concurring in the judgment.
I concur in the judgment of the court’s opinion. I write separately because I am unable to adopt the majority’s analysis of Ferrell v. Trailmobile, Inc., 223 F.2d 697 (5th Cir.1955). In addition, I am compelled to address the implications of the external ambiguity doctrine and the relationship between Central Transport’s successor and guarantor liability.
I.
The majority notes that in United States v. McGaughey, 977 F.2d 1067, 1075-76 (7th Cir.*13161992), this circuit interpreted Ferrell as being an exception to the rule that a new trial cannot be granted under Rule 60(b) on the basis of new evidence unless proper diligence was used to secure that evidence before the original trial. While we declined to formally adopt the Ferrell rule in McGaughey, since it would not have affected the result in that case, id. at 1076, we had no difficulty understanding what Ferrell said. In McGaughey, we interpreted Ferrell as holding that if “practically conclusive evidence” is discovered after trial, a new trial can be granted even if the movant failed to exercise due diligence in obtaining that evidence before trial, in order to prevent a manifest miscarriage of justice. Id. at 1075-76. The majority now maintains that ‘Ferrell itself does not say this — at least not explicitly,” supra Op. at 1314.
The Ferrell case dealt with a judgment for the recovery of an allegedly unmade installment payment toward the purchase of a truck trailer. After judgment was granted for the plaintiff, the defendant came forward with photostatic copies of money orders for the payment, endorsed by the plaintiff, and even an affidavit from the auditor of the plaintiffs bank showing that the money orders were deposited. The Ferrell court concluded,
If, in fact, practically conclusive evidence shows that the appellant had actually paid all eighteen installments for the purchase of the trailer, it is obvious that the judgment should be set aside to prevent a manifest miscarriage of justice. In such a case, the ends of justice may require granting a new trial even through proper diligence was not used to secure such evidence for use at the trial.
Ferrell, 223 F.2d at 698 (citations omitted). This conclusion was based on principles of equity and was made without limitation to any particular kind of Rule 60(b) motion. Johnson Waste Materials v. Marshall, 611 F.2d 593 (5th Cir.1980) (“Our decision in Ferrell is fully justified by equitable considerations.”)
Rules 60(b)(2) and 60(b)(5) are exactly the same today as they were at the time of Ferrell,1 and the majority admits that Ferrell did not limit itself to a particular subsection of Rule 60(b). Yet the majority still attempts to cast Ferrell as a pure 60(b)(5) case. Actually, the Ferrell court’s focus on the nature of the new evidence and mention of “proper diligence” seems more to invoke 60(b)(2) than 60(b)(5). Further, I find unconvincing the majority’s suggestion that Johnson Waste held the Ferrell equitable exception not to apply to 60(b)(2). Although Johnson Waste tentatively sought to cast Ferrell as a 60(b)(5) case, id. at 599, and its implications are less than perfectly clear, Johnson Waste did not overrule Ferrell, nor did it specifically limit Ferrell to Rule 60(b)(5). Other circuits have read Ferrell and Johnson Waste as being consistent and as supporting a general exception to the Rule 60(b) due diligence requirement for the granting of relief from a judgment due to newly discovered evidence.2
*1317Yet all of this is surely not to say that the Ferrell holding should be accepted as consistent with Rule 60(b)(2). Rule 60(b)(2) specifically requires that “newly discovered evidence” have been pursued with “due diligence” prior to the original trial. Although I disagree with the majority’s reading of Ferrell and would prefer to simply reject Ferrell’s, overbroad holding, I fully support the holding of the majority that there is no exception to Rule 60(b)(2)’s due diligence requirement.
II.
In its brief Central Cartage Company (“Cartage”) argued vigorously that the Jack Yager affidavit constituted “objective evidence” that the parties ascribed an “idiosyncratic meaning” to the Teamsters National Master Freight Agreement (“NMFA”), the collective bargaining agreement at issue in this ease. The implication of Cartage’s idiosyncratic meaning argument is that while the NMFA may say that employer contributions are required for “each casual employee,” the parties actually intended the phrase “each casual employee” to cover only “union casual employees.” While this argument may be ultimately unconvincing, the idiosyncratic meaning approach has been recently and explicitly affirmed in this circuit. And though it is not yet clear how this general principle of contract interpretation will ultimately be harmonized with the more specific rule relied upon by the majority (precluding the use of extrinsic evidence that is unknown to a pension or welfare fund to create contract ambiguity in pension and welfare benefits cases), supra at 1315, it is an analytic approach worthy of mention and further consideration.
Recently, in AM International, Inc. v. Graphic Management Associates, Inc., 44 F.3d 572 (7th Cir.1995), we confirmed the vitality of the external ambiguity doctrine in contract law. Citing the classic case of Raffles v. Wichelhaus, 2 H. & C. 906, 159 Eng. Rep. 375 (Ex.1864), we affirmed in AM International that just because the meaning of a contract appears clear on its face (as the designation of the ship “Peerless” did), does not necessarily mean that the contract truly is unambiguous (since there may be two ships named “Peerless”). 44 F.3d at 575. We also clarified what sort of evidence can be used to prove external ambiguity — “objective” evidence is acceptable, “subjective” evidence is not. And we defined these terms as follows:
By “objective” evidence we mean evidence of ambiguity that can be supplied by disinterested third parties: evidence that there was more than one ship called Peerless, or that a particular trade uses “cotton” in a nonstandard sense. The ability of one of the contracting parties to “fake” such evidence, and fool a judge or jury, is limited. By “subjective” evidence we mean the testimony of the parties themselves as to what they believe the contract means. Such testimony is invariably self-serving, being made by a party to the lawsuit, and is inherently difficult to verify.
Id. We further affirmed in AM International that “[i]f the parties agree to an idiosyncratic meaning, the court will honor their agreement.” Id. at 576. We have previously expressed this idiosyncratic meaning doctrine as follows: “[Pjarties, like Humpty Dumpty, may use words as they please. If they wish the symbols ‘one Caterpillar D9G tractor’ to mean ‘500 railroad cars full of watermelons,’ that’s fine — provided [the] parties share this weird meaning.” TKO Equipment Co. v. C & G Coal Co., 863 F.2d 541, 545 (7th Cir.1988).
In lieu of evaluating Cartage’s argument against the standards of this general principle of contract law, the majority apparently concludes that a more narrow principle of disallowing extrinsic evidence of ambiguity in pension and welfare benefits cases trumps, choosing not to directly address the doctrinal tension. It is my view that such a fundamental principle of contract law deserves more scrutiny than this, especially when the foundation of the competing principle is somewhat questionable and its history so limited.
The majority’s claim that Central States Pension Fund v. Gerber Truck Service, Inc., 870 F.2d 1148 (7th Cir.1989) (en banc), “precludes the use of extrinsic evidence that was unknown to a pension or welfare fund,” supra at 1315, overstates the holding of that case. The majority does not cite any partic*1318ular portion of Gerber as containing this holding, probably because Gerber never reaches a conclusion of this generality. Judge Easterbrook’s opinion in Gerber astutely notes many important policy reasons for limiting the use of extrinsic evidence by employers defending against pension fund claims, 870 F.2d at 1151-56, but it never actually finds that these policy considerations mandate a per se rule that extrinsic evidence unknown to a pension fund can never be used in pension contribution disputes. The concern in Gerber was with employers attempting to avoid their obligations to pension funds by pointing to defects in the formation of the contract creating their pension plans, since the obligations of pension funds to covered employees are independent of any employer’s actual payments to the funds. Id. at 1153. Gerber never considered the defense of external ambiguity and actually focussed on defenses of another sort: “[d]efenses based on fraud in the inducement, oral side agreement, course of performance, want of consideration, failure of the union to have majority support ... the defenses most likely to breed litigation even when asserted in good faith, and ... create manifold opportunities for manipulation by crafty operators.” Id. at 1154 (emphasis in original). Yet the defense of external ambiguity based on objective evidence of idiosyncratic meaning, as outlined in AM International, simply does not portend the same sort of fact-intensive litigation or opportunity for fraud as the defenses considered in Gerber.
The majority subsequently states that in Gerber and Central States Pension Fund v. Joe McClelland, Inc., 23 F.3d 1256 (7th Cir.1994), “we emphasized [that] ... a multi-employer pension agreement is not a normal two-party contract for which evidence of idiosyncratic meaning may be used to depart from the objective meaning of the words.” Supra at 1315 (emphasis added). Yet neither of those cases even mentioned the idiosyncratic meaning doctrine or the external ambiguity doctrine. As noted above, the concern in Gerber was with defenses based on defects in contract formation and defenses that breed litigation and create opportunities for manipulation — not the external ambiguity defense set out in AM International. Our decision in Joe McClelland likewise fails to support the majority’s current interpretation of it. Although Joe McClelland never mentioned the idiosyncratic meaning doctrine, it did express the following proposition, which appears to be the foundation for the majority’s treatment of multi-employer pension agreements as a special category of contract law: “extrinsic evidence may not be used to create an ambiguity in a pension or welfare agreement subject to ERISA.” 23 F.3d at 1259 (citing Bidlack v. Wheelabrator Corp., 993 F.2d 603 (7th Cir.1993) (en banc) Id. at 607-08 (lead opinion), Id. at 616-19 (dissenting opinion) (these two opinions speak for a majority of the court on this subject) and Johnson v. Georgia-Pacific Corp., 19 F.3d 1184, 1187 (7th Cir.1994)). Not only is this statement dicta and without analysis or explanation, the citations given for it fail to support it. Johnson is not on point, and Bidlack is to the contrary.
The discussion cited in Johnson deals with an attempt by pensioners to avoid the implications of a plan amendment by putting on “evidence” that, at the time of the amendment, the parties did not consider how it would affect other elements of the plan, i.e., by showing a lack of evidence of reflection on the issue. We concluded that “[o]missions from the ‘legislative history’ of a pension or welfare plan may not be used to undermine or vary the terms of that plan.” 19 F.3d at 1187 (citing Bidlack). Yet evidence that the parties didn’t adequately consider the implications of their contract language is entirely different from evidence that, in using particular contract language, the parties intended a nonconventional meaning, which is what the idiosyncratic meaning branch of the external ambiguity doctrine is about.
This leads me to Bidlack, the seeming fountainhead of the majority’s “rule” that extrinsic evidence cannot be used to create contract ambiguity in pension and welfare benefits cases. While the Bidlack dissent certainly points out the dangers of allowing extrinsic evidence in pension and welfare benefits cases, it does not go so far as to say that such evidence should never be allowed to establish external ambiguity in these cases. 993 F.2d at 616-19 (Easterbrook, J., dissent*1319ing) (joined by Bauer, C.J., Coffey, and Man-ion, JJ.). More importantly and more disturbing, however, is the fact that the lead Bidlack opinion (which is cited in Joe McClelland as support for the rule excluding extrinsic evidence of ambiguity in pension and welfare benefits cases) does not in any way support the claimed “rule.” In fact, the lead opinion in Bidlack explicitly adopts the more traditional contract law principle of idiosyncratic meaning:
Although extrinsic evidence is admissible to show that a written contract which looks clear is actually ambiguous, perhaps because the parties were using words in a special sense, ... there must be either contractual language on which to hang the label of ambiguous or some yawning void ... that cries out for an implied term.
Id. at 608 (Posner, J.) (joined by Flaum, Kanne, J.J.) (internal citations omitted). This court’s very divided decision in Bidlack simply cannot bear the weight that Joe McClelland directly places upon it and the majority implicitly places upon it. In fact, Judge Posner’s opinion in Bidlack lays out the very doctrine of idiosyncratic meaning that today’s majority avoids, despite Cartage’s attempted reliance on it.
The majority may believe that the Yager affidavit so utterly fails to meet the AM International standard for “objective evidence” that the external ambiguity doctrine need not be addressed. If so, I agree that the Yager affidavit does not qualify as “objective evidence” for the purpose of establishing external ambiguity under AM International. While the affidavit is technically not testimony of a “party,” since Yager was the chief negotiator for the Teamsters, not Cartage, it hardly represents the kind of “objective” evidence that AM International described. Rather, the proffered testimony poses all the dangers that we worried about with “subjective” evidence: 1) the testimony concerns what parties to a contract believed it meant, rather than direct evidence of an objective fact observable in the world; 2) such evidence is easy to “fake” and inherently difficult to verify; and 3) such testimony could be self-serving.3 See AM International, 44 F.3d at 575.
Further, the majority may be of the view that the external ambiguity doctrine and the special rule for pension and welfare benefits cases do not actually conflict. I see and have explicitly underscored the tension, which was raised by the parties, but perhaps resolution of the issue is best reserved for another day, when the facts of the case raise it more convincingly.4
*1320The majority concludes that “the fact remains that the [Yager] affidavit does not purport to decode ambiguous language; it expresses an understanding inconsistent with the language.” Swpra at 1315. But the external ambiguity doctrine is not about ambiguous language; and, by definition, an idiosyncratic meaning is inconsistent with general word usage. That’s what makes the ambiguity “external,” as opposed to “internal”; and that’s why I find it inappropriate to respond to an argument based on this doctrine by summarily finding that we cannot even consider extrinsic evidence of idiosyncratic word usage.
III.
One other issue appears worthy of flagging. Central States Pension and Welfare Funds (“Central States”), the plaintiffs in this case, initially sought to recover from defendant Central Transport (“Transport”) on either of two theories: 1) guarantor liability or 2) successor liability. The underlying debt was originally incurred by the now defunct General Highway Express (“GHE”). GHE filed for bankruptcy in 1983, and in 1984 the Bankruptcy Court entered an order confirming GHE’s plan of reorganization. Under this plan, Transport agreed to serve as guarantor for GHE’s debt to Central States. During 1987, however, GHE was merged into Transport, which thereby became the successor to GHE’s claims and liabilities. The subsequent 1988 settlement agreement between Central States and Transport, which purported to release all of Transport’s contribution obligations through their last audit (December 1986), thus appears to raise a quandary: has the GHE debt been released or not? The district court held that Transport’s guarantor obligation, which was established before the 1986 cutoff, was released by the settlement, but that its successor obligation, which commenced with the 1987 merger, was not released. Since the merger occurred after the 1986 cutoff date, the district court held, as we do today, that the successor liability was not released. I agree with the majority’s analysis of the successor liability issue. However, since Transport was one company, which wore these alternative mantles of liability, I deem it appropriate to offer an explanation of why, though it has been stripped of the one, it remains clothed with the other. I would adopt the thorough analysis contained in the district court’s opinion, Central States, Southeast and Southwest Areas Pension Fund v. Central Transport, Inc., 861 F.Supp. 1402, 1413-15 (N.D.Ill.1994).
. Rule 60(b) reads as follows:
On motion and upon such terms as are just, the court may relieve a party or a parly’s legal representative from a final judgment, order, or proceeding for the following reasons: ... (2) newly discovered evidence which by due diligence could not have been discovered in time to move for a new trial under Rule 59(b); ... (5) the judgment has been satisfied, released, or discharged, or a prior judgment upon which it is based has been reversed or otherwise vacated, or it is no longer equitable that the judgment should have prospective application.
Fed.R.Civ.P. 60b(2). The only amendment to rule 60 since 1955 occurred in 1987, and it did not affect the language or meaning of 60(b). Fed. R.Civ.P. 60 advisory committee's note.
. See, e.g., Shook & Fletcher Insulation Co. v. Central Rigging & Contracting Corp., 684 F.2d 1383, 1385 n. 2 (11th Cir.1982) (interpreting both Ferrell and Johnson Waste as allowing a judgment to be reopened where newly discovered evidence is "practically conclusive,” even if not obtained by due diligence); United States v. Philatelic Leasing, Ltd., 794 F.2d 781, 788 (2d Cir.1986) (same interpretation of Ferrell); Niedland v. United States, 338 F.2d 254, 260 (3d Cir.1964) (same). District courts have reached the same conclusion. See, e.g., Wilcox v. Petroleum Helicopters, Inc., 145 F.R.D. 406, 408 (S.D.Tex.1993) (finding that Johnson Waste and Ferrell held that grant of new trial could be appropriate “even though proper diligence was not used to secure [the new] evidence for use at the trial”); Ope Shipping, Ltd. v. Underwriters at Lloyds, 100 F.R.D. 428, 432 (S.D.N.Y.1983) (same for Ferrell).
. While Yager may be a third party, he may not be a disinterested third party. In its brief Central States points to a number of reasons to question the objectivity of the Yager affidavit: 1) other Teamster negotiators of the NMFA categorically deny Yager's claim that only union casuals were intended to be covered; 2) such an agreement would have been illegal and would have exposed the Teamsters to breach of duty of fair representation claims; and 3) since 1993 Yager has been barred for life from holding any position with the Teamsters.
. An example of such a case could be the following. Within a business whose work is very seasonal (lawn care, snow removal, private forest-fire-fighting, etc.) an employer and a union could agree that pension contributions would be based on employee hours worked during a given pay period. While this approach could result in greatly varying pension contributions over the course of a year (since hours worked would vary so greatly), the employer and the union representatives could reasonably agree that over a year's time the total amount paid in would be sufficient to cover the cost of the pension plans. Even if a union in this situation insisted that employees be paid on a salary basis that was consistent throughout the year, it could reasonably agree to let the employer calculate its pension contributions according to this variable rate — thus allowing the employer to decrease its contributions when work and the resulting income were low, and make up the difference when work and income were high. With this understanding, a union and an employer might formulate a collective bargaining agreement that established employer pension contributions at "$20 per day worked by each employee.” Both parties would intend the phrase "per day” to refer not to calendar days, but to 8 hour periods of work. For example, an employee working 4 hours per day for 5 days (20 hours) would have worked only 2.5 days for the purpose of calculating the pension contribution. If the pension fund administering the program later sued the employer for inadequate contribution, insisting that the phrase “per day” is unambiguous and obviously refers to calendar days, could the employer defend against this suit by invoking the idiosyncratic meaning doctrine and putting forward objective evidence (such as other documents reflecting this atypical but agreed-upon use of the word "day”) that the phrase "per day” actually referred to *1320eight hour periods of time? That is the kind of situation that would force us to resolve the doctrinal tension noted here.