence was, we think, more than ample. In representing his client, an attorney has a duty to use that degree of skill, diligence, and judgment ordinarily to be expected of a member of the bar practicing in the same (or a similar) locale. Gans v. Mundy, 762 F.2d 338, 341 (3d Cir.), cert. denied, 474 U.S. 1010, 88 L. Ed. 2d 467, 106 S. Ct. 537 (1985); Kuehn v. Garcia, 608 F.2d 1143, 1147 (8th Cir. 1979), cert. denied, 445 U.S. 943, 63 L. Ed. 2d 777, 100 S. Ct. 1340 (1980); Palmer v. Nissen, 256 F. Supp. 497, 501 (D. Me. 1966) (applying Maine law); Wilson v. Russ, 20 Me. 421, 424 (1841). As part and parcel of this duty, a lawyer must keep his client seasonably apprised of relevant developments, including opportunities for settlement. See Joos and Avery v. Drillock, 127 Mich. App. 99, 106, 338 N.W.2d 736, 739-40 (1983).
Greenberg says that, even if this be true, the sums mentioned to him were too niggardly to be relayed. We need not decide today whether a lawyer has an obligation to transmit a patently unreasonable offer to his client. See Smith v. Ganz, 219 Neb. 432, 436, 363 N.W.2d 526, 530 (1985). The overtures which the defense made in the liability case were neither so totally divorced from a realistic appraisal of the merits nor so unresponsive to the upside and the downside of the litigation that they could blithely be ignored. The ongoing risk/reward calculus had many variables, some of an imponderable nature. These manifold uncertainties added up to at least one bit of certitude: the shipowners' $90,000 offer could not be said, as a matter of law, to be a patently ridiculous one. On this scumbled record, the district court did not err in permitting the jury to determine whether reasonably competent counsel would have informed Moores of the $90,000 offer and whether the client, had he been told, would have clasped it to his bosom.
B. The Jury Trial. The defendant's remaining assignment of error is of a procedural bent. Fed. R. Civ. P. 38(b) provides that a party "may demand a trial by jury . . . by serving upon the other parties a demand therefor in writing . . . not later than 10 days after the service of the last pleading." Greenberg asserts that, inasmuch as no timely Rule 38(b) demand was served, the district court erred in allowing a jury to decide the case. This contention requires careful perscrutation of the record.
When Greenberg removed the malpractice action from state court and invoked federal jurisdiction, it became his responsibility to complete the civil cover sheet established in the district court clerk's office as a guide to docketing the case. In filling out this form, Greenberg (erroneously) checked the box which indicated that a jury trial had been claimed. As the case approached trial-readiness, plaintiff's attorney wrote to the court asking that it be placed on the jury docket. The district judge, an experienced hand, recognized that the scribbling on the cover sheet did not satisfy the literal requirements of Rule 38(b). See Omawale v. WBZ, 610 F.2d 20, 22 (1st Cir. 1979) (per curiam). He treated counsel's letter as a Rule 39(b) request,4 however, and subsequently granted it.
Greenberg challenges the applicability of Rule 39(b) because, he says, no motion was brought. The plea is unavailing. There are no precise formalities for a Rule 39(b) motion. The language and history of the rule indicate that the requirement was inserted by the draftsmen simply as a means of preventing the district court from ordering jury intervention sua sponte when the parties were unanimous in their preference for a bench trial. See 5 J. Moore, W. Taggart & J. Wicker, Moore's Federal Practice para. 39.08 (2d ed. 1985). In our view, Rule 39(b) may be animated by any cognizable affirmation of one party's intent to avail himself of a jury trial. The letter from Moores's lawyer to the court was clearly such an affirmation and was properly processed under Rule 39(b). And there was no prejudice: Greenberg was told of the court's intention to treat the correspondence in this fashion, and was allowed an opportunity to argue in favor of a nonjury trial.
The second string to the defendant's bench trial bow calumnizes the granting of the motion. Yet, the trial court's "discretion under Rule 39(b) is very broad," so much so that it "would be very rare" to find an abuse of discretion in either denying or granting a Rule 39(b) motion. Rowlett v. Anheuser-Busch, Inc., 832 F.2d 194, 199 (1st Cir. 1987). Surely, this instance is not the exception. The cover sheet, with its mistaken indication that a jury had been demanded, can reasonably be viewed as having lulled plaintiff into a false sense of security on the point. The parallel to Pinemont Bank v. Belk, 722 F.2d 232 (5th Cir. 1984), is striking. There, as here, one party had erroneously checked the jury box on the cover sheet. There, as here, the adverse party belatedly discovered the fact and requested that a jury be drawn. The district court denied the Rule 39(b) motion. The Fifth Circuit vacated the ensuing judgment, holding that it was an abuse of discretion to refuse the out-of-time jury request. Id. at 235-38. The court noted, in language which (apart from the inversion of plaintiff and defendant) might have been custom tailored for the case at bar:
We believe . . . that the effect which the mismarked cover sheet had on [defendant] -- to make him believe a [jury] demand had been made -- was probably not unusual. Neither party is free from responsibility for this confusion, but we believe that the [plaintiff's] act of mismarking the cover sheet, which instigated the confusion, is, on balance, more blameworthy than [defendant's] reliance on it. The mismarked pleading circumstance makes us somewhat sympathetic to [defendant's] plight and to the tardiness of his jury trial motion.
Id. at 236. We see Belk as a well-reasoned precedent closely in point. Given the circumstances, the district court did not abuse its discretion in allowing Moores's late request for a jury trial.5
III. THE CLIENT'S APPEAL.
Moores launches two lightening bolts in his effort to incinerate the verdict, both of which relate to damages. From our coign of vantage, these fulgurations strike well wide of the mark.
A. Accounting for the Contingent Fee.
We concisely clarify the critical contours of the contingency conundrum. Had Moores been told of, and accepted, the offer to settle the third-party suit, his proceeds would have been reduced by a $30,000 counsel fee, i.e., by one-third of the gross recovery.6 At the ensuing malpractice trial, the district court instructed the jurors, over objection, that if liability was resolved against Greenberg, they should award Moores only his realizable net proceeds, subtracting the fee. The verdict mirrored this sort of trimming. The instruction, Moores tells us, was error.
We look first to the doctrinal foundation of the action. The amended complaint alleged theories in contract, tort, and malpractice, all of which boiled down to a common thought: that the lawyer was liable for neglecting to communicate the $90,000 offer to the client. In negligence, Moores was entitled to recover "only those damages which [were] a foreseeable consequence of [the] defendant's negligence." Stubbs v. Bartlett, 478 A.2d 690, 693 (Me. 1984); Brewer v. Roosevelt Motor Lodge, 295 A.2d 647, 652 (Me. 1972). Exclusive of other considerations, see supra n.6 and infra Part III-B, that would have amounted to $60,000 ($90,000-$30,000 = $60,000). The jury could also have awarded the plaintiff any other damages actually caused by the negligence, see Wendward Co. v. Group Design, Inc., 428 A.2d 57, 61-62 (Me. 1981), but Moores proved none.7
In contract, the end result is much the same. Maine law provides for damages adequate to "place plaintiff in the same position he would have been in had there been no breach." Forbes v. Wells Beach Casino, Inc., 409 A.2d 646, 654 (Me. 1979); Spitz v. Lamport, 119 Me. 566, 568-69, 112 A. 522 (1921). Had Greenberg not broken the faith, Moores -- exclusive of other considerations -- would have netted $60,000 after paying Greenberg. To replicate plaintiff's position, therefore, the jury should have awarded him only that much. And, although Moores could have recovered such added damages as were "reasonably within the contemplation of the contracting parties when the agreement was made," Winship v. Brewer School Committee, 390 A.2d 1089, 1095 (Me. 1978), it was his burden to introduce evidence of such items. Dairy Farm Leasing Co. v. Hartley, 395 A.2d 1135, 1138 (Me. 1978); McDougal v. Hunt, 146 Me. 10, 14, 76 A.2d 857, 860 (1950). He offered none.
Finally, there is no reason to believe that a Maine court, under a "legal malpractice" rubric, would award damages beyond those available under conventional tort and contract theories. After all, the conceptual foundation on which legal malpractice rests is excerpted from precisely such common law underpinnings. Short of punitive damages -- and none were granted in this case -- "an attorney who [commits malpractice] is liable to his client for any reasonably foreseeable loss caused by his negligence." Fishman v. Brooks, 396 Mass. 643, 646, 487 N.E.2d 1377, 1379 (1986). On this record, it was "reasonably foreseeable" that, by failing to communicate the offer, Greenberg would effectively deprive his client of the net benefit of the tendered bargain -- nothing more.
In terms of a plaintiff's theoretical entitlements, all of the travelled roads lead in the direction of Rome. No matter which of plaintiff's several theories is seen as controlling, logic suggests that Moores cannot recover the fee equivalent in the present action. Unless he can present some cognizable basis for receiving from the lawyer more than he would have netted from the tortfeasors, his assignment of error cannot be credited. It is against this backdrop that we consider plaintiff's caselaw-dependent argument.
There are, as Moores advertises, several cases which suggest that one victimized by legal malpractice should be more generously treated. See, e.g., Andrews v. Cain, 62 A.D.2d 612, 613, 406 N.Y.S.2d 168, 169 (1978) (abjuring deduction for original contingency fee in computation of damages); Strauss v. Fost, 213 N.J. Super. 239, 242, 517 A.2d 143, 145 (1986) (same); Togstad v. Vesely, Otto, Miller & Keefe, 291 N.W.2d 686, 695-96 (Minn. 1980) (same). But see McGlone v. Lacey, 288 F. Supp. 662, 665 (D.S.D. 1968) (fee to be deducted in computing damages) and Sitton v. Clements, 257 F. Supp. 63, 65 (E.D. Tenn. 1966), aff'd, 385 F.2d 869 (6th Cir. 1967) (same). The cases which encourage fee disregard are not, however, persuasive. To the extent that they undertake any analysis, only three justifications are advanced:
1. That a negligent attorney should not "collect" a fee for his shoddy workmanship. E.g., Strauss v. Fost, 517 A.2d at 145; Andrews v. Cain, 406 N.Y.S.2d at 169.
2. That since a plaintiff must pay his attorneys in the subsequent malpractice action, disregarding the original lawyer's fee when calculating damages "cancels out" the extra cost. E.g., Togstad, 291 N.W.2d at 695-96.
3. That attorneys' fees incurred in the malpractice action are recoverable as consequential damages of the negligent lawyering. E.g., Foster v. Duggin, 695 S.W.2d 526, 527 (Tenn. 1985).
In our view, the first two of these arguments cannot withstand scrutiny -- and we need not reach the last.
Restricting the client's recovery in a follow-on malpractice action to the realizable net proceeds from his earlier case does not allow a culpable attorney to "collect" anything. More importantly, the argument to the contrary overlooks that the fundamental purpose of such damages is to compensate a plaintiff, not punish a defendant. Cordeco Dev. Co. v. Santiago Vasquez, 539 F.2d 256, 262 (1st Cir. 1976). Moores's professed fear that professional irresponsibility will be encouraged by focusing upon net recoveries is wholly speculative. The loss of custom and reputation, the availability of compensatory damages, and the prospect of exemplary damages in appropriate cases, provide an array of disincentives which far outstrip this one.
The second proposition is equally unconvincing. It is true that a victimized client will ordinarily hire successor counsel and will incur added expense in pursuing an action against his quondam lawyer.8 But, the assertion that the fees originally to be paid should not be deducted from a malpractice award because the client will then pay twice for the "same" services assumes what it sets out to determine: that plaintiff is entitled to recover the attorneys' fees. To that extent, the argument is an essentially circular ipse dixit ; it supplies no cognitive basis for the result urged by Moores. Moreover, the general rule in the United States, unlike in England, is that each suitor bears his own lawyering costs. Alyeska Pipeline Service Co. v. Wilderness Society, 421 U.S. 240, 248, 44 L. Ed. 2d 141, 95 S. Ct. 1612 (1975). In the absence of a statute, an enforceable agreement, or a recognized juridical exception to the general rule, counsel fees do not accrue in favor of a successful litigant. Id. at 269. In a judicial system which refuses routinely to shift attorneys' fees as a form of incidental damages, it makes little sense to award them by indirection. By barring a jury from considering the antecedent contingent fee obligation when deciding damages in a follow-on malpractice action, we would accomplish exactly that.
The third suggested rationale for deemphasizing a "realizable net proceeds" approach need detain us only momentarily. If one accepts the notion that counsel fees in a malpractice action should be viewed as proximately caused by the original attorney's negligence and therefore recoverable as consequential damages -- a matter as to which we express no opinion, see supra n. 7 -- Moores is not assisted. Whatever form a legal malpractice action takes, the plaintiff has the burden of introducing evidence to justify an award of consequential damages. Wendward Co., 428 A.2d at 61 (tort); Dairy Farm Leasing Co., 395 A.2d at 1138 (contract); Chiaffi v. Wexler, Bergerman & Crucet, 116 A.D.2d 614, 615, 497 N.Y.S.2d 703, 704 (1986) (malpractice). In this instance, Moores offered no proof as to the fees of his newly-retained lawyers.
Nor is there any merit in the suggestion that the original contingency agreement between Moores and Greenberg, without more, can serve as a proxy for actual evidence of consequential damages. There is neither a guarantee nor a likelihood that the reasonable cost of securing competent counsel in a follow-on malpractice action will match (or even approximate) that incurred or anticipated in the original action. The range of matters which lawyers may handle for clients on a contingency basis is close to infinitely variable. Greenberg was engaged to prosecute claims for maritime tort and unseaworthiness, but conceptually, the engagement could easily have been, say, to collect on a negotiable instrument or to redress patent infringement. Of necessity, the fee agreements between attorney and client in the earlier and later actions are largely unrelated. While mathematical certainty in the calculation of damages is unnecessary, McDougal, 76 A.2d at 860, there is no excuse for using an entirely conjectural estimate where an actual amount is easily discernible.
Were this a matter of federal law, we would end the discussion at this point. But it is not. Diversity jurisdiction carries with it an added set of responsibilities. A federal court which finds itself obliged to make an informed prophecy as to state substantive law in an area in which state courts have not spoken, has a duty, we think, to keep its forecast within the narrowest bounds sufficient to permit disposition of the actual case in controversy. Although we believe that the focus on a plaintiff's realizable net proceeds compels deduction of the hypothetical contingent fee in virtually any follow-on malpractice suit, and we suspect that the Maine courts would so hold, this case is a particularly strong one for application of the rule.
The majority of the decisions which -- contrary to our view -- espouse disregard of the deduction are inapposite to the situation at bar. We illustrate briefly. Some cases, like Benard v. Walkup, 272 Cal. App. 2d 595, 77 Cal. Rptr. 544 (1969), dwell on the uncertain dollar amount of counsel's remuneration. There, the contingent fee was on a variable scale and the attorney had failed to file suit within the limitations period. The court refused to require any deduction, remarking that "there is here no way in which we can ascertain what amount of damages would have been produced by full performance of the contract on both sides." 77 Cal. Rptr. at 551. See also Togstad, 291 N.W.2d at 696 (disallowing deduction of "hypothetical contingency fee"); Andrews, 406 N.Y.S.2d at 169 (similar); Duncan v. Lord, 409 F. Supp. 687, 691-92 (E.D. Pa. 1976) (dicta; similar).9
The case before us, however, is altogether different. There was nothing uncertain or problematic about Greenberg's fee. The amount of the gross recovery was fixed: had the lawyer not breached his duty, $90,000 would have been paid. This is a far cry from cases like Andrews v. Cain, supra, where the attorney failed to file suit -- leaving up in the air the question of how much money, if any, his client would have recovered. See also Duncan v. Lord, supra (similar; case involuntarily dismissed for want of prosecution due to counsel's laggardness in answering interrogatories). And the fee arrangement between Moores and Greenberg -- a straight one-third -- was equally definite; had the offer been communicated and accepted, counsel's recompense would have been $30,000. This contrasts sharply with cases such as Andrews, where the court believed "it was impossible to determine what the deduction from plaintiff's award would have been." 406 N.Y.S.2d at 169.
There is another important line of demarcation as well. Many of the cases hawked by Moores, typically ones in which a lawyer neglected to sue before a temporal deadline expired, stress the fact that the attorney-defendant had furnished no services to his client. E.g., Andrews, 406 N.Y.S.2d at 169; Benard, 77 Cal. Rptr. at 551 (defendant had "not established that [the lawyer] performed any part of the contract"). These "do-nothing" cases are also distinguishable. Where a lawyer accepts an engagement and thereafter fails to show up at the starting gate, e.g., id. (failure to file suit within statute of limitations); Winter v. Brown, 365 A.2d 381 (D.C. 1976) (failure to serve mandatory notice of claim within prescribed period), it is arguably equitable to fix damages without regard to a fee entitlement which would only have come into existence had the lawyer performed the contract. Those rough equities are in a different balance, however, where the lawyer -- notwithstanding that he was guilty of some breach of duty -- actually did the work. And the difference in the equities is heightened in a case like this one, where the sum in dispute -- the $90,000 offer -- arose during the trial, presumably in direct response to Greenberg's labors on his client's behalf. Cf. Strauss v. Fost, 517 A.2d at 145 ("We can envision cases where on a quantum meruit basis the efforts of a defendant attorney may have so benefited [sic] a plaintiff . . . that it would be unfair to deny" the deduction); Foster v. Duggin, 695 S.W.2d at 527 ("in an appropriate case, the attorney may be entitled to credit for expenses . . . which ultimately benefitted the client").
In our opinion, the Supreme Judicial Court of Maine, if confronted with the precise question now before us, would rule that where counsel's efforts produced an offer which he then wrongfully failed to relay to the client, the settlement sum should be reduced by the amount of the lawyer's pre-agreed contingent fee (if readily ascertainable) in calculating damages for legal malpractice. This, we think, represents the better-reasoned view of the applicable law and the view most consistent with Maine's expressed jurisprudence. Accordingly, we hold that the district court, on the facts of this case, properly charged the jury to deduct the fee from the offer in arriving at its verdict.
B. Accounting for the LMIC Lien. After his injury, Moores accepted compensation benefits under the Longshoremen's and Harbor Workers' Compensation Act, 33 U.S.C. §§ 901-50 (1982). Though an injured longshoreman may collect such benefits from his employer and nevertheless prosecute a third-party action for personal injuries, the compensation carrier is entitled to reimbursement if the third-party action proves successful. See Bloomer v. Liberty Mutual Ins. Co., 445 U.S. 74, 79-88, 63 L. Ed. 2d 215, 100 S. Ct. 925 (1980). To this extent, the insurer has a subrogation lien on the proceeds of the third-party claim. Id. It is conceded in this case that LMIC's lien was approximately $43,000 and that it would have attached to a settlement of Moores's third-party suit against the shipowners.10 The district judge charged that damages should be reduced by the face amount of the lien. The plaintiff objected contemporaneously and presses the point on appeal.
The objection has two facets. Moores urges that the lien arguably attaches to the proceeds of this action, so that the instruction threatens to take a double dip from his exchequer. Alternatively, he claims that the lienor, LMIC, might have been amenable to some compromise. Thus, the jury should have been free to determine the extent to which LMIC would likely have discounted the lien. We find these contentions to be jejune.
First and foremost, the compensation lien died of natural causes when the supposedly culpable third parties, the shipowners, were exonerated. Moores cites no authority for the proposition that such a lien has a life of its own. He admits that the insurer has not asserted any such claim. The physical injury stemming from the shipboard accident is separate and distinct from the legal injury worked by the lawyer's misfeasance. LMIC's right of reimbursement extends to the damages recoverable in consequence of the former -- and no further. The purpose of the lien law -- to prevent duplicative recovery for the same injuries and damages -- would not be furthered by stretching it to envelop the malpractice award. The damages granted to plaintiff in this case are free and clear of the reach of the statutory lien. Cf. Gartman v. Allied Towing Co., 467 F. Supp. 439, 440 (E.D. Va. 1979) (compensation provider cannot seek reimbursement of payments from proceeds of third-party suit for different injuries).
The plaintiff's second point -- that perhaps less than the face amount of the lien should have been offset -- founders on a straightforward matter of proof. "It is fundamental in [Maine] law that the plaintiff has the burden of proving his damages." Dairy Farm Leasing Co. v. Hartley, 395 A.2d at 1138. Accord Wendward Co. v. Group Design, Inc., 428 A.2d at 61; McDougal v. Hunt, 76 A.2d at 860. In this malpractice action, plaintiff bore the burden of proving, more likely than not, (a) that LMIC would have discounted the lien, and (b) by what amount. Only three witnesses testified directly on the issue. One said that LMIC would not consider negotiating anent the lien unless and until settlement was imminent. Another testified that, come what may, the carrier would insist upon full reimbursement. The third admitted that he was merely an internuncio; he did not know what decisions might or might not be made on such a subject. Other witnesses testified about general industry practices -- but in the most amorphous of terms. There was no evidence whatever as to the approximate percentage of reduction to be anticipated, or as to the dollar amount of the hoped-for compromise.
On this skimpy record, the trial court concluded that the "absence of evidence" on the issue was fatal to Moores's position. We agree. The nisi prius roll was devoid of anything which would have permitted the jury, without engaging in the rankest speculation, to conclude that LMIC would have accepted some definite sum less than $43,000 in satisfaction of its claim. For aught that was proven, plaintiff would have had to repay the entire compensation lien had he settled with the shipowners. In effect, $43,000 out of the $90,000 settlement would have been diverted to the carrier. Consequently, the district court did not err in charging the jury to make a like reduction in Moores's malpractice recovery.
We need go no further. We conclude that this suit was properly assigned to the jury calendar; that the evidence of negligence was adequate to warrant jury submission; and that the liability finding in plaintiff's favor was supportable. We conclude as well that the trial court's instructions on damages were fully consonant with principles of Maine law, as we interpret it. The verdict for $12,000 (the $90,000 offer less  the $30,000 hypothetical contingent fee,  the $5,000 in advanced costs, see supra n.6, and  the $43,000 LMIC lien) possessed satisfactory underpinnings in the record. Accordingly, neither of the cross appeals has merit. We therefore leave the parties as we found them.
Affirmed. No costs.
* Of the Fifth Circuit, sitting by designation.
1. When effected, the removal was improper. See 28 U.S.C. § 1441(b) (resident defendant not permitted to remove on diversity grounds). Yet, the district court would have had diversity jurisdiction over an original complaint; Moores was a citizen of Maine, Greenberg of Massachusetts, and more than $10,000 (exclusive of interest and costs) was in controversy. See 28 U.S.C. § 1332(a). The case having been fully tried and no party having persisted in an objection, jurisdiction has attached. Grubbs v. General Electric Credit Co., 405 U.S. 699, 702, 31 L. Ed. 2d 612, 92 S. Ct. 1344 (1972).
2. Where the parties agree what substantive law controls in a diversity case, we can -- and ordinarily should -- accept such a concession. Mathewson Corp. v. Allied Marine Indus., Inc., 827 F.2d 850, 853 n.3 (1st Cir. 1987).
3. We acknowledge that, in diversity jurisdiction, a federal court -- which has "no roving commission to superimpose federal choices upon the framework of state law," McInnis v. Harley-Davidson Motor Co., 625 F. Supp. 943, 956 n. 7 (D.R.I. 1986) -- should not play "Big Brother". Our mission is to "determine the rule that the [state] Supreme Court would probably follow, not fashion a rule which . . . an independent federal court might consider best." Lowe's North Wilkesboro Hardware, Inc. v. Fidelity Mut. Life Ins. Co., 319 F.2d 469, 472 (4th Cir. 1963). Still and all, if other signposts are blurred, a federal court "may assume that the state courts would adopt the rule which, in its view, is supported by the thrust of logic and authority." Stool v. J.C. Penney Co., 404 F.2d 562, 563 (5th Cir. 1968) (footnote omitted). Accord McInnis, 625 F. Supp. at 956 n. 7.
4. Fed. R. Civ. P. 39(b) reads as follows: Issues not demanded for trial by jury as provided in Rule 38 shall be tried by the court; but, notwithstanding the failure of a party to demand a jury in an action in which such demand might have been made of right, the court in its discretion upon motion may order a trial by a jury on any or all issues.
5. Greenberg also argues that a jury should not have been drawn "because of the strong prejudice against lawyers held by the public." Brief (Appeal No. 86-1599) at 9. The seventh amendment, however, contains no special swaddling for attorneys. In any event, Greenberg had the opportunity during voir dire to challenge any juror who displayed bias. This, in itself, was an adequate safeguard.
6. Greenberg also fronted some $5000 in litigation-related costs, reimbursable only out of the avails of the case. The jury was told, correctly we think, to treat the contingent fee and the expense advance identically. Therefore, what we say concerning the former applies equally to the latter.
7. Particularly, we note that plaintiff offered no proof as to his fee arrangement with successor counsel (the lawyers who essayed on his behalf to bring Greenberg to account). We leave for another day the question of the admissibility of such evidence if proffered. Cf., e.g., Foster v. Duggin, 695 S.W.2d 526, 527 (Tenn. 1985) ("The additional fees necessary to pursue this [legal malpractice] action are in the nature of incidental damages flowing from [the lawyer's] breach of the contract."); Winter v. Brown, 365 A.2d 381, 386 (D.C. 1976) (In legal malpractice action, clients' damages "include the cost of additional litigation in order to recover on their original claim").
8. A malpractice claimant can, of course, appear pro se when seeking redress from careless counsel. None of the decisions which advocate "cancelling out" earlier and later fees explain how damages would be computed in such a situation.
9. Moores rests more weight on Duncan than the decision will support. In that case, the lawyer-defendant "placed in the record no evidence of his fee arrangement with plaintiff," thus precluding definitive consideration of the "net realizable proceeds" point. 409 F. Supp. at 691. Another of Moore's flagship cases is similarly flawed. In Christy v. Saliterman, 288 Minn. 144, 179 N.W.2d 288 (1970), the issue was surfaced for the first time on appeal. 179 N.W.2d at 307. The Minnesota Supreme Court refused to entertain it, holding: "If defendant felt that there was some basis for diminished damages, he should have asserted that theory in the court below." Id. Notwithstanding this procedural default, the ensuing dictum in Christy, id. 179 N.W.2d at 307-08, has become the cornerstone of most of the caselaw upon which plaintiff relies.
10. We caution that the injury occurred, the lien attached, and the third-party suit was tried before the enactment of 33 U.S.C. § 933(f) (West. Supp. 1985). Thus, that statute -- requiring the employer to bear certain of the fees and expenses incurred by the injured employee -- had no bearing upon this case.