concurring.
The writer agrees with the majority that Erickson’s appeal must fail no matter how we interpret the ordinance. If Erickson’s interpretation of the ordinance is correct, then the jury was properly instructed and the breathalyzer results and the officer’s observation of her behavior provided sufficient circumstantial evidence to convict her. If the ordinance establishes presumptions against the accused, permitting the jury to find guilt based only on the breathalyzer, then the instructions were actually more favorable to Erickson than she was entitled to receive. I would agree that at the very least the ordinance establishes the “presumptions” discussed in part I of the opinion. Nevertheless, I am convinced, after reviewing the authorities, that the ordinance, properly construed, is even more favorable to the municipality than a majority of the court is prepared to hold.
The subsection under consideration has two elements.1 First, it must be established that the defendant drove a motor vehicle and, second, that within four hours of his driving a “chemical test” was administered to him resulting in a finding that there was “0.10% or more by weight of alcohol in his blood or 100 milligrams or more of alcohol per 100 milliliters of his blood, or when there is 0.10 gram or more of alcohol per 210 liters of his breath.” If the trier of fact is convinced of these two facts beyond a reasonable doubt, the defendant is guilty as charged. The jury need not determine the precise blood-alcohol level that existed at any given time while the defendant was actually operating his vehicle. A similar statute was considered • by the Delaware Superior Court in State v. Rucker, 297 A.2d 400 (Dd.Super.Ct.1972).
The defendant argues that the essence of the offense is having a blood alcohol level of 0.100, or more, at the time of the driving and not at the time of the test. Defendant says the State failed to prove that the required percentage existed at the time he was driving. This argument fails to recognize that the statute in question spells out what shall constitute proof of the offense.
*969Under the terms of the statute the trier of fact must determine whether the test results show the required percentage of alcohol in the blood. The trier of fact is not free to disregard the mandate of the statute or to question the wisdom of the General Assembly in providing that test results constitute proof of that element of the crime.
The possible variance in results between various types of tests and the possible variances in readings between tests taken while the accused was driving and those taken afterwards may be an inherent weakness of the statutory provisions. The General Assembly could have considered these possible variances when it enacted the legislation but the legislation is so worded as to preclude these factors from being considered as issues of fact.
If there had been evidence that the test was improperly administered, such evidence could cast such doubt on the result as could be considered by the trier of fact in determining whether the statutory requirements had been met. But, as indicated, evidence that the types of tests already approved by the General Assembly when properly conducted are still subject to possible variations in results, is not a matter which is here left to the trier of facts.
297 A.2d at 402-03.
Erickson’s argument fails to distinguish between “legislative” and “adjudicative” facts and “stipulative” and “descriptive” definitions. Adjudicative facts are those that must be found beyond reasonable doubt by the trier of fact before there can be a conviction. Legislative facts are those assumptions of fact, involving social, political, economic or scientific considerations, which a legislature (including a municipal assembly) makes in the course of reaching the policy decisions which it articulates in the form of statutes and ordinances. See, e.g., State v. Erickson, 574 P.2d 1, 4-5 (Alaska 1978). Thus, it is probably true that in enacting the ordinance in question, the municipal assembly made certain factual assumptions. First, that a person driving a vehicle after consuming sufficient alcohol to register a blood-alcohol rate of .10 or more presents an unacceptable risk of danger to the public. Second, assuming that no alcohol is consumed between a person’s ceasing to drive and his taking the test, that the test results will always be equal to or less than the blood-alcohol rate existing at the time of driving, i.e., that the passage of time will more often than not result in a diminished blood-alcohol rate. While both of these factual assumptions may be subject to dispute and, to a- limited extent, were disputed by expert testimony in the eviden-tiary hearing conducted in Anchorage v. Serrano, 649 P.2d 256 (Alaska App.1982), and Cooley v. Anchorage, 649 P.2d 251 (Alaska 1982),2 it has never been necessary that a legislature’s legislative factual assumptions be “probably right” to sustain a statute. It is sufficient if a reasonable legislator could believe them to be true. See State v. Erickson, 574 P.2d at 17-18; Ravin v. State, 537 P.2d 494, 505 n. 44 (Alaska 1975). A reasonable legislator could certainly believe both of the propositions of legislative fact which support enactment of AMC 9.28.020(A) and (B)(2). See People v. Schrieber, 45 Cal.App.3d 917, 119 Cal.Rptr. 812 (Cal.App.1975) (it is rea*970sonable to assume that as time elapses, alcohol will be dissipated from the body resulting in a lower breath or blood-alcohol reading).
A “descriptive definition” purports to describe the way a term is generally used in communication, i.e., in common usage; it is sometimes called a “lexical definition,” see I.M. Copi, Introduction to Logic, 120 (4th ed. 1972). A “stipulative definition” purports to describe the way in which a given speaker intends to use a term; it may or may not correspond to one or more of the descriptive definitions of the term. Id. at 118-19. Generally, a legislative body is assumed to use terms in their “descriptive” sense, i.e., in accord with common usage. See A. Dickerson, The Interpretation and Application of Statutes, at 223 (1975). However, a legislative body may utilize stipulative definitions so long as it makes its intention clear and does not create a constitutionally vague result. See 1A C.D. Sands, Sutherland Statutory Construction, §§ 27.01, 27.02 (4th ed. 1972).
In State v. Erickson, 574 P.2d 1 (Alaska 1978), the legislature defined the term “narcotic” to include “cocaine.” Erickson challenged this definition on constitutional grounds and offered substantial expert testimony to show that cocaine was not a narcotic according to the ordinary usage of the term “narcotic.” The supreme court held that the legislature had used the term narcotic in a “stipulated” way but made its intention sufficiently clear to avoid a vagueness challenge. It consequently rejected Erickson’s constitutional challenge.
In the instant case, the Anchorage Assembly “stipulatively” defined the phrase “driving while intoxicated” to include anyone who (1) drives and (2) within four hours thereafter takes a test showing a .10 blood-alcohol reading (assuming that no alcohol was consumed in the interim). Since there is no vagueness problem with this statutory definition, the ordinance is valid.
Erickson does not deny that she drove a vehicle and that within four hours thereafter a blood-alcohol test was administered resulting in a reading within the statutory guidelines, nor does she contend that the breathalyzer test or the instrument used in that test was in any way defective. Finally, and perhaps most significantly, she does not contend that shé consumed any alcohol from the time she ceased to drive until the time the test was administered.3 Under *971these circumstances, the municipality proved everything it had to prove in order to justify her conviction.
. It is possible that the ordinance has a third element: (3) that the person did not consume additional alcohol from the time he ceased to drive until the time the test was administered. This question is discussed further in n. 3 infra. See Doyle v. State, 633 P.2d 306, 311 (Alaska App.1981).
. In Cooley, we considered a number of constitutional challenges to an ordinance which read in part:
It shall be unlawful for any person to operate, drive or be in actual physical control of an automobile, motorcycle or other motor vehicle in the municipality at such time as the alcohol content of his blood, by weight, is 0.10% or greater as determined by a test of his blood, breath or urine.
Former AMC 9.28.030(A) (Repealed).
In our treatment of these issues, we assumed, though we did not decide, that under that ordinance the test results were evidence from which the trier of fact would be asked to infer a blood alcohol rate at the time of driving. The distinguishing feature which appears in the ordinance under consideration here but which did not appear in the Cooley ordinance is the language: “as determined by a chemical test within four hours of his arrest.” This language serves to shift the trier of fact’s attention from the time of driving to the time of the administration of the test. Our language in Cooley is therefore not inconsistent with the result I would reach in this case.
. The municipal assembly evidently assumed that a .10% blood-alcohol rate rendered a driver unsafe and that a person’s blood-alcohol rate decreases over time if no new alcohol is consumed. Thus, under this interpretation of the ordinance, the assembly considered the precise blood-alcohol rate at the time of driving irrelevant. Erickson did not offer any evidence that she consumed any alcohol from the time she ceased driving until the time the test was administered and the state’s evidence was that she did not consume any alcohol during this time. Therefore, it is not necessary for us to determine the legal effect under the ordinance of testimony that alcohol was consumed in the interim, i.e., whether the absence of interim drinking is an element of the offense or merely a circumstance relevant to the admissibility of the test, i.e., a foundational fact. See A.R.E. 104(b).
Erickson points to expert testimony in the Cooley case which poses the hypothetical possibility that someone could consume a great deal of alcohol, as for example swallowing a bottle of whiskey in a chug-a-lug contest, jump into a vehicle and speed home before the alcohol took effect. Consequently, Erickson con-eludes, it is not always true that a later breathalyzer reading will be lower than the reading at the earlier time when driving took place even if no more alcohol is consumed in the interim. Erickson misses the point of the ordinance. She is arguing over “legislative” not “adjudicative” facts. Thus, the legislative body could well conclude that the person in Erickson’s hypothetical creates an unacceptable risk of harm by her conduct without any compensating social utility even if she does not cause an accident. Such a person is in effect playing a form of Russian roulette with the driving public. Even if we assume Erickson’s hypothetical is possible, it does not follow that the assembly could not rationally determine that the person in the hypothetical should be punished and structure the ordinance accordingly.
In summary, the legislative body evidently wishes to prohibit those who engage in excessive drinking from driving and uses the results of the subsequently administered test to differentiate between those who present an unacceptable risk and those who do not.
Finally, under this interpretation of the ordinance the precise blood-alcohol level at the time of driving is not an element of the offense.