delivered the opinion of the Court.

These cases arise from the Veterans’ Administration’s refusal to grant two recovered alcoholics extensions of time in which to use their veterans’ educational benefits. We must decide whether the Veterans’ Administration’s decision is subject to judicial review and, if so, whether that decision violates § 504 of the Rehabilitation Act of 1973, 87 Stat. 394, 29 U. S. C. § 794, which requires that federal programs not discriminate against handicapped individuals solely because of their handicap.1

*538I

Veterans who have been honorably discharged from the United States Armed Forces are entitled to receive educational assistance benefits under the Veterans’ Readjustment Benefit Act of 1966 (“GI Bill”) to facilitate their readjustment to civilian life. See 38 U. S. C. § 1661. These benefits generally must be used within 10 years following discharge or release from active duty. § 1662(a)(1). Veterans may obtain an extension of the 10-year delimiting period, however, if they were prevented from using their benefits earlier by “a physical or mental disability which was not the result of [their] own willful misconduct.” Ibid.

Petitioners are honorably discharged veterans who did not exhaust their educational benefits during the decade following their military service. They sought to continue to receive benefits after the expiration of the 10-year delimiting period on the ground that they had been disabled by alcoholism during much of that period. The Veterans’ Administration determined that petitioners’ alcoholism constituted “willful misconduct” under 38 CFR § 3.301(c)(2) (1987),2 and accordingly denied the requested extensions.

*539Petitioner Traynor sought review of the Veterans’ Administration’s decision in the United States District Court for the Southern District of New York. The District Court held that it was not foreclosed from exercising jurisdiction over the case by 38 U. S. C. § 211(a), which bars judicial review of “the decisions of the Administrator on any question of law or fact under any law administered by the Veterans’ Administration providing benefits for veterans,”3 because the complaint “requires us to examine constitutional and statutory questions and not merely issues of VA policy.” Traynor v. Walters, 606 F. Supp. 391, 396 (1985). The court rejected Traynor’s claim that the Veterans’ Administration’s refusal to extend his delimiting period violated the Due Process Clause and the equal protection component of the Fifth Amendment.4 However, the court concluded that alcoholism is a handicap within the meaning of the Rehabilitation Act, and that the Veterans’ Administration therefore had engaged in the sort of discrimination on the basis of handicap that is forbidden by that Act.

A divided panel of the Court of Appeals for the Second Circuit reversed on the ground that § 211(a) barred judicial re*540view of the Rehabilitation Act claim. Traynor v. Walters, 791 F. 2d 226 (1986). The court reasoned that, while “many veterans have in the service of our country suffered injuries that qualify them as ‘handicapped individuals]’ for purposes of [the Rehabilitation Act],” Congress evinced no intent in enacting that statute “to grant to ‘handicapped’ veterans the judicial review traditionally denied all other veterans” under §211(a). Id., at 229.5

Meanwhile, petitioner McKelvey sought review of the Veterans’ Administration’s decision in the District Court for the District of Columbia. The District Court exercised jurisdiction over McKelvey’s claims on the ground that §211 (a) permits judicial review of decisions rejecting claims that Veterans’ Administration regulations of general applicability violate a federal statute that is “completely independent of the complex statutory and regulatory scheme for dispersing veterans’ benefits.” McKelvey v. Walters, 596 F. Supp. 1317, 1321 (1984). The court then invalidated 38 CFR § 3.301(c) (2) (1987) as contrary to the Rehabilitation Act. The court ordered the Veterans’ Administration to determine without resort to the regulation whether McKelvey had suffered a disability attributable to his own misconduct.

On appeal, the Court of Appeals for the District of Columbia Circuit agreed that judicial review was not foreclosed by § 211(a), which was held to apply only to claims “resolved by an actual ‘decision of the Administrator.’” 253 U. S. App. D. C. 126, 130, 792 F. 2d 194, 198 (1986) (per curiam) (quoting Johnson v. Robison, 415 U. S. 361, 367 (1974)). The court found that no such decision had been rendered by the Veterans’ Administration as to the validity of 38 CFR *541§3.301(c)(2) (1987) under the Rehabilitation Act.6 On the merits, however, the Court of Appeals reversed, holding that the Veterans’ Administration could consistently with the Rehabilitation Act distinguish between veterans who are at least to some extent responsible for their disabilities and veterans who are not.7 With respect to alcoholism, this distinction could be effected by means of § 3.301(c)(2), said the court, because the Veterans’ Administration could reasonably conclude that alcoholism is a “willfully caused handicap” unless attributable to an underlying psychiatric disorder. 253 U. S. App. D. C., at 132-133, 792 F. 2d, at 200-201. The court expressed disagreement with Tinch v. Walters, 765 F. 2d 599 (CA6 1985), which had invalidated the regulation in light of the Rehabilitation Act. See 253 U. S. App. D. C., at 133, n. 4, 792 F. 2d, at 201, n. 4.

We granted certiorari to resolve the conflicts between the Courts of Appeals as to whether Veterans’ Administration decisions challenged under the Rehabilitation Act are subject to judicial review and, if so, whether that Act bars the Veterans’ Administration from characterizing petitioners’ alcoholism as “willful misconduct” for purposes of 38 U. S. C. § 1662(a)(1). 480 U. S. 916 (1987).

II

We must first consider whether §211(a)’s bar against judicial review of “the decisions of the Administrator on any question of law or fact under any law administered by the Veterans’ Administration providing benefits for veterans” ex*542tends to petitioners’ claim that the Veterans’ Administration regulation defining primary alcoholism as “willful misconduct” discriminates against handicapped persons in violation of the Rehabilitation Act.

We have repeatedly acknowledged “the strong presumption that Congress intends judicial review of administrative action.” Bowen v. Michigan Academy of Family Physicians, 476 U. S. 667, 670 (1986); see also Dunlop v. Bachowski, 421 U. S. 560, 567 (1975); Barlow v. Collins, 397 U. S. 159, 166-167 (1970). The presumption in favor of judicial review may be overcome “only upon a showing of ‘clear and convincing evidence’ of a contrary legislative intent.” Abbott Laboratories v. Gardner, 387 U. S. 136, 141 (1967) (citations omitted). We look to such evidence as “‘specific language or specific legislative history that is a reliable indicator of congressional intent,’ or a specific congressional intent to preclude judicial review that is ‘fairly discernible in the detail of the legislative scheme.’” Bowen v. Michigan Academy of Family Physicians, supra, at 673 (quoting Block v. Community Nutrition Institute, 467 U. S. 340, 349, 351 (1984)).

In Johnson v. Robison, supra, we held that the federal courts could entertain constitutional challenges to veterans’ benefits legislation. We determined that “neither the text nor the scant legislative history of § 211(a)” provided the requisite “clear and convincing” evidence of congressional intent to foreclose judicial review of challenges to the constitutionality of a law administered by the Veterans’ Administration., 415 U. S., at 373-374. In that case, the Veterans’ Administration, acting under 38 U. S. C. §§ 101(21), 1652(a)(1), and 1661(a), denied educational benefits to a conscientious objector who had completed the required alternative civilian service. The claimant brought suit in the District Court, challenging those statutory sections on First and Fifth Amendment grounds. The District Court denied a motion to dismiss based on § 211(a) and gave judgment to the plaintiff. *543Robison v. Johnson, 352 F. Supp. 848 (Mass. 1973). We agreed that § 211(a) did not bar the suit, but reversed the judgment on the merits. On the § 211(a) issue, we reasoned that “[t]he prohibitions [of § 211(a)] would appear to be aimed at review only of those decisions of law or fact that arise in the administration by the Veterans’ Administration of a statute providing benefits for veterans.” 415 U. S., at 367. The'questions of law presented in that case, however, arose under the Constitution rather than under the veterans’ benefits statute and concerned whether there was a valid law on the subject for the Veterans’ Administration to execute. We went on to conclude that the principal purposes of § 211(a)— “(1) to insure that veterans’ benefits claims will not burden the courts and the Veterans’ Administration with expensive and time-consuming litigation, and (2) to insure that the technical and complex determinations and applications of Veterans’ Administration policy connected with veterans’ benefits decisions will be adequately and uniformly made,” id., at 370 — would not be frustrated if federal courts were permitted to exercise jurisdiction over constitutional challenges to the very statute that was sought to be enforced. We noted that such challenges “cannot be expected to burden the courts by their volume, nor do they involve technical consideration of Veterans’ Administration policy.” Id., at 373.

The text and legislative history of § 211(a) likewise provide no clear and convincing evidence of any congressional intent to preclude a suit claiming that § 504 of the Rehabilitation Act, a statute applicable to all federal agencies, has invalidated an otherwise valid regulation issued by the Veterans’ Administration and purporting to have the force of law. Section 211(a) insulates from review decisions of law and fact “under any law administered by the Veterans’ Administration,” that is, decisions made in interpreting or applying a particular provision of that statute to a particular set of facts. Id., at 367. But the cases now before us involve the issue whether the law sought to be administered is valid in light of *544a subsequent statute whose enforcement is not the exclusive domain of the Veterans’ Administration.8 There is no claim that the regulation at issue is inconsistent with the statute under which it was issued; and there is no challenge to the Veterans’ Administration’s construction of any statute dealing with veterans’ benefits, except to the extent that its construction may be affected by the Rehabilitation Act. Nor is there any reason to believe that the Veterans’ Administration has any special expertise in assessing the validity of its regulations construing veterans’ benefits statutes under a later passed statute of general application. Permitting these cases to go forward will not undermine the purposes of § 211(a) any more than did the result in Johnson. It cannot be assumed that the availability of the federal courts to decide whether there is some fundamental inconsistency between the Veterans’ Administration’s construction of veterans’ benefits statutes, as reflected in the regulation at issue here, and the admonitions of the Rehabilitation Act will enmesh the courts in “the technical and complex determinations and applications of Veterans’ Administration policy connected with veterans’ benefits decisions” or “burden the courts and the Veterans’ Administration with expensive and time-consuming litigation.” Id., at 370.9 Of course, if *545experience proves otherwise, the Veterans’ Administration is fully capable of seeking appropriate relief from Congress.

Accordingly, we conclude that the question whether a Veterans’ Administration regulation violates the Rehabilitation Act is not foreclosed from judicial review by § 211(a). We therefore turn to the merits of petitioners’ Rehabilitation Act claim.

Ill

Congress historically has imposed time limitations on the use of “GI Bill” educational benefits. Veterans of World War II were required to use their benefits within nine years after their discharge from military service, while Korean Conflict veterans had eight years in which to use their benefits. See S. Rep. No. 93-977, p. 13 (1974) (letter to Hon. Vance Hartke from Veterans’ Administrator Johnson). The delimiting period under the current “GI Bill” was raised from 8 years to 10 years in 1974. Pub. L. 93-337, § 2(1), 88 Stat. 292, 38 U. S. C. §§ 1712(b)(1), (2). In 1977, Congress created an exception to this 10-year delimiting period for veterans who delayed their education because of “a physical or mental disability which was not the result of [their] own willful misconduct.” Pub. L. 95-202, Tit. II, § 203(a)(1), 91 Stat. 1429, 38 U. S. C. § 1662(a)(1).

Congress did not use the term “willful misconduct” inadvertently in § 1662(a)(1). The same term had long been used in other veterans’ benefits statutes. For example, veterans are denied compensation for service-connected disabilities that are “the result of the veteran’s own willful misconduct.” 38 U. S. C. § 310. See also § 521 (compensation for disabilities not connected with military service). The Veterans’ Administration had long construed the term “willful misconduct” for purposes of these statutes as encompassing primary alcoholism (i. e., alcoholism that is not “secondary to and a manifestation of an acquired psychiatric disorder”). See n. 2, supra.

*546“It is always appropriate to assume that our elected representatives, like other citizens, know the law.” Cannon v. University of Chicago, 441 U. S. 677, 696-697 (1979). Hence, we must assume that Congress was aware of the Veterans’ Administration’s interpretation of “willful misconduct” at the time that it enacted § 1662(a)(1), and that Congress intended that the term receive the same meaning for purposes of that statute as it had received for purposes of other veterans’ benefits statutes. See Sedima, S. P. R. L. v. Imrex Co., 473 U. S. 479, 489 (1985); Morrison-Knudsen Construction Co. v. Director, Office of Workers’ Compensation Programs, 461 U. S. 624, 633 (1983); Bob Jones University v. United States, 461 U. S. 574, 586-587, and n. 10 (1983). In these cases, however, we need not rely only on such assumptions. The legislative history confirms that Congress intended that the Veterans’ Administration apply the same test of “willful misconduct” in granting extensions of time under § 1662(a)(1) as the agency already was applying in granting disability compensation under § 310 and § 521. Specifically, the Report of the Senate Veterans’ Affairs Committee on the 1977 legislation states:

“In determining whether the disability sustained was a result of the veteran’s own ‘willful misconduct,’ the Committee intends that the same standards be applied as are utilized in determining eligibility for other VA programs under title 38. In this connection, see 38 CFR, part III, paragraphs 3.1(n) and 3.301, and VA Manual M21-1, section 1404.” S. Rep. No. 95-468, pp. 69-70 (1977).

The cited regulations include 38 CFR § 3.301(c)(2) (1987), the regulation that characterizes primary alcoholism as “willful misconduct.” The Veterans’ Administration Manual provision states, inter alia, that “[b]asic principles for application in deciding cases involving alcoholism are stated in Administrator’s Decision No. 988,” the decision on which § 3.301(c)(2) is based. VA Manual M21-1, change 132, subch. I, § 14.04c (Jan. 29, 1976). See n. 2, supra. These sources set forth *547the criteria for determining whether a veteran’s alcoholism is the result of “willful misconduct.” These criteria therefore are among the “standards” that, according to the Senate Report, Congress intended to be utilized in determining eligibilty for extended educational benefits.

It is thus clear that the 1977 legislation precluded an extension of time to a veteran who had not pursued his education because of primary alcoholism. If Congress had intended instead that primary alcoholism not be deemed “willful misconduct” for purposes of § 1662(a)(1), as it had been deemed for purposes of other veterans’ benefits statutes, Congress most certainly would have said so.

It was the same Congress that one year later extended §504’s prohibition against discrimination on the basis of handicap to “any program or activity conducted by any Executive agency.” Pub. L. 95-602, Tit. IV, §§ 119, 122(d)(2), 92 Stat. 2982, 2987, 29 U. S. C. § 794. Yet, in enacting the 1978 Rehabilitation Act amendments, Congress did not affirmatively evince any intent to repeal or amend the “willful misconduct” provision of § 1662(a)(1). Nor did Congress anywhere in the language or legislative history of the 1978 amendments expressly disavow its 1977 determination that primary alcoholism is not the sort of disability that warrants an exemption from the time constraints of § 1662(a)(1).

Accordingly, petitioners can prevail under their Rehabilitation Act claim only if the 1978 legislation can be deemed to have implicitly repealed the “willful misconduct” provision of the 1977 legislation or forbade the Veterans’ Administration to classify primary alcoholism as willful misconduct. They must thereby overcome the “ ‘cardinal rule . . . that repeals by implication are not favored.’” Morton v. Mancari, 417 U. S. 535, 549-550 (1974) (quoting Posadas v. National City Bank, 296 U. S. 497, 503 (1936); Wood v. United States, 16 Pet. 342, 363 (1842); Universal Interpretive Shuttle Corp. v. Washington Metropolitan Area Transit Comm’n, 393 U. S. 186, 193 (1968)). “It is a basic principle of statutory con*548struction that a statute dealing with a narrow, precise, and specific subject is not submerged by a later enacted statute covering a more generalized spectrum,” Radzanower v. Touche Ross & Co., 426 U. S. 148, 153 (1976), unless the later statute “‘expressly contradices] the original act’” or unless such a construction “‘is absolutely necessary ... in order that [the] words [of the later statute] shall have any meaning at all.’” Ibid, (quoting T. Sedgwick, The Interpretation and Construction of Statutory and Constitutional Law 98 (2d ed. 1874)). “The courts are not at liberty to pick and choose among congressional enactments, and when two statutes are capable of co-existence, it is the duty of the courts, absent a clearly expressed congressional intention to the contrary, to regard each as effective.” Morton v. Mancari, supra, at 551.

As we have noted, the 1978 legislation did not expressly contradict the more “narrow, precise, and specific” 1977 legislation. Moreover, the 1978 legislation is not rendered meaningless, even with respect to those who claim to have been handicapped as a result of alcoholism, if the “willful misconduct” provision of § 1662(a)(1) is allowed to retain the import originally intended by Congress.

First, the “willful misconduct” provision does not undermine the central purpose of §504, which is to assure that handicapped individuals receive “evenhanded treatment” in relation to nonhandicapped individuals. Alexander v. Choate, 469 U. S. 287, 304 (1985); Southeastern Community College v. Davis, 442 U. S. 397, 410 (1979). This litigation does not involve a program or activity that is alleged to treat handicapped persons less favorably than nonhandicapped persons. Cf. School Board of Nassau County v. Arline, 480 U. S. 273 (1987); Southeastern Community College, supra. Rather, petitioners challenge a statutory provision that treats disabled veterans more favorably than able-bodied veterans: The former may obtain extensions of time in which to use their educational benefits so long as they did not become *549disabled as a result of their own “willful misconduct”; the latter are absolutely precluded from obtaining such extensions regardless of how compelling their reasons for having delayed their schooling might be. In other words, § 1662(a)(1) merely provides a special benefit to disabled veterans who bear no responsibility for their disabilities that is not provided to other disabled veterans or to any able-bodied veterans.

There is nothing in the Rehabilitation Act that requires that any benefit extended to one category of handicapped persons also be extended to all other categories of handicapped persons. Hence, the regulations promulgated by the Department of Health, Education, and Welfare in 1977 with regard to the application of §504 to federally funded programs provide that “exclusion of a specific class of handicapped persons from a program limited by Federal statute or executive order to a different class of handicapped persons” is not prohibited. 42 Fed. Reg. 22676, 22679 (1977), promulgating 45 CFR § 84.4(c) (1986).10 It is therefore not inconsistent with the Rehabilitation Act for only those veterans whose disabilities are not attributable to their own “willful misconduct” to be granted extensions of the 10-year delimiting period applicable to all other veterans. Congress is entitled to establish priorities for the allocation of the limited resources available for veterans’ benefits, cf. McDonald v. Board of Election Comm’rs of Chicago, 394 U. S. 802, 809 (1969), and thereby to conclude that veterans who bear some responsibility for their disabilities have no stronger claim to an extended eligibility period than do able-bodied veterans. Those veterans are not, in the words of § 504, denied benefits *550“solely by reason of [their] handicap,” but because they engaged with some degree of willfulness in the conduct that caused them to become disabled.

Furthermore, § 1662(a)(1) does not deny extensions of the delimiting period to all alcoholics but only to those whose drinking was not attributable to an underlying psychiatric disorder. It is estimated by some authorities that mental illness is responsible for 20% to 30% of all alcoholism cases. Brief for American Medical Association as Amicus Curiae 7. Each veteran who claims to have been disabled by alcoholism is entitled under § 1662(a)(1) to an individualized assessment of whether his condition was the result of a mental illness.

Petitioners, however, perceive an inconsistency between § 504 and the conclusive presumption that alcoholism not motivated by mental illness is necessarily “willful.” They contend that § 504 mandates an individualized determination of “willfulness” with respect to each veteran who claims to have been disabled by alcoholism. It would arguably be inconsistent with § 504 for Congress to distinguish between categories of disabled veterans according to generalized determinations that lack any substantial basis. If primary alcoholism is not always “willful,” as that term has been defined by Congress and the Veterans’ Administration, some veterans denied benefits may well be excluded solely on the basis of their disability. We are unable to conclude that Congress failed to act in accordance with § 504 in this instance, however, given what the District of Columbia Circuit accurately characterized as “a substantial body of medical literature that even contests the proposition that alcoholism is a disease, much less that it is a disease for which the victim bears no responsibility.” 253 U. S. App. D. C., at 132-133, 792 F. 2d, at 200-201. Indeed, even among many who consider alcoholism a “disease” to which its victims are genetically predisposed, the consumption of alcohol is not regarded as wholly involuntary. See Fingarette, The Perils of Powell: In Search of a Factual Foundation for the “Disease Concept of *551Alcoholism,” 83 Harv. L. Rev. 793, 802-808 (1970). As we see it, §504 does not demand inquiry into whether factors other than mental illness rendered an individual veteran’s drinking so entirely beyond his control as to negate any degree of “willfulness” where Congress and the Veterans’ Administration have reasonably determined for purposes of the veterans’ benefits statutes that no such factors exist.11

In sum, we hold that a construction of § 1662(a)(1) that reflects the original congressional intent that primary alcoholics not be excused from the 10-year delimiting period for utilizing “GI Bill” benefits is not inconsistent with the prohibition on discrimination against the handicapped contained in § 504 of the Rehabilitation Act.12 Accordingly, since we “are not at liberty to pick and choose among congressional enactments . . . when two statutes are capable of co-existence,” Morton v. Mancari, 417 U. S., at 551, we must conclude that the earlier, more specific provisions of § 1662(a)(1) were neither expressly nor implicitly repealed by the later, more general provisions of § 504.

*552IV

This litigation does not require the Court to decide whether alcoholism is a disease whose course its victims cannot control. It is not our-role to resolve this medical issue on which the authorities remain sharply divided. Our task is to decide whether Congress intended, in enacting § 504 of the Rehabilitation Act, to reject the position taken on the issue by the Veterans’ Administration and by Congress itself only one year earlier. In our view, it is by no means clear that § 504 and the characterization of primary alcoholism as a willfully incurred disability are in irreconcilable conflict. If petitioners and their proponents continue to believe that this position is erroneous, their arguments are better presented to Congress than to the courts.

The judgment of the Court of Appeals for the District of Columbia Circuit in No. 86-737 is affirmed. The judgment of the Court of Appeals for the Second Circuit in No. 86-622 is reversed, and the case is remanded for further proceedings consistent with this opinion.

It is so ordered.

with whom

I join Parts I and II of the Court’s opinion, for I agree that, under § 504 of the Rehabilitation Act of 1973, 87 Stat. 394, as amended, 29 U. S. C. § 794, the “final and conclusive” language of 38 U. S. C. § 211(a) does not bar judicial review of petitioners’ claims. Similarly, I acknowledge the legality (a) of the 10-year delimiting period imposed by 38 U. S. C. § 1662(a) upon veterans’ educational assistance, and (b) of that statute’s alleviation of the delimiting period in cases of disability except where that disability is the result of a veteran’s “own willful misconduct.”

*553My dispute with the Court centers in its upholding of the regulation, 38 CFR § 3.301(c)(2) (1987), whereby the Veterans’ Administration (VA) presumes, irrebuttably, that primary alcoholism always is the result of the veteran’s “own willful misconduct.” This is the very kind of broad social generalization that § 504 of the Rehabilitation Act is intended to eliminate. The petitioners in these cases ask only that their situations be given individualized evaluation. Because I think this is what the Rehabilitation Act clearly requires, I dissent from the Court’s conclusion to the contrary.1

I

Petitioner Eugene Traynor began drinking when he was eight or nine years old. He drank with increasing frequency throughout his teenage years, and was suffering alcohol-related seizures by the time he was on active military duty in Vietnam. During the four years following his honorable discharge in 1969, Mr. Traynor was hospitalized repeatedly for alcoholism and related illnesses.

By the end of 1974, however, petitioner Traynor had conquered his drinking problem. He attended college part-time beginning in 1977, and continued working toward his degree until the 10-year period for using his veteran’s' educational benefits expired for him in 1979. Mr. Traynor applied for the extension of time available under 38 U. S. C. § 1662(a)(1) to one whose disability had prevented him from completing a program of education within the 10-year period. Because he was unable to establish that his alcoholism was due to an underlying psychiatric disorder, his condition was labeled “primary alcoholism.” Pursuant to the regulation cited above, Mr. Traynor was presumed to have brought his alcoholism upon himself through “willful misconduct.” The requested extension therefore was denied.

*554Petitioner James P. McKelvey also started drinking as a child. He was 13 when he began to develop the alcohol dependency that was common among members of his family. His drinking problem plagued him while he was in the Army, and he was hospitalized frequently during the nine years that followed his honorable discharge in 1966. Despite his disability, however, McKelvey managed, between hospital stays, to attend two educational institutions under the veterans’ educational-benefits program.

Mr. McKelvey took his last drink in 1975, only a year and a half before his 10-year delimiting period expired. Like Traynor, McKelvey sought an extension under 38 U. S. C. § 1662(a)(1) on the ground that his alcoholism had prevented him from using, within the period, the benefits to which he was entitled. And, like Traynor, McKelvey was denied the extension because his disability, primary alcoholism, was conclusively presumed to have been caused by his “own willful misconduct.” The VA’s regulation deprived each of these veterans of any opportunity to establish that, in his particular case, disabling alcoholism was not willfully incurred.

II

The VA’s reliance on its irrebuttable presumption that all primary alcoholism is attributable to willful misconduct cannot be squared with the mandate against discrimination contained in § 504 of the Rehabilitation Act. Just last year, in School Bd. of Nassau County v. Arline, 480 U. S. 273 (1987), this Court explained in no uncertain terms that § 504 bars the generic treatment of any group of individuals with handicaps based on archaic or simplistic stereotypes about attributes associated with their disabling conditions. Instead, § 504 requires an individualized assessment of each person’s qualifications, based on “reasoned and medically sound judgments.” Id., at 285. In sanctioning the VA’s irrebuttable presumption that any veteran suffering from primary aleo*555holism brought the ailment upon himself through willful misconduct, the Court ignores the lesson of Arline, and the clear dictate of the Rehabilitation Act.

In these cases, the Court is called upon, not to make its own medical judgments about the causes of alcoholism, but to interpret § 504. That statute sets forth a simple rule:

“No otherwise qualified individual with handicaps . . . shall, solely by reason of his handicap, be excluded from the participation in, be denied the benefits of, or be subjected to discrimination under any program or activity receiving Federal financial assistance or under any program or activity conducted by any Executive agency. . . .”2

It is beyond dispute that petitioners, as alcoholics, were handicapped individuals covered by the Act. See 43 Fed. Reg. 2137 (1978) (guidelines issued by Department of Health, Education, and Welfare (later the Department of Health and Human Services) reflecting the Attorney General’s specific conclusion, 43 Op. Atty. Gen. No. 12, p. 2 (1977), that an alcoholic is covered by the Act); see also Brief for Respondents 33. Nor is it disputed that § 504 of the Act prohibits federal agencies, such as the VA, from denying benefits to petitioners solely because they are alcoholics. See, e. g., 38 CFR §§ 18.403(j)(1) and (2)(i)(c) (1987).

In 38 U. S. C. § 1662(a)(1), Congress declared that one whose disability resulted from “willful misconduct” is not entitled to the benefit of the extension-of-time provision. As stated above, the VA, by its regulation, has established an irrebuttable presumption that primary alcoholism is the result of willful misconduct. This presumption appears to be a clear violation of § 504’s mandate requiring individualized *556assessment of each claimant’s qualifications. Arline, 480 U. S., at 287-289.

Respondents argue, however, that a case-by-case assessment of whether a claimant’s alcoholism was the result of willful misconduct is not necessary for two reasons. First, respondents contend that Congress, in enacting the 1977 amendment (the extension-of-time provision, see 91 Stat. 1439) to § 1662(a), mandated a conclusive presumption that primary alcoholism is caused by willful misconduct. Second, respondents contend that the VA’s determination that primary alcoholism always is due to willful misconduct is reasonable, and that therefore the presumption is not based on the kind of stereotyping that §504 forbids. The Court today finds each of these arguments persuasive. In my view, each patently is without merit.

Ill

The Court explains:

“As we see it, §504 does not demand inquiry into whether factors other than mental illness rendered an individual veteran’s drinking so entirely beyond his control as to negate any degree of ‘willfulness’ where Congress and the Veterans’ Administration have reasonably determined for purposes of the veterans’ benefits statutes that no such factors exist.” Ante, at 551.

As I see it, § 504 demands precisely the inquiry the Court says is unnecessary. While Congress certainly has the authority to determine that primary alcoholism always should t be attributed to willful misconduct, I find no support whatever for the Court’s conclusion that Congress made that determination when it amended § 1662(a) in 1977.

The Court is correct, of course, see ante, at 546, when it says that we must assume that Congress intended the term “willful misconduct” in § 1662(a)(1) to have the same meaning it had been given in other veterans’ benefits statutes. Indeed, the legislative history indicates that Congress did *557“inten[d] that the same standards be applied as are utilized in determining eligibility for other VA programs under title 38.” S. Rep. No. 95-468, pp. 69-70 (1977). If §504 had not been amended one year later to cover specifically all executive agency programs, including the VA’s benefits programs, see Pub. L. 95-602, §§ 119, 122(d)(2), 92 Stat. 2982, 2987, 29 U. S. C. § 794, there would be little reason to question the application of the VA’s interpretation of the willful-misconduct regulation to § 1662(a)(1).

But the Court goes further and finds that Congress’ reference to the VA’s willful-misconduct regulation in amending § 1662(a) is a congressional adoption of the VA’s rule. The Court transforms Congress’ uncontroversial statement that the willful-misconduct regulation should be given the same meaning throughout the statutory scheme into a “specific] determination]” by Congress that primary alcoholics are presumed to have engaged in willful misconduct. See ante, at 551, n. 11; see also ante, at 547 (Congress’ “1977 determination that primary alcoholism is not the sort of disability that warrants an exemption”); ante, at 548 (Congress had “‘narrow, precise, and specific’” intent to exclude primary alcoholics in enacting § 1662(a)(1)); ante, at 551 (“original congressional intent [in amending § 1662(a)] that primary alcoholics not be excused from the 10-year delimiting period”). This magical transformation is the linchpin in the Court’s analysis, for unless Congress itself actually took a position in 1977 endorsing the association of primary alcoholism with willful misconduct, the subsequent amendment of §504 in 1978 to include benefit programs like the VA’s would simply be read to impose new constraints on the VA’s treatment of alcoholics. There is nothing whatever that is inconsistent about Congress’ willingness, in 1977, to allow the VA to apply its own rules in determining which alcoholic veterans were entitled to benefits, and its decision, one year later, to require such determinations to comply with the antidiscrimination provisions of § 504 then being amended.

*558In order to escape § 504’s requirements, the majority must conclude that in 1977 Congress defined a primary alcoholic as not “otherwise qualified,” within the meaning of § 504, for the extension of time available under § 1662(a)(1). The language of § 1662(a)(1) itself merely establishes that a willfully incurred disability, as a general matter, does not entitle a veteran to the extension of time. And the Senate Report, upon which the Court exclusively relies, makes only passing reference to the relevant regulations — regulations which encompass the VA’s entire policy on the applicability of the willful-misconduct provisions, not just the application of that term to alcoholism. Finally, even those portions of the regulations expressly addressed to alcoholism do not state that primary alcoholism is to be equated with willful misconduct. That interpretation is derived from a 1964 Administrator’s Decision, which itself discusses the VA’s irrebuttable presumption only briefly. Administrator’s Decision, Veterans’ Administration No. 988, Interpretation of the Term “Willful Misconduct” as Related to the Residuals of Chronic Alcoholism 1 (1964).3 *559See 37 Fed. Reg. 20335, 20336 (1972) (proposing regulation and announcing that it was intended to incorporate principles of the 1964 administrative issue).

Surely something more than two sentences quoted from a Senate Report should be required before we interpret general statutory language to conflict with the most natural reading of subsequent specific legislation. It is only the Court’s strained reading of § 1662(a)(1) to embrace a congressional “determination that primary alcoholism is not the sort of disability that warrants an exemption,” ante, at 547, that leads the Court to reject as a disfavored “implicit repeal” § 504’s requirement that qualifications for the exemption be determined on a case-by-case basis. The “ ‘basic principle of statutory construction that a statute dealing with a narrow, precise, and specific subject is not submerged by a later enacted statute covering a more generalized spectrum,’” ante, at 547-548, has no application here, where the earlier enactment is not narrowly or specifically addressed to the matter treated generally in the subsequent enactment: federal agencies’ treatment of alcoholics. I have been no more successful than the VA or the Court in turning up evidence that Con*560gress expressly considered, or intended, in Amending § 1662 (a), to adopt legislatively the VA’s presumption that primary alcoholism always is attributable to willful misconduct. I therefore see no reason to defer to the VA’s rule in interpreting a subsequent and entirely separate congressional enactment that the VA has not been empowered to administrate.

IV

I am reluctant to conclude that anything short of a congressional determination linking all primary alcoholism to willful misconduct could justify the VA’s substitution of its generic rule for the individualized assessment generally required under § 504. It is conceivable that an agency legitimately could eschew individualized assessments of disabled individuals’ qualifications if it were evident, as a matter of medical fact, that a particular disqualifying characteristic always is associated with a particular disability.4 But it is not at all evident that an absolute correlation exists between the condition of primary alcoholism and the disqualifying factor of willful misconduct, as defined by the VA. Nor has the VA successfully demonstrated that such an absolute correlation is medically justified. The VA suggests that it is enough that “although the policy may not produce in an individual case the same conclusion another arbiter might reach, the VA policy provides a reasonable and workable accommodation of modern medico-psychological evidence.” Brief for Respondents 35 (emphasis added). This position is unsatisfactory for several reasons.

*561A

The VA seems to suggest that generalizations about attributes associated with individuals suffering from a particular disability can be relied upon to assess those individuals’ qualifications, as long as the generalizations are shown to be reasonable. But reliance on generalizations, even “reasonable” ones, is clearly prohibited under Arline. In that case, the Court ruled that §504 prevented the Nassau County School Board from generalizing about the contagiousness of tuberculosis. 480 U. S., at 281-286. Acknowledging that in some cases contagiousness would justify altering or perhaps terminating a tuberculosis sufferer’s employment in order to avoid infecting others, id., at 287, n. 16, the Court nevertheless found impermissible a generalization built on that less-than-perfect correlation between disability and qualification. The Court explained:

“The fact that some persons who have contagious diseases may pose a serious health threat to others under certain circumstances does not justify excluding from the coverage of the Act all persons with actual or perceived contagious diseases. Such exclusion would mean that those accused of being contagious would never have the opportunity to have their condition evaluated in light of medical evidence and a determination made as to whether they were ‘otherwise qualified.’ Rather, they would be vulnerable to discrimination on the basis of mythology — precisely the type of injury Congress sought to prevent.” Id., at 285 (emphasis in original).

The myth to which the Court was referring was not that some tuberculosis sufferers were contagious, but that they all were. The parallel myth in the present cases, of course, is that all primary alcoholics became disabled as a result of their own willful misconduct. Just as § 504 entitles each person suffering from tuberculosis to an individualized determination, based on sound medical evidence, as to whether *562that person is contagious and therefore not “otherwise qualified” for a job, 29 U. S. C. § 794, the statute entitles each alcoholic veteran to an individualized determination, based on the medical evidence in his own case, of the causes of his disability. If this individualized assessment leads the adjudicator to conclude that the particular veteran’s alcoholism was brought on by willful misconduct, that veteran will have been adjudicated to be not “otherwise qualified” to collect the education benefits. But only after this individualized inquiry has been conducted, can the VA deprive him of benefits available to all whose disabilities were not caused by willful misconduct.

B

The VA’s attempt to justify its reliance upon the irrebuttable presumption that primary alcoholism is caused by willful misconduct is further undermined by the meagerness of the medical support it summons. Nothing in the record suggests that the VA based its continuing reliance on the presumption, after § 504 was amended, on any factual findings of the kind found to be required in Arline. And its post hoc rationalization of that reliance in this litigation consists of a hodgepodge of medical conclusions, some of only marginal relevance. For example, the VA relies upon the comments of a number of “medical writers” who note that “volition plays a significant role” in the treatment of alcoholism. See Brief for Respondents 43; see id., at 45-46, and nn. 32 and 33, citing, among others, G. Vaillant, The Natural History of Alcoholism 299 (1983), and S. Zimberg, The Clinical Management of Alcoholism 67-69, 118 (1982). While cure and cause are likely to be somewhat related, the fact that alcoholism is “highly treatable, but . . . will require great responsibility from the patient,” G. Vaillant, supra, at 299, provides little assistance in assessing whether the original onset of the disability can always be ascribed to willful misconduct.

In contrast, ample evidence supports petitioners’ contrary contention that the degree of willfulness associated with the *563onset of alcoholism varies from case to case. Recent medical research indicates that the causes of primary alcoholism5 are varied and complex, only some of which conceivably could be attributed to a veteran’s will.6 Indeed, even the VA acknowledges that “alcoholism is not a unitary condition [but rather] has multiple forms and ranges of severity.” Brief for Respondents 34, and nn. 21, 22, citing, among others, Bohman, Sigvardsson, & Cloninger, Maternal Inheritance of Alcohol Abuse: Cross-Fostering Analysis of Adopted Women, 38 Archives Gen. Psychiatry 965, 968 (1981) (describing genetically different types of alcoholism, each producing a different form of the condition); Cloninger, Bohman, & Sigvardsson, Inheritance of Alcohol Abuse: Cross-Fostering Analysis of Adopted Men, 38 Archives Gen. Psychiatry 861, 867 (1981) (identifying two types of alcohol abuse with different genetic and environmental causes); G. Vaillant, supra, at 17 (“[A]lcohol abuse reflects a multidetermined continuum of drinking behaviors *564whose determinants are differently weighted for different people and include culture, habits, social mores, and genes”). A sensitivity to this case-to-case variation is precisely what § 504 requires of employers and federal agencies in their assessments of the qualifications for employment or benefits of an individual with handicaps. As the medical community’s understanding of the causes of alcoholism continues to develop, § 504 requires the VA to take these new developments into account in making “sound medical judgments” about the source of a particular veteran’s alcoholism.7 Presumably, evidence concerning the circumstances surrounding a veteran’s development of alcohol dependence — including his age, home environment, and psychological health8 — always will be relevant to this assessment.

C

Finally, in asserting that its automatic association of primary alcoholism with willful misconduct is supported by medical evidence, the VA adopts, perhaps for purposes of this litigation alone, a definition of willful misconduct which is inconsistent with the definition articulated in the VA’s own regulations and practices. According to the VA, primary alcoholism is appropriately attributed to willful misconduct because medical evidence suggests that “many alcoholics are not completely helpless,” in controlling their disability. Brief for Respondents 47 (emphasis added). But a “not completely helpless” test is not the standard the VA has estab*565lished for determining whether other disabilities are incurred willfully.

The VA defines willful misconduct as “an act involving conscious wrongdoing or known prohibited action,” 38 CFR § 3. l(n) (1987), and “the intentional doing of something either with the knowledge that it is likely to result in serious injury or with a wanton and reckless disregard of its probable consequences.” VA Manual M21-1, change 239, subch. I, § 14.04a (Aug. 21, 1979).9 This definition of willful misconduct is a far cry from a “not completely helpless” standard. While some primary alcoholics may well owe their disability to willful misconduct, as delineated by the regulation, the VA has failed to demonstrate that all primary alcoholics had any awareness that their initial drinking was likely to result in serious injury. Nor, in many cases, would it be appropriate to describe one’s gradual development of alcohol dependency as evidence of “wanton and reckless disregard of [drinking’s] probable consequences.” Indeed, I wonder how one meaningfully can ascribe such intent and appreciation of long-range consequences to a 9- or 13-year-old boy who follows the lead of his adult role models in taking his first drinks.10

The awkwardness of attributing all primary alcoholism to willful misconduct is made apparent in the Administrator’s Decision No. 988, which elaborates on the meaning of the term in the context of explaining why the VA does not bar recovery for those suffering organic secondary effects of the continued use of alcohol:

*566“[The] development of the secondary condition does not meet the definition of intentional wrongdoing with knowledge or wanton disregard of its probable consequences. Secondary results are not the usual and probable effects of drinking alcohol as a beverage. By the time there is sufficient awareness of any probable deleterious consequences, the process has developed to a point where it is irreversible without professional help. At such time, the person by himself, may lack the capacity to avoid the continued use of alcohol. While it is proper to hold a person responsible for the direct and immediate results of indulgence in alcohol, it cannot be reasonably said that he expects and wills the disease and disabilities which sometimes appear as secondary effects.” Decision No. 988, p. 2 (1964) (emphasis in original).

All of this surely can be said of some primary alcoholics, whose drinking begins innocently enough and who feel only much later the effects of a dependency so disabling that it requires repeated hospitalization.

Individuals suffering from a wide range of disabilities, including heart and lung disease and diabetes, usually bear some responsibility for their conditions. And the conduct that can lead to this array of disabilities, particularly dietary and smoking habits, is certainly no less voluntary than the consumption of alcohol. Nevertheless, the VA has expressed an unwillingness to extend the definition of willful misconduct to all voluntary conduct having some relation to the development of a disability. In justifying the exclusion of secondary organic effects of alcoholism, such as cirrhosis of the liver, from the reach of the willful-misconduct presumption, the VA has explained:

“[Historically, the question of willful misconduct has never been raised in other related situations where personal habits or neglect are possible factors in the incurrence of disability. For example, the harmful effects of tobacco smoking on circulation and respiration were *567known long before tobacco was incriminated as a causative factor in the high incidence of cancer, emphysema and heart disease. Yet smoking has not been considered misconduct. It is unreasonable and illogical to apply one set of rules with respect to alcohol and a different one in a situation closely analogous.” Ibid.

In deferring to the VA’s “reasonable” determination that all primary alcoholism is attributable to willful misconduct, the Court obscures the meaning of “willful misconduct” in a similar fashion. The Court discusses the propriety of denying benefits to those who “bear some responsibility for their disabilities,” and suggests that the attribution of all primary alcoholism to willful misconduct is justified because “the consumption of alcohol is not regarded as wholly involuntary.” Ante, at 549, 550. The degree of personal responsibility for their disability attributed to alcoholics by the VA in its brief and echoed by the Court in its opinion is clearly not of the magnitude contemplated by the VA’s general definition of willful misconduct.

V

Section 504 guarantees Eugene Traynor and James P. Mc-Kelvey federal benefits absent a demonstration that they, as individuals, fail to satisfy the legitimate qualifications Congress has imposed upon receipt of those benefits. The VA has failed to demonstrate that any legislative or medical determinations justify its conclusive presumption that Mr. Traynor’s and Mr. McKelvey’s alcoholism was incurred willfully. Both cases therefore should be remanded to the VA for individualized determinations, based on “sound medical judgments” whether these men are “otherwise qualified” to receive veterans’ educational benefits beyond the 10-year period.

I dissent.