OPINION

I. INTRODUCTION

The rape victim of a parolee sued the State for negligent parole supervision. The State sought summary judgment on the grounds that (1) it owes no duty of care in supervising parolees; (2) it is immune from suit; and (3) it is entitled to judgment as a matter of law on the issue of causation. We conclude that the State owes a duty of care to a parolee’s victim, but vacate the denial of summary judgment and remand to the superior court to address the questions of discretionary function immunity and causation in light of the principles set forth in State of Alaska, Department of Corrections v. Cowles,1

This case also presents the question whether the cap on noneconomic damages in AS 09.17.010 is constitutional and, if so, whether multiple sexual penetrations should be treated as a single injury for purposes of the cap. We hold that the cap on noneco-nomic damages in AS 09.17.010 does not constitute a violation of equal protection under the Alaska Constitution. We also hold that each sexual assault constitutes a separate incident under the statute.

II. FACTS AND PROCEEDINGS

A. Factual History

In 1987 Luke Carter committed a violent rape. Carter was convicted of sexual assault in the first degree and sentenced to a fifteen-year presumptive term. He was released on mandatory parole on July 29,1997.

Joseph Murphy, Carter’s institutional parole officer, met with him before his release from prison to discuss his plans. Carter initially planned to settle near his home village of Noorvik. Carter was informed that he could not be released to the Kotzebue area because his parole conditions required him to live in a place where sex offender treatment was available. Carter chose Anchorage. Carter and Murphy did not develop a detailed release plan; Carter’s notification of release to supervision did not list an address or employer for Carter in Anchorage.

On July 29,1997, Carter was released from prison subject to twenty-seven standard and special conditions of mandatory parole. Pri- or to his release, Carter was told that he was required to report to his field parole officer *376the day after his release and to register as a sex offender within seven days.

Field Parole Officer Linda Heyward was assigned to Carter’s case and reviewed his field file a few days before his release. On the morning of July 30, Carter reported as required for the intake interview, but Hey-ward was not present. The office clerk asked Carter to complete a monthly report form and instructed him to return the next day for the interview. On the form, Carter listed no residence: he provided only a Noor-vik mailing address and an unidentified Anchorage telephone number. Carter failed to report on July 31. Carter also failed to register as a sex offender within seven days of his release. Heyward took no immediate action to report these violations or pursue revocation.

On August 11 a state trooper contacted Heyward regarding a report that Carter might have stolen a key to a hobby locker. Heyward told the officer that she had not been in contact with Carter for the past several weeks and gave him some Anchorage telephone numbers at which Carter might be reached. The officer was unable to contact Carter and placed a “locate” for Carter on the state police computer system. Also on August 11 Heyward called the phone number listed on Carter’s form, but Carter was not there. Heyward noted in her log that she would give Carter two weeks to remedy his violations before taking any further action.

Two days later, on August 13, Heyward received Carter’s complete institutional file. After reviewing it, Heyward copied medical testimony into her log regarding Carter’s dangerous personality. She attempted to contact Carter’s cousin again, as well as Carter’s father, without success. That same day, Heyward completed a parole violation report which described Carter’s failure to report and to register as a sex offender and stated her opinion that Carter was a danger to the public and likely to flee. This report led to the immediate issuance of an arrest warrant. Also on August 13 Heyward notified the Kot-zebue probation office that Carter might be headed in their direction and promised to fax them a copy of the arrest warrant. Once the arrest warrant was issued, Heyward did not take any further action to locate or arrest Carter.

At approximately 8:30 a.m. on August 27, 1997, Carter attacked C.J. while she was jogging along Anchorage’s Tony Knowles Coastal Trail.2 Carter grabbed her by the neck and dragged her into the bushes. The court of appeals described the incident in the criminal action:

He said that he had a weapon and would kill her if she screamed. He pushed her sweatshirt and sports bra over her head. He penetrated her vagina with his fingers. He then removed her shorts and underwear, and he performed cunnilingus. He penetrated her vagina with his penis. These three distinct penetrations led to the three counts of first-degree sexual assault.[ 3]

Carter was charged and convicted by a jury of three counts of first-degree sexual assault. At sentencing, Carter’s attorney argued that even though Carter had been convicted of three sexual assaults he should only be subjected to a single sentence because there “was no break in the event; it was a continuous episode.” The State sought thirty years on each count with five years suspended, imposed consecutively. The judge rejected the defense’s argument, finding it “significant that this victim was assaulted in different ways.” As a result, Carter received three partially consecutive sentences, totaling forty-five years.4

B. Procedural History

On July 13, 1999, C.J. filed an action against the State for negligence. On February 14, 2000, the State filed a third-party complaint against Carter for equitable apportionment of fault under AS 09.17.080.

On September 15, 2003, the State moved for summary judgment. In its accompanying *377memorandum, the State argued that (1) our decision in Division of Corrections v. Neakok5 should be overruled and therefore that it owed no duty of care to C.J.; (2) discretionary function immunity barred plaintiffs claims; and (3) DOC’s conduct did not, as a matter of law, cause C.J.’s injuries.

On September 30, 2003, C.J. filed a motion asking the court to find that (1) the caps on noneconomic damages in AS 09.17.010 are unconstitutional, and, in the alternative, that (2) C.J. is entitled to recover damages up to the $400,000 cap “for each of the three separate incidents of rape she suffered.” The same day, the State filed a motion asking the court to limit any potential recovery by C.J. to one award, limited to the $400,000 cap on noneconomic damages, because the repeated sexual assaults occurred in a “continuous criminal episode.”

On November 17, 2003, Superior Court Judge Sharon L. Gleason held oral argument on the motions regarding summary judgment and the damages cap. On November 18, 2003, Judge Gleason denied the State’s motion for summary judgment. Judge Gleason also issued an order finding that C.J.’s injuries “shall be treated as a single injury for purposes of AS 09.17.010 and the non-economic damages may not exceed $400,000.”

On January 27, 2004, we granted both the State’s petition for review of the denial of summary judgment and C.J.’s petition for review of the superior court’s ruling on the damages cap.

III. STANDARD OF REVIEW

“We review denials of summary judgment motions de novo to determine whether there are genuine issues of material fact and whether the moving party is entitled to judgment as a matter of law, viewing all facts in the light most favorable to the non-movant.”6 Issues relating to a statute’s constitutionality and interpretation are also legal questions subject to de novo review.7

IV. DISCUSSION

A. Duty, Immunity, and Causation

The State argues that it is entitled to summary judgment because it owed no duty to protect C.J. from Carter. It urges us to overrule our contrary holding in Neakok.8 The State also argues that it is immune from suit under AS 09.50.250(1). The State raised similar arguments in Cowles,9 a case that also involves a claim of negligence against the State based on crimes committed by a parolee under its supervision. Because these cases present related issues, we announce our decision in these two cases together.

As we discuss in our opinion in Cowles, we decline to overrule our holding in Neakok that the State has a duty to exercise due care in supervising parolees. We also set forth in Cowles the boundaries of discretionary function immunity as it relates to decisions involving parole supervision. Our discussion in Cowles of when the State is immune from liability for a parole officer’s failure to seek parole revocation is particularly relevant since C.J. alleges that Heyward should have pursued revocation when Carter failed to report on July 31 and failed to register as a sex offender. As we explained in Cowles, a parole officer’s decision whether or not to pursue revocation in response to a known parole violation is immune from liability unless DOC policy explicitly requires the officer to seek revocation under these circumstances.10 On remand, the superior court should determine whether each of DOC’s allegedly negligent decisions is a discretionary function based on the principles *378outlined in Cowles and our other decisions concerning discretionary function immunity.11

The State also argues that it is entitled to summary judgment because no reasonable jury could find that DOC’s alleged negligence caused Carter’s assault. Our decision requires the superior court to reexamine whether the State is entitled to discretionary function immunity for some of its allegedly negligent acts. That reexamination may affect the superior court’s causation analysis, for C.J. may only rely on non-immune acts to establish causation. We decline to decide in the first instance whether the State is entitled to judgment as a matter of law on the issue of causation and remand the case to the superior court for a ruling on this question.

B. The Cap on Noneconomic Damages in AS 09.17.010 Does Not Violate C.J.’s Right to Equal Protection Under the Alaska Constitution.

C.J. argues that the noneconomic damages cap in AS 09.17.010 violates the equal protection clause of the Alaska Constitution12 because two classes of successful tort plaintiffs are treated differently: (1) those suffering serious injuries resulting in noneconomic damages that exceed the cap, and (2) those who receive “full compensation for their loss” because their damages are below the cap.

We analyze equal protection claims under a “sliding scale approach”13 which “placets] a greater or lesser burden on the state to justify a classification depending on the importance of the individual right involved.” 14 If the right impaired by the challenged legislation is not very important, the State need only show that its objectives are “legitimate” and that the legislation bears a “substantial relationship” to its purpose.15 At the other end of the continuum, legislation that impairs one of the most important individual interests will be upheld only if it furthers the State’s “compelling interest” and if it is the “least restrictive means” available to achieve the State’s objective.16

We considered a facial equal protection challenge to the noneconomic damages caps in AS 09.17.010 in Evans v. State.17 An equally divided court affirmed the superior court’s holding that the caps were constitutional.18 The plurality opinion in Evans concluded that the interest impaired by the caps is merely economic and therefore justifies only minimum scrutiny.19 The plurality next determined that the stated purposes of the legislation, which include controlling liability insurance premiums, are legitimate.20 Finally, the plurality found that the caps bear a “substantial relationship” to the legislature’s *379objective of lowering insurance rates.21

C.J. urges us to reject the reasoning of the plurality opinion in Evans. She also argues that the particular circumstances of her case enhance the cap’s unfairness and support a finding that the cap is unconstitutional as applied to her. We adopt the equal protection analysis of the plurality opinion in Evans, and respond to each of C.J.’s arguments in turn.

C.J. first makes the same argument as the Evans dissent that the interest implicated in this case is not merely an economic interest in money damages but the more fundamental right of equal access to the courts.22 But we have consistently held that restrictions on the types or amounts of damages that a plaintiff can pursue in court impair economic interests only and are therefore subject to minimum scrutiny review.23 As we noted in Evans, the right of access to the courts “is impaired only by state action that actually limits or blocks access to the courts,” not by regulations that “simply limit plaintiffs recovery in civil court.”24 C.J. also contends that the cap “creates a practical and effective barrier to filing suit” because plaintiffs with little or no economic loss will have insufficient incentive to file suit and will have difficulty finding a lawyer to take the case. But in every case where the cap is applied, the plaintiffs recovery for noneconomic damages will be at least $400,000 plus Civil Rule 82 attorney’s fees, guaranteeing sufficient incentive to exercise the right to access the courts.25

C.J. repeats the Evans dissent’s argument that Alaska’s sliding-scale approach permits economic interests to qualify for close scrutiny in some circumstances.26 But the cases cited by C.J. and the Evans dissent for the proposition that economic interests can qualify for heightened scrutiny involve the “right to engage in an economic endeavor,” which is a special and quite different kind of economic interest from the one at issue in this case.27 As we explained in Wilkerson v. State, the right to engage in an economic endeavor is important because it involves a source of sustaining income that individuals depend on to supply “the basic necessities of life.”28 In Wilkerson, we explicitly contrasted the important interest in engaging in an economic endeavor with the interest in other payments, such as the permanent fund dividend or a foster parent stipend, that are not “a source of sustaining income” and are therefore mere economic interests entitled to mini*380mum scrutiny.29 Because a noneconomic damages award is, by definition, not a source of sustaining income, a replacement for the ability to earn income, or a source of money to provide for basic needs such as past or future medical care,30 the economic interest impaired by the cap is subject to minimal protection.

C.J. also contends that the damages cap cannot be sustained even under minimum scrutiny analysis because the State has failed to show the required fit between the cap on damages and the State’s objective of lowering insurance premiums.31 In making this argument, C.J. does not question whether the damages cap will actually result in lower insurance premiums. Rather, her argument is that it is irrational to single out the most severely injured tort victims to pay for the reduction in premiums. The dissent in Evans made the same argument.32

To begin, we agree that the legislature may not use a suspect or arbitrary method in identifying the group that will bear the brunt of an otherwise legitimate policy.33 But because the damages cap imposes only economic burdens, and allocates these burdens using criteria that are not presumptively suspect, our scrutiny of the underlying rationale is minimal. Minimum scrutiny has traditionally meant two things in assessing the “means” chosen to effectuate the statutory goal. First, although the scheme should be “substantially” related to some legitimate purpose, we have not required “a perfect fit between a legislative classification and the government objective it is intended to further.”34 For example, we have upheld school funding schemes that were not the “most protective of tax equality,” but that were “close enough.”35 We have also upheld foster parent application restrictions even though we conceded the restrictions “do[ ] not have to be meaningful in all cases,”36 and have emphasized that policies reviewed under minimum scrutiny are not unconstitutional simply because a litigant is able to propose a regulation that would further the legislative goal in a more rational manner.37 The second principle is that although the rationale underlying the legislation should be logically plausible, there is no requirement that it be proved in court. For example, in considering a challenge to a new limit on prejudgment interest, we upheld the limit because we agreed that a policy of reducing malpractice insurance “can reasonably be thought to be furthered” by the scheme.38 Similarly, we accepted, without evidence, the assertion that legislation narrowing employers’ liability for stress-related mental injuries would save employers money by eliminating unusually susceptible claimants and minimizing fraud and abuse in claims for stress-related mental injuries.39

*381Viewed in light of these precedents, the damages cap satisfies the means-end fit requirement. First, the legislature appears to have viewed large noneconomic damage awards as susceptible to over-estimates of the dollar value of a victim’s noneconomic loss.40 Whether or not we agree with this conclusion, we think it is one that the legislature was entitled to make. Second, the legislature could reasonably have concluded that any alternative method of lowering insurance costs would have been less fair than a cap on noneconomic damages. The most obvious alternative would have been to reduce all tort awards by a fixed percentage. But such a policy would create its own inequities. A significant across-the-board reduction in awards might have the effect of destroying the viability of smaller claims, thereby burdening victims’ ability to seek redress to a much greater degree than would result from a cap applicable only to high-damage cases, which are viable almost by definition. And the legislature could determine that an across-the-board cut would not necessarily reduce insurance costs as effectively as a cap because very large, uncapped awards may have a disproportionate effect on insurance premiums. The fit between the noneconomic damages cap and reducing insurance premiums may not be perfect, but it satisfies minimum scrutiny.

C.J. argues that the “arbitrariness and harsh effect” of the damages caps is “exacerbated” when applied to her because (1) as a rape victim, her injury involves a “disproportionate amount of pain and suffering compared to monetary loss” and (2) as a low-wage earner her “recovery will depend almost entirely on noneconomic damages.”41 We are sympathetic to C.J’s particular circumstances. But while limiting the noneco-nomic damages for such a grievous injury may seem harsh, we have held that “under a minimum scrutiny analysis, we do not determine if a regulation is perfectly fair to every individual to whom it is applied.”42 Rather, “we must decide only if the regulation bears a fair and substantial relationship to a legitimate government objective.”43 As we have explained, the damages cap satisfies this standard.

A final consideration is that the arguments advanced by C.J. and the Evans dissent would have far-reaching implications that are inconsistent with established law. Legislation frequently reduces or eliminates certain tort remedies, and often the rationale for making a subgroup of injured people pay the price of legitimate legislative goals could not survive close scrutiny. We have already mentioned some precedents that illustrate this point, but some more general examples also come to mind. Workers’ compensation statutes base damages entirely on wages, essentially eliminating all noneconomic damages.44 At oral argument, C.J. responded to this example by noting that while a damages cap is an unmitigated detriment to tort victims, workers’ compensation has given workers a streamlined, no-fault system in return for the lower recovery. But C.J.’s characterization works only at a high level of generality. Workers whose wages are low, who have been the victims of blatantly negligent conduct, or who suffer exceptional noneconomic injuries bear the brunt of a system that may benefit their co-workers or employers but certainly does not benefit them. Similarly, where someone is injured by the negligence of a state official, sovereign immunity and official immunity may bar recovery entirely.45 This immunity exists to promote vigorous *382government, but it involves no other benefit to tort victims generally, and to the victims of government torts it must seem arbitrary that they bear the entire cost. Going further back, historical developments such as strict products liability have greatly expanded the tort liability of certain actors, arguably “singling” them out to pay for social problems in ways that might have seemed irrational or illegitimate at the time.46 But tort law has not been so “constitutionalized” as to prevent these developments, and society has benefit-ted from the experimentation and debate that have resulted.47

Notwithstanding the foregoing, we appreciate that there will be severely injured persons who are under-compensated as a result of this legislation, and we are under no illusion that this result will seem fair to them. In deciding that the cap is constitutional, we express no opinion about whether damages caps are good policy.48 As we noted in Evans, “[i]t is not a court’s role to decide whether a particular statute or ordinance is a wise one; the choice between competing notions of public policy is to be made by elected representatives of the people.”49 Because the State has met its burden under minimum scrutiny of showing that the damages caps are substantially related to the legitimate interest of reducing insurance premiums, we hold that the cap does not violate C. J.’s right to equal protection.

C. CJ.’s Damages Should Not Be Limited to a Single Cap Amount Under AS 09.17.010(d).

C.J. argues that the superior court erred in limiting her potential damages to the single cap amount of $400,000. She contends that if the damages cap is applied to her she is entitled to recover the cap amount for each of the three types of sexual penetrations that formed the basis of Carter’s three convictions. The State argues that the superior court correctly applied a single cap because C.J.’s injuries occurred as part of a single criminal episode.

Alaska Statute 09.17.010(b) imposes a cap on claims “arising out of a single injury or death,” and AS 09.17.010(d) provides that “[m]ultiple injuries sustained by one person as a result of a single incident shall be treated as a single injury for purposes of this section.” In interpreting a statute, we look “to the meaning of the language, the legislative history, and the purpose of the statute in question.”50 C.J. argues that each type of penetration by Carter resulted in a separate “incident” under the statute. We agree.

*383The State relies on the use of the word “incident” in criminal eases that do not interpret AS 09.17.010 to argue that C.J.’s claims arise out of a “single incident” under the statute. But a closer look at the statute and its purposes reveals each penetration constitutes a separate “incident.” Alaska Statute 09.17.010(d) is intended to limit recovery for a single tortious act that causes multiple injuries.51 Where a car accident is caused by tortious behavior, for example, the statute seeks to prevent the accident victim who suffers multiple injuries from recovering more than one cap amount. But a cap on damages resulting from “a single incident” does not evince a legislative intent to absolve a tortfeasor of liability for committing multiple, separate tortious acts that result in distinct compensable injuries simply because those acts occur in a compressed period of time. Such a rule would allow intentional tortfeasors to reduce their liability by committing torts in rapid succession — undermining the compensatory function of tort law. Eliminating liability for distinct tortious acts that cause distinct injuries would run counter to the stated purpose of the statute to decrease the cost of litigation “without diminishing the protection of innocent Alaskans’ rights to reasonable, but not excessive, compensation for tortious injuries caused by others.” 52

The Indiana Court of Appeals reached a similar conclusion in a case involving a statute capping damages for each “occurrence of [medical] malpractice.”53 The court interpreted “occurrence” to mean each negligent act and its resulting injury,54 and held that where a doctor committed two separate acts of negligence resulting in two separate injuries in the course of a single medical procedure, the plaintiff was entitled to the equivalent of two cap amounts.55 The court noted that to impose a single cap on two distinct acts resulting in two distinct injuries “would undermine fundamental principles of tort law.”56

Here, Carter committed three distinct acts, each an intentional tort, each causing a separate injury. In applying AS 09.17.010 to this case, we look to how the legislature and the courts have defined the crime of sexual assault57 Our criminal law recognizes that “different forms of sexual penetration constitute different forms of indignity and violation ... meriting] separate punishment,”58 and Carter was found to have committed three separate criminal acts. Limiting C.J. to one cap amount in this ease would deny her any recovery for her injuries as a result of two of the three types of penetration, a result contrary to the statute’s purpose of protecting a plaintiffs right to “reasonable, but not excessive, compensation” for tortious injuries.59 The State argues that the number of “incidents” under AS 09.17.010 should not be tied to the number of convictions, pointing out *384that the disposition of a criminal case may depend on tactical decisions made by the prosecutor. But our analysis relies not on the number of Carter’s convictions, but on the fact that our criminal law treats different forms of sexual penetration as separate criminal acts.

The multiple sexual penetrations in this case are analogous to the multiple sexual violations that occurred in Kodiak Island Borough v. Roe.60 We held in Kodiak Island that each occurrence of rape caused a separate injury and therefore the plaintiff was entitled to damages for each assault.61 Although the sexual assaults in this ease occurred within a more compressed time frame, each sexual assault constitutes a separate incident for purposes of the cap.

Moreover, it would be anomalous to conclude one damages cap applies because the rape was a single continuous episode when the State successfully defeated precisely that theory in the criminal case and obtained separate sentences for each penetration.62 At sentencing, Carter’s attorney argued that even though Carter had been convicted of three sexual assaults he should only be subjected to a single sentence because there “was no break in the event, it was a continuous episode.” The State opposed this argument, seeking thirty years on each count imposed consecutively. The judge rejected the defense’s argument for a single sentence, finding that it was “significant that this victim was assaulted in different ways.” As a result, Carter received three partially consecutive sentences totaling forty-five years. Given that each type of penetration was found to constitute a separate criminal act meriting separate punishment, it would be unjust to conclude that the victim of these separate violations is limited to a single cap amount.

We therefore hold that if C.J. prevails in her civil suit, she is entitled to recover up to the cap amount for each of the three types of sexual penetration.

V. CONCLUSION

We hereby REVERSE the superior court’s order that C.J.’s potential noneco-nomic damages be capped at $400,000 under AS 09.17.010. We VACATE the superior court’s order denying the State’s motion for summary judgment. We REMAND for further proceedings consistent with this opinion.

BRYNER, C.J., and CARPENETI, J., concurs and dissents.

MATTHEWS, Justice, dissents in part.

concurring and dissenting.

I write separately to address two points.

First, although I agree with the court’s decision, I rely on a somewhat different rationale. In today’s companion case, State, Department of Corrections v. Cowles,1 we construed Neakok2 to impose an actionable duty on parole officers only when the officer knew or reasonably should have known of a focused danger to an identifiable victim or class of victims — a present danger focusing on something “more than simply members of the general public.”3 As I indicated in my concurring opinion in Cowles, I would hesitate to conclude that the population of Anchorage qualifies as an identifiable class of victims under this standard.4

*385Because I doubt that the current record in this case establishes the focused danger needed to trigger an actionable Neakok duty, it seems possible that C.J. might ultimately be unable to meet the Neakok duty standard. I nonetheless believe that the record does raise a genuine issue of material fact as to whether Carter posed an obvious and present danger of committing a sexual assault. And since the state did not assert as a basis for summary judgment the absence of evidence showing that the danger was sufficiently focused — simply arguing, in relevant part, that Neakok was bad law and should be overruled — C.J. was never called upon to develop and offer evidence that might have established a particularized risk to an identifiable class of victim.

Given the state’s failure to demonstrate that it was entitled to summary judgment on the specific ground it asserted below, I agree that the superior court properly declined to grant the state’s motion for complete summary judgment; I further agree that the superior court’s ruling on the issue of summary judgment must be reconsidered on remand in light of our opinions here and in Cowles. But in addition to other questions today’s opinion identifies for consideration on remand, I would include the question whether C.J. can meet Neakok’s focused-danger requirement. I concur in Part IV.A. of today’s opinion on this basis.

Second, as to the opinion’s ruling on the issue of capping noneconomic damages, I adhere to my dissenting opinion in Evans v. State5 and thus dissent from Part IV.B. of the opinion.

I join the opinion in all other respects.

concurring and dissenting.

I agree with the court’s opinion in all respects except Part IV.B. For the reasons expressed in Justice Bryner’s dissent in Evans ex rel. Kutch v. State,1 in which I joined,2 I would conclude that AS 09.17.010, in capping noneconomic damages, violates the Alaska Constitution’s guarantee of equal protection,

dissenting in part.

In my dissenting opinion in State of Alaska, Department of Corrections v. Cowles, I concluded that tort liability should not be imposed on parole officers or the State for negligent supervision of parolees.1 The main premise of the case holding to the contrary, Division of Corrections v. Neakok-that parole officers have a substantial ability to control the conduct of parolees2 — has been shown in subsequent years to be false and therefore Neakok should be overruled. I would thus reverse the superior court’s denial of the State’s motion for summary judgment concerning the duty issue and remand for an entry of judgment in favor of the State.

That result would moot the additional questions addressed in today’s opinion as to whether the cap on noneconomic damages in AS 09.17.010 is constitutional and whether the rape victim in this case is entitled to a separate cap for each penetration or whether the single-incident limitation for multiple injuries should apply. But since these issues are not mooted, I address them as well. I agree with the opinion of the court that the damages cap is constitutional for the reasons expressed by the court. But I do not agree that the victim in this ease is entitled to multiple caps. In my view, the rape was a “single incident” resulting in multiple injuries and thus under AS 09.17.010(d) the resulting injuries are subject to one cap. I develop this point in the paragraphs that follow.

Alaska Statute 09.17.010 caps noneconomic losses at the greater of $400,000 or the injured person’s life expectancy multiplied by $8,000, except in cases of severe permanent physical impairment or severe disfigurement. The cap is triggered by “a single injury or *386death,” but multiple injuries sustained by one person as a result of a single incident are counted as a single injury. Such injuries are thus subject to a single cap rather than multiple caps.3 The question in this case is, given that each penetration inflicted on the victim of the rape in this case is a separate injury, whether each injury was “sustained ... as a result of a single incident” within the meaning of AS 09.17.010(d).

The term “single incident” should be interpreted in a manner that is consistent with its common usage.4 In common usage, a rape or sexual assault occurring in a brief period of time is referred to as an “incident” without regard to the details of the assault. This usage is well illustrated by the opinion of the court of appeals in Yearly v. State.5 The defendant there had raped a woman at a supermarket.6 In the process he had engaged in several types of penetrations.7 The court referred to this assault as “the Carrs Supermarket incident.”8 Similarly, the same defendant had engaged in another mul-ti-crime sexual assault near a bike path at Goose Lake.9 The court of appeals described this assault as “the Goose Lake incident.”10

The same usage is reflected in another single incident, multiple-penetration case of the court of appeals, Erickson v. State.11 There the court stated:

In the present case, the jury found that Erickson had engaged in four distinct types of sexual penetration with the victim. To paraphrase Dunlop, “[wjhen several [distinct types of sexual penetration] occur in the course of a single incident, the offense prohibited by the statute has been violated several times over.”[12]

In Kodiak Island Borough v. Roe a mentally disabled resident of a residential facility was sexually assaulted by two employees of the facility.13 There were four separate assaults, two by each employee, each on different days and in different locations in the facility.14 We referred to each event as an “incident of assault.”15 We employed the same usage in referring to a Montana case where there had been three rapes of a patient on three separate occasions, describing *387that ease as “holding each incident of rape was a separate wrongful act.”16

The usage reflected by these eases applies here. Just as the multi-crime bike path sexual assault in Yearty was described as an “incident,” and just as the multi-penetration assault in Erickson was described as a “single incident,” the rape that took place in this ease is most readily described in the same way.

This interpretation is also consistent with the purpose of the damage cap statute. The statute is meant to limit damage awards in order to lower, or at least control, insurance premiums.17 This purpose is furthered by employing the common-usage meaning reflected in the above cases because the resulting noneconomic injuries suffered by a victim are limited to a single cap. On the other hand, subdividing a sexual assault into discrete penetrations runs counter to the purpose of the statute by effectively taking the damage cap out of play.

Apart from the general usage of the word “incident” and the purpose of the statute, how else could one determine whether a series of acts consist of a single incident or two or more separate incidents? A sensible approach is to use the same test used to determine whether a set of facts constitutes a single transaction or occurrence, or more than one. This is the test that we use when determining whether a counterclaim is compulsory and thus must be asserted in response to a complaint, and when determining whether a cause of action has been impermis-sibly split. “Incident” and “occurrence” are synonyms,18 and our endeavor here is much like what we try to do when we use the same transaction or occurrence test: determine what combinations of facts should be tried together, if they are tried at all.

The “same transaction” or “occurrence” test consists of a list of factors that are useful in determining whether various claims arise from the same transaction or occurrence. These factors “include whether the facts are related in time, space, origin, or motivation; whether they form a convenient trial unit; and whether their treatment as a unit conforms to the parties’ expectations or business understanding or usage.”19

These factors indicate that the rape that occurred in this case is one transaction or occurrence rather than more than one. Most importantly the facts here are closely related in time, space, origin, and motivation. Further, they would form a convenient trial unit. As to the parties’ expectations or usage, I think it would be unprecedented to conduct separate trials for each type of penetration that took place in the course of a sexual assault. So these factors also point to the conclusion that what we have here is a single transaction or occurrence. Based on all these factors, it seems clear that a victim of a multiple-penetration rape could not try a civil case based on a complaint limited to only one penetration and then subsequently bring a new case based on a second penetration. Doing so would run afoul of the rule prohibiting splitting a cause of action. It follows that each penetration is part of the same transaction or occurrence and, by analogy, also a part of a single incident.

The fact that the criminal law may count each type of penetration as a separate crime seems to me to be irrelevant. Alaska Statute 09.17.010 applies to civil not criminal cases. There is no analogous statute or legal principle in criminal law that mandates that multiple affronts sustained by one person as a result of a single incident must be considered a single crime.20 The purposes of the crimi*388nal law, briefly stated, are to condemn, punish, and deter wrongful conduct.21 These purposes are unrelated to the purposes of AS 09.17.010 — limiting tort recoveries for non-economic losses in order to control the cost of liability insurance.

In summary, the brutal act that gave rise to this case was a single incident within the common usage of that term. This usage controls the meaning of AS 09.17.010(d) and is consistent with its purposes, whereas a narrower usage would not be. The test we use to define the synonymous word “occurrence” is also relevant to the meaning of “incident” here and reveals that the several penetrations constitute one incident. Again, it seems beyond question that the plaintiff here would not be able to bring a lawsuit based on one penetration and then later sue separately on another penetration. The fact that the several penetrations that occurred were separate crimes is irrelevant because no principle similar to the “single incident” limitation of AS 09.17.010(d) applies in criminal law, and the criminal law rule permitting the imposition of separate convictions for each penetration in a sexual assault has purposes that are unrelated to the purposes of AS 09.17.010(d).

For these reasons I would uphold the determination of the superior court that the injuries suffered by the victim in this case are subject to a single cap.