September 11, 1975.
 Infants — Intoxicating Liquors — Regulation — Equal Protection — Criminal Penalties. Under U.S. Const. amend. 21, which grants special authority to states over public welfare regarding intoxicating liquor, state action regulating alcoholic beverages on the basis of age and providing criminal penalties meets equal protection requirements if there is a rational relationship between it and a legitimate state purpose.
 Constitutional Law — Equal Protection — Rational Relationship — Determination. The question of whether a rational relationship exists between state action and a legitimate state purpose is one of law; there is no need to inquire into the facts of a particular case. The restrictive rules of judicial notice are not applicable to scientific studies considered by the court as a part of its legal reasoning.
 Infants — Intoxicating Liquors — Regulation — Validity. A rational basis exists between the establishment of 21 as a minimum age for the consumption of alcohol and a legitimate state purpose. Such a limitation established by the legislature is valid.
Appeal from a judgment of the Superior Court for Thurston County, No. 48784, Gerry L. Alexander, J., entered January 5, 1974. Affirmed.
Action challenging the age limitation for consumption of alcoholic beverages. The plaintiff appeals from a summary judgment in favor of the defendants.
Buzzard Glenn, P.S., by Jerome L. Buzzard, for appellant.
Slade Gorton, Attorney General, and Dennis R. Reynolds, Assistant, for respondents.
Charles Houser III, brought this action on his own behalf and that of the class of all 18- to 20-year-olds in this state, challenging the constitutionality of the legislation that established a minimum age of 21 for the consumption of alcoholic beverages. The trial court rendered summary judgment in favor of the defendant state agencies, and Houser appeals. We affirm the ruling of the trial court.
Appellant sought from the court below a declaratory judgment that the 21-year-old drinking age deprived persons between the ages of 18 and 20 of the equal protection of the laws in violation of the Fourteenth Amendment and Const. art. 1, § 12. He contended that no rational basis exists for the present statutory scheme under which Washington citizens are considered to be adults at the age of 18 for all purposes except the possession and consumption of alcohol. He supported this contention with evidence in the form of an expert's affidavit that indicated that several traditional arguments for maintaining the drinking age at 21 were without scientific support. The State countered this claim by submitting to the court's notice two technical studies which contained data supporting the statutory age discrimination, and by citing the single federal case in point, Republican College Council v. Winner, 357 F. Supp. 739 (E.D. Pa. 1973), which upheld Pennsylvania's drinking age limitation against equal protection attack. The trial court found the State's studies and the reasoning of the Republican College case adequate to uphold the drinking-age statutes, regardless of the truth of the statements in appellant's expert's affidavit. It therefore granted the State's motion for summary judgment. Appellant's appeal challenges both the trial court's judicial notice of the State's studies and its refusal to grant him a trial on the merits of the contradictory factual claims.
The evidence cited by the affidavit indicated (1) that lowering the drinking age to 18 would not lead to increased alcohol consumption, juvenile delinquency and alcoholism, and (2) that it was not true that persons between 18 and 20 are more profoundly affected by alcohol than others and less able to exercise moderation in alcohol consumption.
These studies indicated (1) that drinking and youth are both positively correlated with traffic accidents, (2) that reduction of the drinking age in some states has been followed by an increased number of accidents, and (3) that persons between 18 and 20 are more vulnerable to traffic accidents at a given blood alcohol level than those over 21.
In order to consider whether the trial court's disposition of the case was correct, we must first determine what standard of constitutional review appellant's equal protection claims properly invoked. The trial court assumed that the discrimination between 18- to 20-year-olds and those 21 and over with regard to drinking should be upheld if a "rational relationship" was found to exist between it and a legitimate State purpose. We agree.
 Certainly the "strict scrutiny" applicable to suspect classifications and those which burden fundamental rights would not be appropriate here: age discriminations are not inherently "suspect" ( Oregon v. Mitchell, 400 U.S. 112, 295 n. 14, 27 L.Ed.2d 272, 91 S.Ct. 260 (1970) (Stewart, J., concurring); United States v. Duncan, 456 F.2d 1401 (9th Cir. 1972), and the "right" to consume alcohol is far from "fundamental" ( Randles v. State Liquor Control Bd., 33 Wn.2d 688, 694, 206 P.2d 1209, 9 A.L.R.2d 846 (1949). The criminal penalties that drinking laws carry impact on the right to be free from imprisonment and would, in other contexts, make them subject to some more demanding scrutiny than "minimum rationality" review. State v. Martinez, 85 Wn.2d 671, 538 P.2d 521 (1975); cf. Jackson v. Indiana, 406 U.S. 715, 723-31, 32 L.Ed.2d 435, 92 S.Ct. 1845 (1972). However, the Twenty-First Amendment, which confers on the states "something more than the normal state authority over public health, welfare, and morals" ( California v. LaRue, 409 U.S. 109, 114, 34 L.Ed.2d 342, 93 S.Ct. 390 (1972)), modifies the demands of the Fourteenth as to regulation of alcohol use and distribution and makes state action in this area subject only to the most lenient equal protection review, the "rational relationship" standard the trial court applied.
Persons between the ages of 18 and 21 do not comprise a "discreet and insular minority" ( United States v. Carolene Prods. Co., 304 U.S. 144, 152 n. 4, 82 L.Ed. 1234, 58 S.Ct. 778 (1938)) for which unusual constitutional protection is appropriate, especially in light of the fact that they have now been given the power to protect and enforce their class interests through the electoral process. U.S. Const. amend. 26. Further, we recognize that judicial scrutiny of age classifications should not be overly demanding, if only because distinctions based on age are inevitably rough and arbitrary. See State v. Koome, 84 Wn.2d 901, 911-12, 530 P.2d 260 (1975).
When a legal distinction is determined . . . between . . . childhood and maturity . . . a point has to be fixed or a line has to be drawn . . . to mark where the change takes place. Looked at by itself without regard to the necessity behind it the line or point seems arbitrary. It might as well or nearly as well be a little more to one side or the other. But when it is seen that a line or point there must be, and that there is no mathematical or logical way of fixing it precisely, the decision of the legislature must be accepted unless we can say that it is very wide of any reasonable mark.
Louisville Gas Elec. Co. v. Coleman, 277 U.S. 32, 41, 72 L.Ed. 770, 48 S.Ct. 423 (1928) (Holmes, J., dissenting).
Not only is the freedom to consume alcohol not "explicitly or implicitly guaranteed by the Constitution" ( San Antonio Independent School Dist. v. Rodriguez, 411 U.S. 1, 33, 36 L.Ed.2d 16, 93 S.Ct. 1278 (1973)), it is specifically made subject to restriction and regulation by the Twenty-First Amendment. See Republican College Council v. Winner, 357 F. Supp. 739, 740 (E.D. Pa. 1973).
 To ascertain whether a rational relationship existed between the 21-year-old drinking age and a legitimate state purpose, the trial court took judicial notice of the studies submitted to it by the State. Appellant contends these studies were not judicially noticeable because the facts they contain were not "well established and authoritatively settled." This argument misconceives the function the court was performing in ruling on the constitutional issue before it. The State's summary judgment motion required the court to inquire not into the facts of the particular case at bar but into the general relationship between attainment of the age of 21 and the effect of alcohol consumption. The question it presented was essentially one of law, not fact: whether there was a "rational relationship" between the statutory distinction and the state purposes it was alleged to serve. The finding that it was rational to believe that the discrimination did correspond to a permissible state objective was a step in the court's legal reasoning, not a conclusion regarding the factual background of the particular dispute before it.
A court "may ascertain as it sees fit any fact that is merely a ground for laying down a rule of law . . ." Chastleton Corp. v. Sinclair, 264 U.S. 543, 548, 68 L.Ed. 841, 44 S.Ct. 405 (1924). The restrictive rules governing judicial notice are not applicable to factual findings that simply supply premises in the process of legal reasoning. See Fed.R. Evidence 201 (a), Advisory Committee's note; 2 K. Davis, Administrative Law Treatise §§ 15.02-15.06 (1958); C. McCormick, Handbook of the Law of Evidence 768-69 (E. Cleary 2d ed. 1972). In interpreting and developing the constitution and laws, courts cannot operate in a vacuum. In order to determine whether there is a "rational relationship" between a statutory classification and an objective said to justify it, a court must look beyond the case reports and statute books into a world that is rich with probability and conjecture and almost devoid of settled certainty. It must make the best assessment it can from the best information it can obtain. Reputable scientific studies are one source of such information, increasingly utilized by courts in constitutional decision making. The trial court thus did not err in noticing the studies submitted to it in this case.  In light of the information it had before it, the trial court was clearly correct in its grant of summary judgment for the State. Regardless of the arguments and allegations made by appellant, there was no "genuine issue as to any material fact" (CR 56 (c)) before the court. The sole material factual question on the summary judgment motion was whether there was a rational basis for the 21-year-old drinking age. The studies the State submitted established that basis. They showed, among other things, that there is reputable scientific evidence indicating that the impact of alcohol on driving ability is more profound in persons between the ages of 18 and 20 than in persons over 21. Appellant challenged this conclusion, but did not claim that the data on which it was based were wholly unscientific, fraudulent, or otherwise so unreliable that it would be irrational to give them credence. The trial court therefore correctly ruled that a "`state of facts reasonably [could] be conceived that would sustain the classification'" ( Sonitrol Northwest, Inc. v. Seattle, 84 Wn.2d 588, 590, 528 P.2d 474 (1974)): the authors of the studies conceived such a state of facts and provided sufficient data in support of it to make that conception reasonable. Nothing appellant or his expert could have added would have altered that.
See, e.g., Furman v. Georgia, 408 U.S. 238, 33 L.Ed.2d 346, 92 S.Ct. 2726 (1972) (studies of various aspects of capital punishment); Brown v. Board of Educ., 347 U.S. 483, 494 n. 11, 98 L.Ed. 873, 74 S.Ct. 686, 38 A.L.R.2d 1180 (1954) (studies regarding effects of school segregation); State v. Koome, 84 Wn.2d 901, 905, 530 P.2d 260 (1975) (studies showing dangers of childbearing by minors); DeFunis v. Odegaard, 82 Wn.2d 11, 33, 507 P.2d 1169 (1973) (facts showing minority underrepresentation in legal profession).
It is noteworthy, though not crucial, that this point (the special impact of alcohol consumption on teenage driving) was stressed by opponents of 1973 Referendum 36, a measure which would have lowered the drinking age along with the age of majority for other purposes. See Official Voters Pamphlet, 1973 General Election, November 6, 1973, at pages 8-9. A slightly different argument, that a lowered drinking age would result in an increased number of accidents simply because of the increased number of drinkers it would produce, was also made in the Referendum 36 campaign. This claim is supported by the studies submitted by the State below, but it alone cannot justify a legal distinction between persons under 21 and those over that age. The fact that more drinking causes more accidents may warrant, even demand, strong State action to control its use by all persons, but not State action aimed at only one class indistinguishable from the population at large. Only the evidence that alcohol more profoundly affects the driving of persons under 21 supplies the correlation between age, accidents and alcohol necessary to uphold the classification on this basis.
Appellant's expert did contend in his affidavit that "there is not one shred of evidence that in its action on the bodily function and performance [alcohol] is more severe in those who are 18, than in older individuals." He did not, however, explain away the State's data showing an increased impairment of driving ability for persons under 21. Similarly, though the affidavit cited evidence that the legal status of 18- to 21-year-old drinking would not affect its frequency, it did not impugn the findings in several studies that, to the contrary, lowering of the drinking age in several states has been followed by an increase in the number of alcohol-related accidents involving 18- to 20-year-old drivers. These data in turn suggest that the change in the law did indeed result in an increased rate of alcohol consumption within that group.
The decision of the trial court is affirmed.
STAFFORD, C.J., and FINLEY, ROSELLINI, HUNTER, HAMILTON, WRIGHT, BRACHTENBACH, and HOROWITZ, JJ., concur.