Dissenting:
I would affirm the decision of the district court and therefore respectfully dissent from the conclusion reached by the majority. The facts and background of this appeal have been fully and clearly stated. My disagreement with the conclusion reached is based upon my reading of Judge Knapp’s decisions of May 1, 1989 and August 31,1993.
*125In my view, Judge Knapp’s decision of May 1, 1989 reflects a studied analysis of the relevant statistics upon which the EEOC relied and which the defendant failed to challenge. That opinion, in my view, also reflects a consideration of the causal connection between the offending employment practice and its resulting disparate impact. In that opinion Judge Knapp wrote:
The EEOC has conducted exhaustive discovery and analysis of JAC’s applicant records for the years 1980-rl984, covering three successive apprenticeship programs ... The EEOC has presented statistical data, gleaned from these records as well as from outside sources such as census data, revealing disparity between blacks and whites and between women and men in several categories.
Joint Appendix (“JA”), Vol. I at 346. Further on in that opinion, Judge Knapp found that “[d]efendants have failed to cast doubt on the validity of EEOC proffered statistics.” Id. at 348. His decisions based upon those findings are related in the majority opinion.
On appeal, this Court vacated Judge Knapp’s May 1, 1989 decision concluding that a remand was warranted for the reason that his decision “appears not to conform to the applicable legal standard” (emphasis mine) announced in Wards Cove Packing Co. v. Atonio, 490 U.S. 642, 109 S.Ct. 2115, 104 L.Ed.2d 733 (1989). See EEOC v. Joint Apprenticeship Committee, 895 F.2d 86, 91 (2d Cir.1990). That standard, simply put, requires the plaintiff, in establishing a prima facie case, to show that there is a causal connection between the employment practice and the disparate impact. See, 490 U.S. at 656, 109 S.Ct. 2115. I would suggest that the requisite causal connection appeared in that decision and its appearance inadvertently overlooked. Thus, in discussing the high school diploma requirement, Judge Knapp wrote: “[T]he EEOC has presented statistics showing a significant disparity between the percentages of blacks with high school diplomas and whites with high school diplomas in the geographical area from which JAC draws its applicants .... ” JA, Vol. I at 346. After citing the relevant statistics, Judge Knapp continued:
This disparity may explain the under-representation of blacks in the applicant pool during the year's in question: blacks made up 12.2% of the applicants to the three programs, but comprise 18.3% of the available pool of “operatives” and “laborers” in the relevant geographical area, based on 1980 Census data. Yet it is not even necessary to speculate on the degree to which the high school requirement deters blacks without high [school] diplomas from applying, because the data concerning actual applicants to the three programs show that blacks without high school diplomas apply, and are turned away, at a higher rate than whites without high school diplomas. For example, with respect to Program 12, 6.2% of the black applicants lacked high school diplomas, as compared with 2.13% of white applicants.... According to the EEOC’s statistical analysis, all of the above-listed disparities are statistically significant well beyond the two to three standard deviations suggested by the Supreme Court as determinative of whether a legally significant disparate impact has been shown. Castaneda v. Partida, 430 U.S. 482 n. 17, 97 S.Ct. 1272, 51 L.Ed.2d 498 (1977).
JA Vol. I, at 347 (emphasis added).
That Judge Knapp was aware of and found the requisite causal connection is perhaps most clearly manifest from this portion of his opinion:
Second, [defendants] argue that an insufficient “nexus” has been shown between the age maximum and the disparity between the percentages of male and female applicants, suggesting that other factors, such as gender stereotyping in education, may be responsible. The main flaw in this argument is that it *126ignores the import of the bare numbers: there were consistently higher percentages of overage female applicants than male applicants, and thus the age maximum had a disparate impact on female applicants. It is not necessary for plaintiffs to explain the disparity on which their prima facie case rests; they must only show that its existence is more probable than not. Eldredge v. Carpenters 46 Northern California Counties Joint Apprenticeship and Training Committee, 833 F.2d 1334, 1340 n. 8 (9th Cir.1987) cert, denied 487 U.S. 1210, 108 S.Ct. 2857, 101 L.Ed.2d 894 (1988) (“For purposes of [plaintiffs] prima facie case, ... the issue is not why the [requirement] has a discriminatory impact, rather the issue is simply whether it had such an impact.”)
JA Vol. I at 350-51 (emphasis added).
The “appearance” of a causal connection between the offending employment practice and the disparate impact to which it gave rise is, I suggest, readily found and not made invisible by the absence of words such as “causal connection” an insistence upon which would be to exalt form over substance and to require a mantra-like recitation of a formula to express a finding otherwise made.
In his second opinion dated August 31, 1993, following the remand, Judge Knapp wrote:
Turning to the key question on remand, whether the EEOC has demonstrated a causal nexus between the challenged practices and the statistical disparities, we are persuaded by a careful consideration of the record that the educational
requirement and age maximum had a chilling effect on potential black and women applicants, which caused the above-described statistical disparities. Thus we conclude that the EEOC has carried its burden and made out a prima facie case of disparate impact with respect to both requirements. 828 F.Supp. at 266 (S.D.N.Y.1993)
Two conclusions follow inevitably from the foregoing. First, Judge Knapp initially found the requisite causation because its absence was specifically argued by the defendant and rejected by him. Second, the defendant deliberately calculated not to pursue a business necessity justification but elected instead, to disparage, albeit unsuccessfully, the statistical data upon which the EEOC relied. Their awakening to the realization that they blundered strategically1 and then belatedly (after the record was closed and oral argument concluded) attempted to correct it by submitting affidavits, which Judge Knapp quite properly foreclosed,2 does not entitle them to revisit that issue nor did this Court invite the issue to be revisited on remand. In remanding to the district court, this Court wrote: “In doing so, we express no view as to the sufficiency of EEOC’s statistics to establish the requisite disparities, or as to whether summary judgment is appropriate with respect to the issue of causation.... If it is determined that EEOC has advanced a prima facie case of disparate impact, the burden of production shifts to JAC to produce evidence rebutting the prima facie case. Thus, JAC would then bear the burden of showing that there are ‘legitimate nondiscriminato*127ry reasons’ for the specific employment practice in question.” 895 F.2d at 91.
The direction to remand does not require the district court to revisit the issue of business necessity. Rather, it is a direction to remand solely for the purpose of determining whether the plaintiff has indeed set forth a prima facie case, given that Wards Case requires a showing of a causal connection between the statistics presented and the practices being challenged. Whether a plaintiff can establish a prima facie case of disparate impact and whether a defendant can show business necessity are two separate inquiries. The fact that plaintiffs prima facie case was revisited and reaffirmed on remand should in no way obligate the district court to again address the issue of business necessity which the defendant elected to forego. To do so would simply permit the defendants two bites at the apple. This litigation, commenced fourteen years ago, should not be further prolonged.
Judge Knapp’s musings in a footnote, 828 F.Supp. at 267 n.3, gives the majority some pause, namely “even if we were to allow affidavits, neither refers to any study or scholarly work (or indeed ivork of any kind) to support its conclusions .... ” (emphasis added). First, it is dicta and second, the phrase “or indeed work of any kind” I construe as a muted echo of Matsushita Electric Industrial Co. v. Zenith Radio Corp. 475 U.S. 574, 586-87, 106 S.Ct. 1348, 89 L.Ed.2d 538 (1986) where the Court wrote that in opposing a motion for summary judgment the “opponent must do more than simply show that there is some metaphysical doubt as to the material facts.... In the language of the Rule, the nonmoving party must come forward with ‘specific facts showing that there is a genuine issue for trial.’ ” Judge Knapp considered “the record taken as a whole” and found in essence that it “could not lead a rational trier of fact to find for the nonmoving party, there is no genuine issue for trial.” Id. at 587, 106 S.Ct. 1348. Coming forward with no “work of any kind” is not even to raise a metaphysical doubt.
As regards the issue of back pay, I would also affirm the judgment of the district court.
. As Judge Knapp expressed it in his order of May 1, 1989 at JA 352:[A]l oral argument [defendants] admitted that a strategic decision had been made to proceed in this manner, placing all of their chips on their (now failed) effort to undercut the EEOC's prima facie case. See also 828 F.Supp. 264, 267 n. 3 (S.D.N.Y.1993)
. Although in theory, a motion may be amended at any time before the Judge has acted upon it, "it is particularly inappropriate after briefs have been interposed by the opposing parties and oral argument heard ...” Wright & Miller, Federal Practice and Procedure: Civil 2d § 1194, cited with approval in Select Creations, Inc. v. Paliafito America, Inc., 828 F.Supp. 1301, 1357 (E.D.Wis.1992). It is important to note that this Court made no finding that Judge Knapp abused his discretion in foreclosing the submission of affidavits after briefs were submitted and oral argument heard.