ALLBEE, Justice.

This case stems from a drag race between defendant Michael Dwayne McFadden and another driver, Matthew Sulgrove, which occurred on a Des Moines city street in April 1980. During the course of the two vehicles' southbound progression, Sulgrove lost control of his automobile and swerved into a lane of oncoming traffic, where he struck a lawfully operated northbound vehicle. This third vehicle contained a six-year-old passenger, Faith Ellis, who was killed in the collision along with Sulgrove. Defendant's automobile did not physically contact either of the two colliding vehicles. Further details concerning the race and the accident will be related as necessary for treatment of the issues raised by defendant.

Defendant was charged with two counts of involuntary manslaughter, a violation of section 707.5(1), The Code 1979. Having waived a jury, defendant was tried to the court and convicted and sentenced on both counts. In this appeal, he challenges the validity of his convictions and the sentences imposed.

I. Theory of liability.

Section 707.5(1) defines involuntary manslaughter as follows:

A person commits a class "D" felony when the person unintentionally causes the death of another person by the commission of a public offense other than a forcible felony or escape.

Although not expressly stated in the statute, we have held that the underlying public offense must be committed recklessly in order to convict a person under section 707.5(1). State v. Conner, 292 N.W.2d 682, 686, 689 (Iowa 1980).

Proof at trial here was primarily directed toward the public offenses of reckless driving, § 321.277, The Code 1979, and drag racing, § 321.278, The Code 1979. On appeal, defendant does not argue that the evidence was insufficient for trial court to find that both he and Sulgrove recklessly committed those public offenses prior to the accident. Rather, defendant's main contention is that proof of the causation element of section 707.5(1) was lacking. See Divisions II and III, infra.

Trial court found that defendant was guilty of involuntary manslaughter under each of three separate theories: (1) that defendant aided and abetted Sulgrove in Sulgrove's commission of involuntary manslaughter, see § 703.1, The Code 1979 (defining aiding and abetting); (2) that defendant was vicariously responsible for Sulgrove's commission of involuntary manslaughter by reason of their joint participation in the public offense of drag racing, see § 703.2, The Code 1979 (defining joint criminal conduct); and (3) that defendant himself committed the crime of involuntary manslaughter by recklessly engaging in a drag race so as to proximately cause the Sulgrove-Ellis collision.

We note that aiding and abetting and joint criminal conduct are theories of vicarious liability, based on Sulgrove's commission of involuntary manslaughter. Although a vicarious liability theory may be sufficient to convict defendant for the death of Faith Ellis, the same is not true with regard to the death of Sulgrove. This is because the involuntary manslaughter statute requires proof that the perpetrator caused the death of "another person." See § 707.5(1). Obviously, Sulgrove could not have committed involuntary manslaughter with respect to his own death. Therefore, a theory under which defendant is only vicariously liable for Sulgrove's crime would be inadequate to convict defendant for Sulgrove's death.

We turn, then, to consideration of the third theory of liability, i.e., that defendant's reckless commission of the public offense of drag racing was a proximate cause of the Sulgrove and Ellis deaths.

II. Causation: legal principles.

As stated earlier, most of the issues raised by defendant on this appeal concern the causation element of section 707.5(1). In addition to challenging the sufficiency of the evidence of causation, defendant raises certain issues concerning governing legal principles. We will address those legal questions in this division.

Preliminarily, we note that the fact that defendant's automobile did not physically contact either of the other two vehicles does not, standing alone, preclude his conviction. This rule was established in Iowa in another drag-racing case, State v. Youngblut, 257 Iowa 343, 132 N.W.2d 486 (1965), where a defendant was held to have been properly charged with involuntary manslaughter under similar facts. Having taken initial note of Youngblut, we proceed to address defendant's legal arguments concerning causation.

A. Legal effect of Sulgrove's voluntary participation.

Defendant asserts that because Sulgrove was a competitor in the drag race, he assumed the risk of his own death, and therefore defendant could not be convicted or sentenced for that death. This question was not raised in Youngblut because the only victim there was an innocent third party who had been traveling in a nonracing vehicle.

Defendant's position finds some support in State v. Petersen, 270 Or. 166, 167-68, 526 P.2d 1008, 1009 (1974), a drag-racing case in which the court held that Oregon's involuntary manslaughter statute "should not be interpreted to extend to those cases in which the victim is a knowing and voluntary participant in the course of reckless conduct." Drag-racing cases from other jurisdictions, however, have held defendants liable for manslaughter in the death of a co-participant. State v. Melcher, 15 Ariz. App. 157, 159, 161-62, 487 P.2d 3, 5, 7-8 (1971); Campbell v. State, 285 So.2d 891, 892, 895 (Miss. 1973); Commonwealth v. Peak, 12 Pa.D. C.2d 379, 381-82 (1957). Although Peak appears to have been effectively overruled by Commonwealth v. Root, 403 Pa. 571, 170 A.2d 310 (1961), a case we will discuss further below, we find ourselves in agreement with Peak rather than with Root. Therefore, we quote with approval the following discussion from Peak which is pertinent to the issue at hand:

Defendants by participating in the unlawful racing initiated a series of events resulting in the death of Young. Under these circumstances, decedent's own unlawful conduct does not absolve defendants from their guilt. The acts of defendants were contributing and substantial factors in bringing about the death of Young. The acts and omissions of two or more persons may work concurrently as the efficient cause of an injury and in such case each of the participating acts or omissions is regarded in law as a proximate cause.

12 Pa.D. C.2d at 382. See also State v. Shimon, 182 N.W.2d 113, 115-16 (Iowa 1970).

We hold that the fact of Sulgrove's voluntary and reckless participation in the drag race does not of itself bar defendant from being convicted of involuntary manslaughter for Sulgrove's death.

B. Standard of causation.

Next, defendant contends trial court erred in applying the civil standard of proximate cause in a criminal prosecution, rather than adopting the more stringent standard of "direct causal connection" used by the Pennsylvania court in Commonwealth v. Root, 403 Pa. 571, 580, 170 A.2d 310, 314 (1961). In Root, the court held that "the tort liability concept of proximate cause has no proper place in prosecutions for criminal homicide and more direct causal connection is required for conviction." Id. The court cited the facts of another Pennsylvania case as an example of such direct causation:

In [Commonwealth v. Levin, 184 Pa. Super. 436, 135 A.2d 764 (1957)] two cars were racing on the streets of Philadelphia at speeds estimated at from 85 to 95 miles per hour. The defendant's car, in the left hand lane, was racing alongside of the car in which the deceased was a passenger when the defendant turned his automobile sharply to the right in front of the other car thereby causing the driver of the latter car to lose control and smash into a tree, the passenger being thrown to the road and killed as a result of the impact. . . . Levin's act of cutting his automobile sharply in front of the car in which the deceased was riding directly forced that car off of the road and into the tree. The defendant's reckless and unlawful maneuver was the direct cause of the crucial fatality.

Id. at 576, 170 A.2d at 312.

We had occasion to consider a similar standard-of-causation issue in State v. Marti, 290 N.W.2d 570, 584-85 (Iowa 1980), which upheld the involuntary manslaughter conviction of a man who provided an obviously intoxicated, suicidal woman with the means to shoot herself by loading a gun for her and placing it within her reach. As here, the defendant in Marti argued that the trial court "inappropriately adopted the standards of proximate cause applied in civil cases." Id. at 584. Unlike the Pennsylvania court in Root, however, we said in Marti that we were "unwilling to hold as a blanket rule of law that instructions used in civil trials regarding proximate cause are inappropriate for criminal trials." Id. We explained:

One reason for this is the similar functions that the requirement of proximate cause plays in both sorts of trials. The element of proximate cause in criminal prosecutions serves as a requirement that there be a sufficient causal relationship between the defendant's conduct and a proscribed harm to hold him criminally responsible. [Citations omitted.] Similarly, in the law of torts it is the element that requires there to be a sufficient causal relationship between the defendant's conduct and the plaintiff's damage to hold the defendant civilly liable. [Citation omitted.]

Id.

We did note in Marti, however, that legal causation (as opposed to factual causation) is "essentially a question of whether the policy of the law will extend the responsibility for the conduct to the consequences which have in fact occurred." Id. at 585 (quoting W. Prosser, Handbook of the Law of Torts § 42, at 244 (4th ed. (1971)). Further, we recognized "that different policy considerations may come into play in criminal prosecutions than in civil trials," and that an "argument could be made that these differences should be reflected in the proximate cause instructions used in the different kinds of trials." Id. Nevertheless, we had no occasion to consider whether such differences existed in that case because the defendant there

failed to indicate any differences in policy considerations relevant to his case, much less how such differences might have prejudicially affected any particular instruction given here. For a party to preserve for appeal his objections to a trial court's instructions on this amorphous and multifaceted matter of legal causation, he or she must specify the deficiency in narrower terms than a mere assertion that they copy civil instructions.

Id. We find no error preservation problem in this bench-tried case, and thus proceed to address the standard-of-causation question on its merits.

First, although defendant does not cite or appear to rely on State v. Rullestad, 259 Iowa 209, 212-13, 143 N.W.2d 278, 280 (1966), we are aware of Rullestad's holding that to sustain an involuntary manslaughter conviction based on the public offense of drunk driving, it is necessary to show a "direct causal connection" between the drunk driving and the death. Rullestad cites Root among a list of cases "having some bearing" on the point. While Rullestad does use the phrase "direct causal connection," the case contains no discussion of the distinction between the ordinary concept of proximate cause and the more stringent standard adopted in Root. Elsewhere in Rullestad, the term "proximate cause" is used repeatedly in referring to the appropriate legal standard. 259 Iowa at 212-14, 143 N.W.2d at 280-81. Thus, we believe Rullestad's use of the phrase "direct causal connection" was intended to convey nothing more than the ordinary notion of proximate cause.

This view is reinforced by the fact that Youngblut, 257 Iowa 343, 132 N.W.2d 486, implicitly rejects the causation standard adopted in Root. The facts outlined in the minutes of testimony attached to the indictment in Youngblut did not include any specific act by defendant which directly caused his fellow racer to collide with the hapless third vehicle. Thus, those facts would not have met the direct causation test of Root. Yet this court held that those facts were legally sufficient to satisfy the elements of involuntary manslaughter.

Moreover, we observe that although Root had been decided four years earlier and was the subject of an American Law Reports annotation, see 82 A.L.R.2d 452 (1962), Youngblut did not cite the Root case or follow its lead. Rather, Youngblut relied primarily on the drag-racing cases of People v. Kemp, 150 Cal.App.2d 654, 659, 310 P.2d 680, 683 (1957), and State v. Fennewald, 339 S.W.2d 769, 773 (Mo. 1960). The facts of neither of those cases would have met the direct causation standard of Root. In addition, we note that in Kemp, at least, the theory of liability was based on the concept of proximate causation. See 150 Cal.App.2d at 658-59, 310 P.2d at 682-83.

Furthermore, of all the involuntary manslaughter cases we have studied which involve drag racing, we have found only one that applies a standard of causation as stringent as that in Root: Thacker v. State, 103 Ga. App. 36, 37-39, 117 S.E.2d 913, 914-15 (1961). Other cases, when not based on a theory of vicarious liability, appear to apply a proximate cause standard similar to that in tort cases. E.g., State v. Melcher, 15 Ariz. App. 157, 161-62, 487 P.2d 3, 7-8 (1971); Campbell v. State, 285 So.2d 891, 893-95 (Miss. 1973).

Finally, defendant has suggested no specific policy differences, nor can we think of any, that would justify a different standard of proximate causation under our involuntary manslaughter statute than under our tort law. The Root court opined that "[l]egal theory which makes guilt or innocence of criminal homicide depend upon such accidental and fortuitous circumstances as are now embraced by modern tort law's encompassing concept of proximate cause is too harsh to be just." 403 Pa. at 576, 170 A.2d at 312. We do not agree. Proximate cause is based on the concept of foreseeability. We believe the foreseeability requirement, coupled with the requirement of recklessness, see State v. Conner, 292 N.W.2d 682, 686 (Iowa 1980), will prevent the possibility of harsh or unjust results in involuntary manslaughter cases. We disagree with the Root court's apparent opinion that drag racing on a public street is "not generally considered to present the likelihood of a resultant death." 403 Pa. at 575, 170 A.2d at 311.

Accordingly, we hold that trial court did not err in applying ordinary proximate cause principles to determine whether the causation element of section 707.5(1) had been met, and in declining to adopt the more stringent "direct causal connection" standard of Root.

C. Withdrawal as a defense.

Defendant argues that trial court erred in failing to find that defendant had withdrawn from the drag race prior to the collision between the Sulgrove and Ellis vehicles. The legal underpinning for his asserted withdrawal defense was State v. Fair, 209 S.C. 439, 40 S.E.2d 634 (1946). In that case, it was recognized that one drag racer could be held responsible for a death resulting from a collision between his competitor and a third party, under the vicarious liability theory of joint criminal enterprise; however, the defendant's conviction was reversed because of the trial court's failure to instruct the jury on the effect of withdrawal. Id. at 443-45, 40 S.E.2d at 636-37.

The withdrawal defense is usually associated with the theory of joint criminal enterprise, as it was in Fair. As noted earlier, however, that theory of liability is not sufficient in this case to justify defendant's conviction for the death of Sulgrove; thus, the theory we have been examining is based on the idea that defendant's own reckless commission of a public offense "cause[d]" the Sulgrove and Ellis deaths within the meaning of section 707.5(1). Under this this theory, the notion of withdrawal is pertinent only insofar as it relates to the element of proximate cause. For example, if one drag racer were to abandon the race by slowing down to normal speeds or stopping, and his competitor became aware of the defendant's withdrawal but still chose to continue driving fast and recklessly, that fact might have a bearing on whether the defendant's drag racing was a proximate cause of a subsequent collision between his competitor and a third party; it might also bear on whether the competitor's decision to continue his reckless driving was an intervening, superseding cause. See Saisa v. Lilja, 76 F.2d 380, 381 (1st Cir. 1935); Jones v. Northwestern Auto Supply, 93 Mont. 224, 229-31, 18 P.2d 305, 306-07 (1932); Boykin v. Bennett, 253 N.C. 725, 732, 118 S.E.2d 12, 17 (1961); Lemons v. Kelly, 239 Or. 354, 356-60, 397 P.2d 784, 785-87 (1964). Thus, the point we wish to make here is that under the theory of direct liability we are considering, a defendant's asserted withdrawal should not be viewed as an absolute defense, but only as a factor affecting the determination of proximate cause. We will so consider it in the following division when we review the sufficiency of the evidence of proximate causation.

III. Causation: sufficiency of the evidence.

A. Standard of review.

Before considering the sufficiency of the evidence of causation in this case, we must address defendant's contention that this court should adopt a "strict scrutiny" standard in reviewing the sufficiency of the evidence in a criminal bench trial. A similar proposal was made by a defendant and rejected by this court in State v. Hall, 287 N.W.2d 564, 564-65 (Iowa 1980). In that case, this court held to the rule that in court-tried criminal cases, "findings of fact, if supported by substantial evidence, have the same effect as a special jury verdict." Id. at 565. Defendant has shown no good reason why we should depart from this "substantial evidence" standard of review in bench trial cases.

Accordingly, we will apply the usual rules for reviewing the sufficiency of the evidence, namely, that the evidence is viewed in the light most favorable to the State; that all of the evidence must be considered, and not just that which supports the verdict; that the verdict must be upheld if supported by substantial evidence; and that substantial evidence means such evidence as could convince a rational trier of fact that the defendant is guilty beyond a reasonable doubt. State v. Robinson, 288 N.W.2d 337, 338-40 (Iowa 1980).

B. Sufficiency of the causation evidence in this case.

With the foregoing principles in mind, we review the evidence in this case. Testifying for the State were three witnesses who were driving in the vicinity on the morning of the accident, and two police department accident reconstruction experts who had been called to the scene. Defendant's only witness was a private investigator who, as a former police officer, also had accident reconstruction training and experience. In addition, the trial judge viewed the scene of the drag race and accident under the stipulation of both parties that anything he observed could be evidence in either the State's or defendant's case.

On Saturday morning, April 5, 1980, a green car driven by Sulgrove was observed moving at a high rate of speed off the exit ramp of the MacVicar Freeway onto Southeast Fourteenth Street, a multi-lane city street. In making his exit, Sulgrove hit a curb, "fishtailed," and then entered the southbound traffic on Southeast Fourteenth. Traffic was fairly heavy at this point, and Sulgrove moved from one southbound lane to another trying unsuccessfully to pass cars and get ahead. Farther down the street, which had narrowed to two northbound and two southbound lanes, Sulgrove's car was seen traveling at a "fairly high" rate of speed; it came up behind a red car driven by defendant, an acquaintance of Sulgrove's, just past the Gratis Street intersection. Defendant, traveling at an estimated speed of 40-47 m.p.h., had been in the left southbound lane until just before Sulgrove came up behind him; at that time, defendant pulled his car into the right southbound lane alongside a car in the left lane, thereby preventing Sulgrove from passing. Defendant's maneuver forced Sulgrove to brake suddenly, and Sulgrove's car again "fishtailed." After getting past the vehicle which blocked the left lane, the red and green vehicles were observed engaging in what a witness described as a "cat-and-mouse" game: Defendant's red car, traveling generally in the right lane just ahead of Sulgrove's green car, repeatedly blocked Sulgrove's attempts to pass by moving partially into the left lane and then back to the right again when Sulgrove would change lanes. Starting at the Pioneer Street intersection, where the speed limit is 40 m.p.h., a witness who had some experience in drag racing under controlled conditions saw the two cars traveling "head to head" at an estimated speed of 70-75 m.p.h., with defendant in the right lane and Sulgrove in the left. The witness testified that when the two cars came upon a gold car in the left lane near the Lacona Street intersection, "neither driver was willing to give up the [competitive] edge in order to negotiate the gold car until at the very last second," when Sulgrove braked, swerved into the right lane behind defendant, bounced off a curb, and passed the gold car. Sulgrove then got back into the left lane and the two cars continued southward at a high rate of speed. Because the two vehicles were "traveling as a pair" in a competitive manner, the witness concluded they were drag racing.

Beyond the Lacona intersection, Southeast Fourteenth slopes upward and crests twice, the first crest being between Creston and King streets and the second crest being 1126 feet beyond the first. A witness named Jamison who was traveling behind the red and green vehicles testified that he watched both vehicles continue to speed and participate in "cat-and-mouse" activity up to the top of the second crest, where they dropped out of his sight. Just beyond the second crest, a pickup truck was traveling in the left southbound lane at a normal speed. The pickup driver testified that Sulgrove's car "came around" her truck at a high rate of speed in the right lane, lost control and skidded across the left lane in front of her. Sulgrove then crossed over the median and struck the northbound Ellis vehicle at a point 263 feet beyond the second crest. The pickup driver did not recall seeing defendant's red car.

When the police arrived, they found 84 feet of skid marks left by Sulgrove's vehicle. It was determined from physical evidence that Sulgrove had been going about 80 m.p.h. just before he went into the skid. Defendant was present at the accident site, and his car was parked in the lot of Godfather's Pizza, which is located on the west side of the street. The southernmost driveway to this lot is 351 feet beyond the second crest. There were no skid marks left by defendant's car anywhere in the vicinity.

Defendant challenges the sufficiency of the foregoing evidence under both perspectives of causation, factual and legal. See State v. Marti, 290 N.W.2d 570, 584-85 (Iowa 1980). Factual causation is determined under the sine qua non test: "but for the defendant's conduct, the harm or damage would not have occurred." Id. at 585. Defendant argues that because Sulgrove was driving fast and recklessly even before defendant entered the picture, the accident would have occurred even without defendant's participation in the drag race. Viewing the evidence in the light most favorable to the State, however, we find there was substantial evidence from which trial court could conclude that Sulgrove's speed and recklessness both increased once he entered the heat of competition with defendant, and that the accident would not have occurred but for their joint racing.

Defendant next asserts that proof of proximate or legal causation was lacking because there was no credible evidence that defendant continued to race beyond the first crest of the hill. Witness Jamison, mentioned earlier, testified unequivocally that he observed the red and green cars go over both crests of the hill. He said: "Well, there is two small hills there as you are going up that hill, and I never lost sight of them over the first hill because it is high enough up that I could see them, and the last hill, whichever one it is where they go over the crest was where I saw them last." Jamison was the only witness who claimed to have observed defendant's activity between the first and second crests. Defendant attempted to discredit that testimony by proving that Jamison could not have observed the cars once they passed over the first crest. It was shown that a person sitting in a stationary vehicle at the Lacona intersection can see only the first crest, and that traffic disappears from sight as it goes over that crest. When a person in a moving vehicle comes within 150 feet of the first crest, however, he can see over that crest all the way to the second crest. Thus, adding 150 feet to the distance between the crests, it can be seen that Jamison would have had to be no more than 1276 feet behind the red and green cars in order to watch them go over the second crest. Jamison estimated that he was roughly 1/4 to 1/2 mile (1320 to 2640 feet) behind the two cars when they went over the second crest, but he also disclaimed any ability to estimate distance accurately. Indications that the distance was not great included Jamison's testimony that he saw dust rise over the hill ahead of him when the collision occurred, and that he arrived at the accident site just as defendant was getting out of his car in the Godfather's lot.

Even if Jamison did only see the cars go over the first crest, the evidence of proximate causation would not be insufficient. As noted earlier, the distance between the two crests is 1126 feet. Evidence presented at trial concerning the number of feet per second traveled at various speeds indicates to us that it would take less than 10 seconds to go from crest to crest at 80 m.p.h. and less than 13 seconds at 60 m.p.h. There was substantial evidence that defendant and Sulgrove were traveling in the 60-80 m.p.h. range when they topped the first crest. Based on this evidence, trial court could find that it would have been impossible for defendant, in such a short period of time, to effect a timely withdrawal which would have allowed Sulgrove an opportunity to slow down and avoid the accident. As stated in a civil drag-racing case:

[R]acing on a highway is hazardous to all other persons upon the highway and . . . the actor participates at his peril. . . . One who does participate in setting in motion such hazardous conduct cannot thereafter turn his liability off like a light switch. From the authorities cited we conclude that one who participates in setting such hazardous conduct in motion cannot later be heard to say: "Oh! I withdrew before harm resulted even though no one else was aware of my withdrawal." It would be a reasonable probability that the excitement and stimulus created by this race of several miles had not dissipated nor, in fact, terminated at all, in the fraction of a minute in time between the act of passing and the accident. The state of mind of the participants was material. We cannot gauge that state of mind to the point of saying that the stimulus or intent had ended. The evidence warrants a finding that it did continue. It would be for the jury to decide if the racing were the cause of the accident.

Lemons v. Kelly, 239 Or. 354, 360, 397 P.2d 784, 787 (1964). Thus, even if defendant had started to slow down between the first and second crests, as his counsel argues he did, trial court could find that any such last-minute effort to withdraw would not break the chain of causation set in motion by the drag race. See also Jones v. Northwestern Auto Supply, 93 Mont. 224, 229-31, 18 P.2d 305, 306-07 (1932).

For the same reason, trial court could find that the lack of skid marks from defendant's car into the Godfather's parking lot failed to create a reasonable doubt as to defendant's guilt. A defense expert testified that it would have been impossible for defendant to turn into that lot without leaving skid marks if he, like Sulgrove, was traveling 80 m.p.h. as he came over the second crest. From what we have said in the preceding paragraph, however, it was not necessary for the State to prove that defendant was going 80 m.p.h. when he topped that hill. One of the accident reconstruction experts who testified for the State theorized that defendant and Sulgrove continued to race almost side-by-side until they neared the second crest and saw the pickup truck ahead blocking Sulgrove's path, at which point defendant slowed down to allow Sulgrove into the right lane. This theory would explain why the pickup driver did not notice defendant's car, and would also be one explanation for the lack of skid marks from defendant's car.

We have examined defendant's remaining complaints concerning asserted flaws in the State's case and in trial court's findings. We conclude that the few flaws or errors which do exist are nonprejudicial and that the remainder of defendant's assertions are without merit.

Viewing the evidence in the light most favorable to the State, we hold that the record contains substantial evidence that defendant's participation in a drag race with Sulgrove was a concurring proximate cause of the accident in which Sulgrove and Faith Ellis were killed. We therefore affirm defendant's convictions.

IV. Sentences on each count of involuntary manslaughter.

Trial court's imposition of a separate sentence of no more than five years confinement and a $500 fine on each of the two counts of involuntary manslaughter charged by the trial information is also challenged by defendant. The sentences were to run concurrently.

Defendant argues that only one sentence is permissible because both deaths arise out of a single act of recklessness. In support of this argument, defendant relies on State v. Wheelock, 216 Iowa 1428, 250 N.W. 617 (1933), where three indictments of involuntary manslaughter were returned against the defendant predicated upon a single accident in which three persons in an automobile struck by defendant's vehicle died. The defendant was tried on one of the charges and acquitted, and then pleaded former jeopardy when the State sought to try him on another of the charges. In recognizing the validity of the former jeopardy plea under those circumstances, Wheelock adopted the rule that "an act of negligence on the part of a tort-feasor, which results in the involuntary killing of two or more human beings, is ordinarily a single offense and is subject to one prosecution." Id. at 1448, 250 N.W. at 625 (emphasis in original). The Wheelock court explained its rationale and underscored the narrowness of the holding in this passage:

Except as to cases of involuntary manslaughter, it is not the purpose, or the purport, of this opinion, to establish any new precedent or to extend or restrict what has already been said in our own cases. Our opinion herein shall go no farther than to determine whether an involuntary manslaughter of two or more persons attributable to the single negligence of the defendant, without any intent on his part to cause any injury, is a single or a multiple offense. If the respective courts differ in their conclusions in cases involving other offenses, such difference arises nevertheless on the question of fact as to whether the intent of the perpetrator was single or plural. In the case before us there is no basis for the claim of multiple intent either as a question of fact or of law.

Id. (emphasis in original).

The procedural posture in Wheelock can be distinguished from that presented here: Wheelock involved the issue of successive trials where a single act produced multiple victims; this case presents multiple charges in a single prosecution arising from a single act. Nevertheless, Wheelock does lend support for defendant's argument that only a single sentence may be imposed because both charges, even though contained in a single prosecution, arise out of a single act.

The Wheelock holding, we believe, must be re-examined in light of developments since it was decided. When announcing the rule that multiple-victim involuntary manslaughter resulting from a single act is but a single offense, the Wheelock court stated that in so holding it was "[i]n line with all the other courts, which have passed upon the specific question. . . ." Id. This expression in that opinion followed an extensive survey of authorities from across the country. It is clear, however, that today the overwhelming weight of authority recognizes that there are as many separate and distinct offenses as there are deaths resulting from a single incident of vehicular manslaughter. See, e.g., State v. Miranda, 3 Ariz. App. 550, 557-58, 416 P.2d 444, 451-52 (1966); McHugh v. State, 160 Fla. 823, 824, 36 So.2d 786, 787 (1948), cert. denied, 336 U.S. 918, 69 S.Ct. 640, 93 L.Ed. 1081 (1949); State v. Lowe, 130 So.2d 288 (Fla.Dist.Ct. App. 1961); People v. Allen, 368 Ill. 368, 379, 14 N.E.2d 397, 405 (1937), appeal dismissed, 308 U.S. 511, 60 S.Ct. 132, 84 L.Ed. 436 (1939); Fleming v. Commonwealth, 284 Ky. 209, 210-11, 144 S.W.2d 220, 221 (1940); Burton v. State, 226 Miss. 31, 45, 79 So.2d 242, 249-50 (1955); State v. Whitley, 382 S.W.2d 665, 667 (Mo. 1964); Jeppesen v. State, 154 Neb. 765, 768-69, 49 N.W.2d 611, 613-14 (1951); State v. Martin, 154 Ohio St. 539, 541-42, 96 N.E.2d 776, 778 (1951); Fay v. State, 62 Okla. Cr. 350, 357, 71 P.2d 768, 771 (1937); State v. Irvin, 603 S.W.2d 121 (Tenn. 1980); State v. Rabe, 96 Wis.2d 48, 72-76, 291 N.W.2d 809, 821-22 (1980); see also 7A Am.Jur.2d, Automobiles and Highway Traffic § 391 (1980).

For several reasons, we are persuaded that we should adopt the majority view. First, we believe Wheelock's focus upon the lack of intent to inflict multiple deaths is faulty. The role of intent pertains only to the commission of the act which causes the death of others, not to the number of deaths which result from the act. By the very nature of the offense, involuntary manslaughter deaths are unintentional, the fortuitous result of an act found to be reckless. Next, we recognize that a single reckless act can foreseeably victimize several persons. This case serves as an example. See State v. Rabe, 291 N.W.2d at 822. Finally, the Wheelock rule, once "[i]n line with all the other courts," now finds scant acceptance. See State v. Irvin, 603 S.W.2d at 121, 124 n. 2.

We now hold that a separate and distinct offense arises from each death caused by a single act of vehicular involuntary manslaughter. Wheelock and other decisions holding to the contrary are overruled to the extent that they conflict herewith. Insofar as Wheelock addressed the issue of successive trials following an initial acquittal in the situation where a single criminal act produces multiple victims, the decision has continuing vitality. See Ashe v. Swensen, 397 U.S. 436, 90 S.Ct. 1189, 25 L.Ed.2d 469 (1970) (simultaneous robbery of six poker players: defendant tried and acquitted on one charge; trials on second and subsequent charges barred).

V. Failure to give reasons for the sentences.

Defendant also asserts that trial court erred in failing to state on the record its reason for selecting the particular sentences imposed, as required by Iowa R. Crim. P. 22(3)(d). The State concedes lack of compliance with the rule and that the sentences must be vacated. State v. Smith, 309 N.W.2d 454, 457 (Iowa 1981); State v. Luedtke, 279 N.W.2d 7, 8 (Iowa 1979). Consequently, the sentences imposed are vacated and this case is remanded for resentencing. This does not affect the convictions, which remain as rendered.

AFFIRMED IN PART; SENTENCES VACATED; REMANDED FOR RESENTENCING.