841 F.2d 1067
46 Fair Empl.Prac.Cas. 1542,
46 Empl. Prac. Dec. P 37,898
Charles R. BARBER, Plaintiff-Appellee,
v.
INTERNATIONAL BROTHERHOOD OF BOILERMAKERS, IRON SHIP
BUILDERS, BLACKSMITHS, FORGERS, AND HELPERS,
DISTRICT LODGE # 57, Defendants-Appellants.
No. 87-7080.
United States Court of Appeals,
Eleventh Circuit.
April 4, 1988.
George C. Longshore, Birmingham, Ala., for defendants-appellants.
Carol Ann Rasmussen, Birmingham, Ala., for plaintiff-appellee.
Appeal from the United States District Court for the Northern District of Alabama.
Before VANCE and CLARK, Circuit Judges, and GARZA*, Senior Circuit Judge.
CLARK, Circuit Judge:
The district court in this case has entered two judgments against appellant, District Lodge No. 57 of the International Brotherhood of Boilermakers ("the union"), finding on both occasions that the union intentionally discriminated against appellee Charles R. Barber in violation of Title VII of the Civil Rights Act of 1964, 42 U.S.C. Secs. 2000e to 2000e-17 (1982). After the entry of the first judgment, the union appealed to this court, the judgment was vacated, and the case was remanded to the district court for proceedings consistent with our opinion. See Barber v. International Brotherhood of Boilermakers, 778 F.2d 750 (11th Cir.1985). On this appeal, finding that the district court exceeded our mandate by admitting a considerable amount of new evidence and that the evidence that was authorized on remand did not support a finding of intentional discrimination, we reverse.
I. BACKGROUND
The International Brotherhood of Boilermakers is a construction union that negotiates with employers concerning the working rules and conditions of employment in the boilermaking trade. The collective bargaining agreement governing the boilermakers in District Lodge No. 57 is the Southeastern States Articles of Agreement. Under the Southeastern States Agreement, locals of the union maintain an "out-of-work" list from which union members and others are referred to various jobs.
The rates union members are paid depend primarily on their status in the union. "Full-fledged," or journeymen, boilermakers are paid 100% of the rate specified in the Southeastern States Agreement. Those who are not fully qualified boilermakers--those in the union's helper-trainee program--are usually paid less. The trainee program, as it is set forth in the Agreement, provides that union members with less than 2000 hours experience are paid 70% of the specified rate and those with 2000-8000 hours 80% of the specified rate. Once a trainee reaches 8000 qualified hours, he or she becomes a full-fledged boilermaker and is always paid the 100% rate.1
Appellee Barber first became affiliated with the union in July 1977. He had been working as a laborer and cement finisher with several members of the union. These members approached Carl Phillips, the business agent of Local 455, about starting Barber in the trainee program. After talking with Barber, and on the belief that he had no boilermaking experience, Phillips signed Barber up as a 70% trainee and the only black in the local.
In the years after 1977, Barber attended a welding school in Ohio and gained considerable experience in welding and rigging. In early 1978, he began receiving referrals at the 80% rate. In 1980, he joined the International Union's National Transient Division (NTD). For this he received a book indicating that he was a full-fledged boilermaker under the terms of NTD's separate collective bargaining agreement. In 1982, Barber asked the union to credit him for his rigging and welding experience as well as hours he had spent boilermaking in 1973. He also asserted that his NTD book entitled him to the 100% rate. The union responded that his NTD book did not alter his trainee status with Local 455 and that it would need documentary proof of his experience in 1973. It is unclear from the record whether Barber was granted hours of credit for his rigging and welding experience. In any event, at the time of the first trial, he was still consistently being referred out at the 80% rate.
On November 2, 1981, Barber filed a complaint with the EEOC charging that whites no more qualified than he were receiving referrals at the 100% rate. The complaint was not Barber's first. He had brought an EEOC claim in 1978, and that claim led to a lawsuit in 1979. On June 30, 1980, Barber and the union settled, "in full compromise settlement and satisfaction of any and all claims and causes of action raised or for which could have been raised." Record, Defendant's Exh. 12 at 3.
On the first trial of this case, the district court found in Barber's favor. The court acknowledged that the June 1980 settlement agreement barred any consideration of discrimination that occurred prior to that date, but found that after June 30, 1980, Barber had been referred at the 80% rate while similarly situated whites had been referred at 100%. The district court's findings were dictated into the record the day after the trial, and they mentioned three white men specifically: John Sharit, Damon Temple, and James Ray. See Supp. Record, Vol. I at 157-58.
On the union's appeal, this court vacated the district court's judgment and remanded the case "for further proceedings consistent with this opinion." Barber, 778 F.2d at 762. In reaching that result, this court closely examined the evidence concerning each of the three white men addressed by the district court. With respect to John Sharit, the court held that he had become a full-fledged boilermaker prior to June 30, 1980, and thus was not situated similarly to Barber during the only period the June 30 settlement permitted the district court to consider. See id. at 756-57. With respect to Damon Temple, the court found that the district court clearly erred when it relied on "internally inconsistent" work records to conclude that Temple, a trainee like Barber, was referred out at 100% after the date of the settlement. The court stated that "[o]n remand, further proceedings will be necessary to establish whether Temple was referred at the 100% rate after June 30, 1980." Id. at 762. Finally, with respect to James Ray, the court declined to upset the district court's finding that Ray's 100% referrals constituted disparate treatment. The court noted, however, that those referrals were not discriminatory if, unlike Barber, Ray had been admitted initially as a full-fledged boilermaker, and it remanded for clarification of that issue. See id. at 759-60.
On remand, the district court reopened discovery, admitted evidence, over the union's objection, concerning tens of additional white union members, and again entered judgment for Barber. See Barber v. International Brotherhood of Boilermakers, 651 F.Supp. 265, 268 (N.D.Ala.1986). In this decision, the court altered its earlier findings with respect to John Sharit and James Ray and found additionally that John McGhee, Williams Burroughs, Ted Gerrard, William Clifford James, and Carl Arrington were examples of similarly situated white union members who received 100% referrals while Barber, because of his race, received 80% referrals. The district court implied that it was not limited to considering only those whites mentioned in this court's order because the district court "never indicated that they were the only similarly situated white employees." Id. at 266 n. 3.
The union argues that in admitting, considering, and basing its finding of discrimination on evidence not put forth in the first trial, the district court exceeded our mandate and violated the law of the case. The union also argues that the evidence at the second trial that was within the mandate's scope failed to establish the union's intent to discriminate. We agree on both points, and reverse for entry of a judgment in favor of the union.
II. THE EFFECT OF THE PRIOR APPEAL
This case is governed by two related doctrines. The first is what commentators have denominated the "mandate rule." See, e.g., J. Moore, J. Lucas & T. Currier, Moore's Federal Practice p 0.404 (2d ed. 1983); see also Litman v. Massachusetts Mutual Life Insurance Co., 825 F.2d 1506, 1511 (11th Cir.1987) (in banc). The mandate rule simply embodies the proposition that "a district court is not free to deviate from the appellate court's mandate." Wheeler v. City of Pleasant Grove, 746 F.2d 1437, 1440 n. 2 (11th Cir.1984); see Baumer v. United States, 685 F.2d 1318, 1321 (11th Cir.1982). The lower court may consider anew issues not " 'within [the mandate's] compass.' " Quern v. Jordan, 440 U.S. 332, 348 n. 18, 99 S.Ct. 1139, 1148 n. 18, 59 L.Ed.2d 358 (1979) (appropriateness of notice relief could be considered where Court had addressed only constitutionality of notice relief) (quoting Sprague v. Ticonic National Bank, 307 U.S. 161, 168, 59 S.Ct. 777, 781, 83 L.Ed. 1184 (1939)); see Perkins v. Standard Oil of California, 399 U.S. 222, 224, 90 S.Ct. 1989, 1990, 26 L.Ed.2d 534 (1970) (lower courts could consider award of attorneys' fees where Court had addressed only merits of Clayton Act claim). However, where an appellate court remands for "resolution of a narrow factual issue," the lower court may not circumvent the mandate by approaching the identical legal issue under an entirely new theory. Baumer, 685 F.2d at 1321 (district court could not take evidence of fair market value of option when granted where appellate court remanded for determination of option's fair market value when exercised); see EEOC v. International Longshoremen's Ass'n, 623 F.2d 1054, 1058 (5th Cir.1980) (district court could not hold hearing on the equity of merging one of four locals of a union where appellate court directed that all four locals be merged), cert. denied, 451 U.S. 917, 101 S.Ct. 1997, 68 L.Ed.2d 310 (1981). As should be apparent, the application of these mandate rule principles will, in accord with the rule's purpose of promoting finality, depend considerably on the stage a case has reached when it goes up on appeal and on the language of the appellate court's mandate and/or opinion.
It seems rather clear to us that the district court violated the mandate rule when it reopened the case to accept the records of and testimony concerning a number of white union members not mentioned in this court's prior opinion and ostensibly not a part of the first trial. The first opinion of this court directed the district court to resolve four narrow factual issues: (1) whether Damon Temple received any 100% referrals after June 30, 1980; (2) whether James Ray was admitted to the union as a trainee or a full-fledged boilermaker; (3) if so, whether Ray was given credit for experience that would justify the 100% referrals; and (4) if so, whether there was any evidence that might suggest that the union's claim of reliance on Ray's experience was a pretext for discrimination. The case had been fully tried below, and this court's opinion clearly specified the only remaining areas of inquiry. See Litman, 825 F.2d at 1513. Although we recognize that the district court remarked on remand that it had "never indicated that [Sharit, Ray, and Temple] were the only similarly situated white employees," Barber, 651 F.Supp. at 266 n. 3, we note that even on remand, the only white employees the district court referred to that were not mentioned in its first order were men whose records were not in evidence at the first trial. In short, even if it had been appropriate for us to sift through the record on the first appeal trying to find the other, unnamed, similarly situated white employees, the district court's own order on remand convinces us that we would not have been able to do so.2
Barber argues that the admission of the additional records did not violate the mandate rule because the union was permitted on remand to amend certain aspects of the pretrial order. Specifically, the union was permitted to indicate that a provision of the Southeastern States Agreement that trainees could receive the 100% rate while doing ASME Code welding was no longer in force when Barber was doing such welding. See supra note 1. We reject Barber's argument that this amendment reopened the case because permitting the amendment was itself a violation of the mandate. The subject matter of the amendment was addressed in this court's first opinion, see Barber, 778 F.2d at 753 n. 2, the court evidently felt there was not a need for any further fact-finding,3 and the district court was not directed to resolve the ambiguity. We conclude that the district court violated the mandate rule by taking evidence as to Carl Arrington, William Burroughs, William James, Ted Gerrard, and John McGhee, and that to the extent its finding of intentional discrimination was based on the referrals of these men, that finding was clearly erroneous.
We further conclude that the district court violated the mandate rule by considering "new" evidence concerning John Sharit. This court determined on the first appeal that Sharit became a full-fledged boilermaker in February or March 1980 and thus could not be considered situated similarly to Barber during the relevant time period, i.e., post-June 30, 1980. We cannot help but note that we did so because Sharit himself, a witness for the plaintiff, testified to the February-March date, and in its first ruling, the district court expressly credited Sharit's testimony.4 Supp.Record, Vol. I at 65, 157. Under these circumstances, and the fact, as described above, that we specifically directed the district court to resolve only the four factual issues concerning Damon Temple and James Ray, any consideration of Sharit on remand was foreclosed. Thus, insofar as the district court relied on referrals of Sharit, it exceeded the mandate and its finding of intentional discrimination was clearly erroneous.5
The second doctrine implicated here is that of the "law of the case." Under this doctrine, "both the district court and the court of appeals generally are bound by findings of fact and conclusions of law made by the court of appeals in a prior appeal of the same case." United States v. Robinson, 690 F.2d 869, 872 (11th Cir.1982); see Dorsey v. Continental Casualty Co., 730 F.2d 675, 678 (11th Cir.1984). The courts are bound not only by explicit findings, but by those matters decided by necessary implication. Carpa v. Ward Foods, Inc., 567 F.2d 1316, 1319 (5th Cir.1978). The law of the case does not bar reconsideration of factual findings or legal conclusions when (1) a subsequent trial produces substantially different evidence, (2) controlling authority has since made a contrary decision of law applicable to that issue, or (3) the prior decision was clearly erroneous and would work manifest injustice. Wheeler, 746 F.2d at 1440; Robinson, 690 F.2d at 872.
In its conclusion regarding James Ray--that he had been admitted as a full-fledged boilermaker but that allowing him such status was itself discriminatory--the district court violated the law of the case. This court determined on the prior appeal that Barber might have a valid claim of disparate treatment if he could establish that James Ray was being referred at 100% while still a trainee like Barber. Because the record was clear that Ray had been referred at 100%, but was not clear on Ray's status, we remanded the case for clarification of the latter point. In the course of our discussion, we stated flatly that if Ray had been admitted to the union as a full-fledged boilermaker, Barber's claim necessarily would fail: "If the court finds on remand that Ray had been considered a full-fledged boilermaker all along, ... then it must conclude that his referrals in this status do not support Barber's prima facie case." Barber, 778 F.2d at 762. Implicit in this statement was our conclusion that Ray was admitted to the union in December 1979, and thus, even if it was discriminatory to give him full-fledged status, the discrimination occurred during the period covered and rendered non-actionable by Barber's settlement with the union. No exceptions to the law of the case doctrine permitted the district court to contradict this conclusion, see Westbrook v. Zant, 743 F.2d 764, 768 (11th Cir.1984), and its finding of intentional discrimination based on referrals of Ray was thus clearly erroneous.
III. CONCLUSION
Our dispositions of the district court's attempts to compare Barber with Carl Arrington, William Burroughs, Ted Gerrard, William James, John McGhee, John Sharit, and James Ray6 make it clear that the court's overall finding of disparate treatment was clearly erroneous. Accordingly, the judgment of the district court is REVERSED, and judgment should be entered for District Lodge No. 57.
REVERSED.
Honorable Reynaldo G. Garza, Senior U.S. Circuit Judge for the Fifth Circuit, sitting by designation
It has not always been necessary to advance hour by hour through the training program to be paid the 100% rate. According to testimony at the first trial, certain people thought to have "sufficient experience along our line" were deemed 100% "probationaries" (apparently the term used prior to the establishment of the helper-trainee program), or at least were given credit toward their 8000 hours. Supp.Record, Vol. I at 91, 105 (testimony of local business agent). For a time, workers who were proficient welders could receive 100% while welding up to the standards of the American Society of Mechanical Engineers code. In certain periods, when workers were especially needed, the 8000-hour requirement was reduced to 6000 hours. Finally, in November 1979, with the revision of the local's referral rules, persons who had consistently, even if mistakenly, received 100% pay were "grandfathered," i.e., deemed entitled to that rate from then on
We are comforted in this conclusion by the fact that Barber's counsel requested neither rehearing by the panel nor rehearing by the court en banc. The first opinion of this court should have made clear to counsel that the case had come down to a comparison with three white employees, and yet no objection was filed. See Litman, 825 F.2d at 1513 (noting appellee's failure to seek any modification of appellate court's prior decision, which limited issues on remand). We can only presume that this was because counsel was aware that Sharit, Temple, and Ray constituted Barber's best case
Implicit in the court's discussion of the ambiguity regarding the actual dates between which trainees could receive 100% for Code welding was a finding that neither side could rely on the temporary advancement: Barber could not claim that he was qualified for the rate because he was Code welding, and the union could not justify a differential between Barber and a white employee by asserting that the white was welding
It is possible to read the district court's first order as crediting only Sharit's testimony that he was always sent out at the 100% rate. However, because we can discern no reason why the court might have credited only part of the same witness' testimony--it would seem that one's rate of pay and the date one finally accumulated 8000 hours would be equally memorable--we assume that it credited Sharit's testimony in full
We are mindful that our decision in this regard may seem harsh. Still, it is well settled that plaintiffs in all cases are to be given their day in court, nothing less but nothing more. Barber was given such an opportunity, as the case was fully tried the first time, and the "new" records reflecting Sharit's status were available then. See Baumer, 685 F.2d at 1321 ("There is nothing in the record to indicate that the evidence produced at the hearing after remand was unavailable to the taxpayers during the first trial.")
We are also aware that the law of the case doctrine, discussed infra, includes two exceptions that Barber would argue cover the new evidence concerning Sharit: prior findings by a court of appeals are not binding if (1) evidence in a subsequent trial produces substantially different evidence, or (2) the appellate decision was clearly erroneous and would work a manifest injustice. United States v. Robinson, 690 F.2d 869, 872 (11th Cir.1982). The fact remains, however, that there should have been no opportunity for substantially different evidence to appear, as Sharit's referrals were not to be considered on remand. The law of the case exceptions apply only when substantially different evidence comes out in the course of a subsequent trial authorized by the mandate. Moreover, even if we deemed it appropriate to consider the new evidence under the exceptions to the law of the case doctrine, we do not believe that it was substantially different or rendered our earlier holding clearly erroneous. Although we cannot be certain precisely what evidence the district court was relying upon, it appears to be another of the "internally inconsistent" work records we disapproved on the first appeal. See Barber, 778 F.2d at 762. Additionally, a finding cannot be considered clearly erroneous simply because it relies on a witness' testimony--testimony expressly credited by the district court--over documents that are highly likely to be inaccurate.
The district court was free on remand to inquire further into the referrals of Damon Temple. See Barber, 778 F.2d at 762. Because the court did not include any discussion of Temple in its second order, we assume that the evidence showed that Temple did not receive any 100% referrals after June 30, 1980 and thus did not support Barber's prima facie case