(dissenting).
Washington v. Davis, 426 U.S. 229, 239, 242, 96 S.Ct. 2040, 2047, 2049, 48 L.Ed.2d 597 (1977) holds:
[0]ur cases have not embraced the proposition that a law or other official act, without regard to whether it reflects a racially discriminatory purpose, is unconstitutional solely because it has a racially disproportionate impact. [Emphasis in original.]
# sf( # * # $
. [W]e have not held that a law, neutral on its face and serving ends otherwise within the power of government to pursue, is invalid under the Equal Protection Clause simply because it may affect a greater proportion of one race than of another.
The majority today determines that Washington v. Davis, supra, supports their previous holding that the Massachusetts Veterans’ Preference statute, Mass.Gen. Laws ch. 31, § 23, deprives women of equal protection of the laws in violation of the Fourteenth Amendment in all areas of civil service employment in the Commonwealth. Although recognizing that “[a] facially neutral statute may not be deemed vulnerable to equal protection challenge solely because it has a disproportionate impact”, ante at 146, Judge Tauro reaches this determination by finding that “[w]e are dealing here with a statute that is not facially neutral”, ante at 147, fn. 7, and that it is the Commonwealth’s intent to achieve the purpose of benefiting its veterans “by subordinating employment opportunities of its women”. Ante at 149-150. Judge Campbell concurs in the judgment of unconstitutionality, finding that the inevitability and degree of disproportionate effect make the statute non-neutral and that the inevitability of effect suggests discriminatory intent. With respect, I disagree that these findings and the result reached are demonstrably tenable.
I
The Veterans’ Preference statute is not on its face gender-based. Anthony v. Commonwealth of Massachusetts, 415 F.Supp. 485, 501 (1976) (Campbell, C. J., concurring). Clearly the statutory “division between veterans and non-veterans is not drawn along sex lines and does not provide for dissimilar treatment for similarly situated men and women. On its face the statute is neutral Id. at 503 (Murray, J., dissenting). Most persons favored by the statutory preference are males, although a substantial number of those not so favored are also males. Non-veteran women in larger numbers share with non-veteran men the disfavor of the statute, but a number of those aided by the statute indeed are women. The statute explicitly includes women in its requirement for service during time of war, but not combat duty. Mass.Gen. Laws ch. 4, § 7, cl. 43; ch. 31, § 21; 1958 Op.Atty.Gen., 25-26. Although in operation it favors males in greater proportion than females for the higher civil service *153positions,1 the statutory classification has not been shown to be a mere pretext to accomplish the purpose of invidiously discriminating against women. See Geduldig v. Aiello, 417 U.S. 484, 94 S.Ct. 2485, 41 L.Ed.2d 256 (1974); General Electric Co. v. Gilbert, 429 U.S. 125, 97 S.Ct. 401, 50 L.Ed.2d 343 (1976). Moreover, it is not disputed that the statutory preference was not enacted for the purpose of disqualifying women from receiving civil service appointments. Anthony v. Commonwealth of Massachusetts, supra at 495.
The attempted distinction between the test in Davis and the statute here is totally unconvincing: one is no more neutral than the other. In each case the classification is facially neutral, and in operation the effects are uneven; the only difference is that the statute here has a weightier impact on the relevant group, and impact alone is not determinative, Washington v. Davis, supra, 426 U.S. at 239, 96 S.Ct. 2040.2
II
In Arlington Heights v. Metropolitan Housing Corp., 429 U.S. 252, 264-266, 97 S.Ct. 555, 563, 50 L.Ed.2d 450 (1977), the Court said:
Our decision last Term in Washington v. Davis, 426 U.S. 229 [96 S.Ct. 2040, 48 L.Ed.2d 597] (1976), made it clear that official action will not be held unconstitutional solely because it results in a racially disproportionate impact. “Disproportionate impact is not irrelevant, but it is not the sole touchstone of an invidious racial discrimination.” Id., at 242, 96 S.Ct. at 2049. Proof of racially discriminatory intent or purpose is required to show a violation of the Equal Protection Clause. . . . [Emphasis supplied.]
[I]t is because legislators are properly concerned with balancing numerous competing considerations that courts refrain from reviewing the merits of their decisions, absent a showing of arbitrariness or irrationality. But racial discrimination is not just another competing consideration. When there is a proof that a discriminatory purpose has been a motivating factor in the decision, this judicial deference is no longer justified. [Emphasis supplied.]
The record before the court, to the extent that it provides direct and circumstantial evidence of intent, does not show the operation of the statute and its effect to be a clear pattern, unexplainable on grounds other than an intent to limit the employment opportunities of women. This is so, whether the relevant facts are viewed totally or separately. Conceding the factor of unequal impact and that it was foreseeable, a showing of unconstitutional action has not been made. Even in Davis the government officials there might well have foreseen that blacks would not do as well on the test as whites. See Boston Chapter, N.A.A.C.P. v. Beecher, 504 F.2d 1017, 1021 (1st Cir. 1974). Awareness on the part of the legislature that disproportionate impact would follow is not enough.3 Awareness, like *154foreseeability, is not proof of discriminatory intent, and other evidence is required. The legislative history of the statute with its unequal impact on women is clearly explainable as having the purpose of preferring qualified veterans for consideration for civil service jobs.4
The preference statute is not vulnerable to the claim that discriminatory intent may be inferred because there is no relationship between the preference and job performance. In the first place, the contention of no such relationship is open to dispute, see Feinerman v. Jones, 356 F.Supp. 252, 260 (M.D.Pa.1973), but even if that contention were to prevail, it would bear on intent only if job performance were the only goal the legislature could serve by means of the preference. That is obviously not the case here, for it is in the national interest that enlistment in the armed services be encouraged, see, e. g., H.Rpt. No. 93-857, 93rd Cong., 2d Sess. (1974) (Armed Forces Enlisted Personnel-Bonus Revision Act of 1974), and hiring preferences are well-established means for furthering that purpose. See, e. g., Anthony v. Commonwealth, supra at 496, 497; 42 U.S.C. § 2000e-11.
The statistical evidence presented by plaintiff provides no support for an inference of a discriminatory purpose. This is an impact argument, and Arlington Heights (and Davis) requires proof of intent as “a motivating factor”. Plaintiffs systematic exclusion argument analogizes the jury-selection cases, but those cases do not apply in the context of this case. Arlington Heights pointed out that “[bjecause of the nature of the jury selection task, however, we have permitted a finding of constitutional violation even when the statistical pattern does not approach the extremes of Yick Wo [v. Hopkins, 118 U.S. 356, 6 S.Ct. 1064, 30 L.Ed. 220 (1886)] or Gomillion [v. Lightfoot, 364 U.S. 339, 81 S.Ct. 125, 5 L.Ed.2d 110 (1960)] . . . ”. 429 U.S. at 266, n.13, 97 S.Ct. at 564.5 Whatever the exact focus of the Court in jury-selection cases, the Court makes it clear that even in those cases impact alone is determinative only when it emerges as “a clear pattern, unexplainable on grounds other than race”, Arlington Heights, supra at 266, 97 S.Ct. at 564. The facts here do not fit into that mold: it is undisputed that the preference here is based on a determination to help veteran men and women and not non-veterans.
Plaintiff’s reliance on Castaneda v. Partida, 430 U.S. 482, 97 S.Ct. 1272, 51 L.Ed.2d 498 (1977), is misplaced; Castaneda, which Judge Tauro finds no need to address, ante at 148, n.ll, is distinguishable from the case before us. In that case statistics were used to show that the number of Mexican-Americans on certain grand juries normally to be expected, had the jurors been chosen randomly, was so much higher than the actual number of Mexican-Amerieans called that plaintiff had made out a prima facie case of equal protection violation. The sta*155tistics were presented in the context of the operation of the “key man” system of jury selection, which allows jury commissioners to select jurors from a list on which Spanish surnames are easily identifiable, and the system is thus “susceptible of abuse”. 430 U.S. at 497, 484-85, 495, 97 S.Ct. 1272. No evidence was presented by the State, and the Court recognized that there would be no constitutional violation were the State to explain the numerical discrepancy on neutral grounds. As pointed out above, the preference statute is clearly explainable as having the purpose of preferring veteran men and women at the expense of non-veteran men and women.
Ill
The principle applied in tort and criminal actions, that an actor is presumed to intend the natural and foreseeable consequences of his deeds, must yield to the entirely different considerations at work when a federal court is addressing an equal protection challenge to state legislation. Principles of federalism involve a “recognition of the value of state experimentation with a variety of means for solving social and economic problems”, Anthony, supra at 502 (Murray, J., dissenting), and considerations of federalism require that an impermissible motive in enacting state legislation be not lightly inferred. See Note, Developments in the Law: Equal Protection, 82 Harv.L.Rev. 1065, 1093-94, n.101; A. Bickel, The Least Dangerous Branch, 214; P. Brest, Palmer v. Thompson : An Approach to the Problem of Unconstitutional Legislative Motivation, 1971 Sup.Ct.Rev. 95, 129-30. Inevitability of effect, even coupled with disproportionate impact, “absent a pattern as stark as that in Gomillion or Yick Wo” is not evidence of discriminatory purpose or intent.6 See Davis, supra 426 U.S. at 242, 96 S.Ct. 2040; Arlington Heights, supra 429 U.S. at 266, 97 S.Ct. 555. A legislature’s choice of preferring veterans implies invidious intent only if it appears inconsistent with expected and valid considerations.7 In most hiring situations the scores of those certified would likely be very little different were the veterans’ preference not in effect.8 There is here no indication that the legislature de*156parted from usual considerations in enacting the preference. To the extent, however, that the legislature wishes to use civil service hiring practices to favor veterans, any effort to diminish the impact on women by diluting the preference necessarily results in a diminution of the benefit to veterans. Because of this nature of the hiring benefit, use of the “absolute” preference instead of a point preference, like the use of any preference at all, provides no ground for indictment of the legislature’s motive.
IV
Since Washington v. Davis, three veterans’ preference provisions have been subjected to equal protection challenge; all three have been upheld. Bannerman v. Dept. of Youth Authority, 436 F.Supp. 1273 (N.D.Cal.1977); Branch v. DuBois, 418 F.Supp. 1128 (N.D.Ill.1976); Ballou v. State, Dept. of Civil Service, 148 N.J.Super. 112, 372 A.2d 333 (App.Div.1977), aff'd, 75 N.J. 365, 382 A.2d 1118 (1978). Three of the decisions distinguish Anthony v. Commonwealth, supra, as having been based on a stronger negative effect on women than those courts faced. The California court, however, states that the approach used in Anthony was “rejected in Washington v. Davis", Bannerman, 436 F.Supp. at 1280. Each court had' little trouble in concluding that no intent to harm women was present, even in the “absolute” preference at issue in New Jersey. The Illinois court’s language is representative.
While those who never served in the armed forces, those who served at times not within the statutory periods and women who are not veterans suffer a disadvantage in hiring and promotion, this is an incidental result of a statute intended to reward veterans and not one intended to discriminate against men and women who are not veterans or those whose service was in times of limited military action.
Branch v. DuBois, 418 F.Supp. at 1133.9
The impact of the statute at issue here does not approach the extremes described in Arlington Heights, supra 429 U.S. at 266, 97 S.Ct. 555, and plaintiff must prove intent by other evidence. This she has not done. The question: “Would the veterans’ preference statute have been enacted if women were represented in the armed services in such numbers that the preference would have no discriminatory effect?” has not been addressed by plaintiff, and she has given the court absolutely no reason to answer this question in the negative. She has failed to make out a prima facie case of discriminatory intent. See Mt. Healthy City Board of Ed. v. Doyle, 429 U.S. 274, 287, 97 S.Ct. 568, 50 L.Ed.2d 471 (1977). In light of Washington v. Davis I would not hold, as the majority does, that the Massachusetts Veterans’ Preference statute violates the Equal Protection Clause of the Fourteenth Amendment. I respectfully dissent.
. Unequal treatment of plaintiffs interest in the opportunity for public employment under a statute serving ends otherwise within the power of the state to pursue violates no fundamental interest guaranteed to plaintiff by the federal constitution. Massachusetts Bd. of Retirement v. Murgia, 427 U.S. 307, 96 S.Ct. 2562, 49 L.Ed.2d 520 (1976). Since the statute here is neutral on its face, and since it is undisputed that the statute was not enacted to harm women, the statutory scheme to benefit veteran men and women in the area of public employment to the disadvantage of non-veteran men and non-veteran women does not offend the equal protection clause of the Fourteenth Amendment.
. Judge Campbell states this result is an “inescapable and ‘built-in’ ” feature of the law, ante at 151. But in weighing his argument that the statute is for that reason, inter alia, impermissibly discriminatory against women, it cannot be overlooked that the unfavorable impact of the statute is shared alike by non-veteran women and a large number of non-veteran men.
. See the concurring opinion of Mr. Justice Stewart, joined by Mr. Justice Powell, in United Jewish Organizations of Williamsburgh, Inc. v. Carey, 430 U.S. 144, 180, 97 S.Ct. 996, 1017, 51 L.Ed.2d 229 (1977):
That the legislature was aware of race when it drew the district lines might also suggest a discriminatory purpose. Such awareness is not, however, the equivalent of discriminatory intent.
. The effect of certain statutory enactments would appear to be protective of women. See St. 1895, c. 501, § 1 and St.1896, c. 517, § 2. Each sets out details of the preference and concludes: “But nothing herein contained shall be construed to prevent the certification and employment of women.” See Opinion of the Justices, 166 Mass. 589, 592-593, 44 N.E. 6251 (1896). The legislature in 1971 revised the provision allowing single sex requisitions, with the result that the number of “women’s” jobs protected from the preference was severely limited, but the purpose of the revision would appear to be the prevention of occupational sex discrimination: the statute allows single sex requisitions only after approval has been obtained from the Massachusetts Commission Against Discrimination. Mass.Gen.Laws ch. 31, § 2A(e). See also G. Blumberg, De Facto and De Jure Sex Discrimination Under the Equal Protection Clause: A Reconsideration of the Veterans’ Preference in Public Employment, 26 Buff.L.Rev. 3, 38 (1976-77).
. The Court may be referring to the difference between an inference of intent from the cumulative impact of a series of administrative determinations and an inference from the impact of a rule promulgated by prior legislative or administrative action, see Shield Club v. City of Cleveland, 14 E.P.D. ¶ 7763 at 5772 (N.D.Oh. 1976); it may be referring to the presumption, more likely in jury cases than in other cases, that the result of selection will be random, see J. Ely, Legislative and Administrative Motivation in Constitutional Law, 79 Yale L.J. 1205, 1263-66.
. An important point in Judge Campbell’s analysis is the following:
To be sure, the legislature did not wish to harm women. But the cutting-off of women’s opportunities was an inevitable concomitant of the chosen scheme — as inevitable as the proposition that if tails is up, heads must be down. Where a law’s consequences are that inevitable, can they meaningfully be described as unintended?
Ante at 151. The answer to his question must be that inevitability of effect is relevant only where it bears on intent, and to find intent as that word is used in Washington v. Davis one must find motive. Judge Campbell agrees that “[w]hile the harm to female employment opportunities is extensive and, given the statutory scheme, inevitable, it was not this harm which prompted passage of the law . . . Ante at 150, n.*. Where, as here, a law’s consequences were inevitable, but there is no evidence at all that those particular consequences motivated the legislature, they can indeed be described as unintended.
. See P. Brest, supra, 1971 Sup.Ct.Rev. at 121— 122; Note, Reading the Mind of the School Board: Segregative intent and the De Facto/De Jure Distinction, 86 Yale L.J. 317, 332 — 43 (1976).
. For one of the positions applied for by plaintiff, that of Solomon Head Administrative Assistant, the three applicants certified of whom one would be chosen, had scores of 77.40, 93.-28, and 90.20. Without the veterans’ preference, the top three scores would have been 94.88, 93.28, and 92.32 (plaintiff). Agreed Statement of Facts (hereinafter “Statement”) 12, 13, Exhibits 2, 4. For another position, that of Administrative Assistant, there were seven positions available. Eleven persons would be certified, Statement ¶ 9, and were the top eleven all to indicate interest, the positions would be filled from a group with scores of 88, 86, 86, 84, 94, 92, 92, 92, 90, 90, and 90. Without the preference, the selections would be from a group with scores of 94, 92, 92, 92, 91, 90, 90, 90, 90, 89, and 89. Statement 111) 16, 17, Exhibit 7. For a third position, Assistant Secretary, Board of Dental Examiners, the top three scores were 89.72, 78.08, and 83.64; without the preference, the top three scores would have been 89.72, 86.68 (plaintiff), and 83.98. Statement !| 27, Exhibit 61. That the appointee for this position had a score of 78.08, the lowest of the three certified, indicates that there are other important qualifications besides test scores and thus that there is little reason to believe that the quality of the employee pool is significantly lowered by its containing persons with slightly lower test scores than would be present absent the veterans’ preference statute.
. This court would seem to have agreed in its earlier opinion, where the majority stated that
[t]he Massachusetts Veterans’ Preference was not enacted for the purpose of disqualifying women from receiving civil service appointments.
Anthony v. Commonwealth of Massachusetts, 415 F.Supp. 485, 495 (1976).
Nowhere in his opinion has Judge Tauro said that the Massachusetts legislature intended to harm job opportunities for women or that limiting such opportunities was a motive in enactment of the legislation, and that, of course, is precisely what must be shown. All Judge Tauro will say is that the legislature’s “clear intent was to benefit veterans even at the expense of women”, ante at 146. This says nothing about motive and is entirely consistent with a finding that the legislature saw the impact on women as extremely regrettable but unavoidable.