concurring in part and dissenting in part.
The majority concludes that the trial court erred in dismissing the writ of habeas corpus in this case because petitioner is entitled to be released on parole on his life sentence for aggravated murder. According to the majority, Norris v. Board of Parole, 331 Or 194, 13 P3d 104 (2000), cert den, 534 US 1028 (2001), compels that conclusion. The majority is wrong. It focuses on one aspect of Norris that arguably conflicts with other aspects of the decision without making any effort to reconcile the conflict. In fact, the conflicting provisions can be harmonized. When they are harmonized, Norris is not authority for the majority’s conclusion that petitioner is entitled to be released on parole, and there is no other support for that conclusion. I therefore dissent from the majority’s decision to reverse the dismissal of the writ of habeas corpus. I concur in its resolution of the balance of the appeal.
A jury convicted petitioner of aggravated murder for a murder that he committed in 1984. Pursuant to ORS 163.105(2) (1983), the circuit court entered a judgment in December 1984 that sentenced petitioner on that conviction to life in prison with a minimum term of 20 years’ imprisonment without the possibility of parole or work release.1 Under ORS 163.105(3), a person who has served 15 years of a *34520-year minimum term for aggravated murder can require the Board of Parole and Post-Prison Supervision (board) to hold a hearing to determine if he is a person who is “likely to be rehabilitated within a reasonable period of time.” Petitioner requested such a hearing, which the board held. As a result of the hearing, the board issued an order on February 15, 2000, that found that petitioner met the statutory rehabilitation standard and that converted the terms of petitioner’s confinement to life imprisonment with the possibility of parole. The order also set a projected parole release date for petitioner of May 18, 2004.
In November 2000, petitioner filed a petition in circuit court for a writ of habeas corpus, contending that he was entitled to be released on parole when the board issued its February 15, 2000, order. The circuit court issued the writ but later dismissed it. Whether the court erred in dismissing the writ turns on whether the board’s order entitled petitioner to be released on parole when the board issued the order. The resolution of that issue depends, in turn, on Norris.
Norris involved a petitioner who was serving consecutive life sentences for aggravated murder, both of which had 20-year minimum terms. Pursuant to ORS 163.105, the board held a hearing to determine if the petitioner was likely to be rehabilitated within a reasonable time. The board issued an order that found that he was such a person. The board also issued an order in which it decided not to modify the consecutive 20-year minimum terms. As a consequence, it set the petitioner’s projected parole release date for a date in 2018, which was 40 years after he had begun serving his first aggravated murder sentence.
On judicial review of the board’s final order, we reversed the order. On Supreme Court review of our decision, the court also concluded that the board had erred in issuing its final order, but it relied on grounds different from ours. As framed by the parties, the issue for the court was the effect to be given to the board’s finding that the petitioner was likely to be rehabilitated within a reasonable time. The petitioner contended that the effect of the finding was to eliminate both of the 20-year minimum sentences and to require the board *346to apply the matrix system to set his parole release date. The board contended that the sole effect of the finding was to make the petitioner a person who could be paroled. It did not affect the consecutive 20-year minimum sentences, which could be altered only if the board made additional findings to override them. Consequently, the petitioner would be required to serve 40 years in prison before he would become eligible to be released on parole.
The court did not agree with either party’s position. It concluded that the finding affected only the first of the two consecutive aggravated murder sentences. The consequences of the finding on that sentence were (1) to remove the restriction that barred the petitioner from parole or work release and (2) to eliminate the 20-year minimum term. Norris, 331 Or at 207-08.
The modification of the first aggravated murder sentence did not affect the second aggravated murder sentence that was consecutive to it. The second sentence still barred the petitioner from parole or work release, and that bar would remain in effect until the board found on that sentence that the petitioner was likely to be rehabilitated within a reasonable period of time. Furthermore, the board could not make that finding until the petitioner had served at least 15 years on the second sentence, at which point the petitioner could petition the board to conduct a hearing to determine whether he met the statutory rehabilitation standard. Only if the board made a rehabilitation finding after such a hearing would the second aggravated murder sentence be modified, which would then make the petitioner eligible for parole. See id. at 208; ORS 163.105(4).
The majority and I appear to agree on the foregoing statements about Norris and ORS 163.105. We part company on the import of the court’s statement in Norris that the petitioner began serving his second aggravated murder sentence on the date that the board issued its rehabilitation order on the first aggravated murder sentence, which converted the first sentence to one on which he could be paroled. Normally, a person does not start serving a sentence that is consecutive to a preceding indeterminate sentence until the person has been paroled on the first sentence. Because the court said in *347Norris that the petitioner began serving his second aggravated murder sentence when the board issued its rehabilitation order that modified the terms of the first aggravated murder sentence, the majority reasons that the Norris court’s statement has to mean that the petitioner not only became eligible for parole on the first sentence on entry of the rehabilitation order but had to be paroled on the first sentence on entry of the order. It therefore concludes that anyone who is serving an aggravated murder sentence that is subject to ORS 163.105, such as petitioner, has to be paroled when the board enters an order that finds that the person is likely to be rehabilitated within a reasonable time. In other words, although the statute and Norris say that the finding converts the relevant sentence to one in which the person has a possibility of parole, the majority interprets Norris to establish that the sentence is converted to one in which the person has a certainty of immediate parole.
The flaw in the majority’s reasoning is that it fails to recognize the significance to be attached to the fact that the petitioner in Norris was serving consecutive aggravated murder sentences that were subject to ORS 163.105. Until the board made a rehabilitation finding on the second aggravated murder sentence, the petitioner was not someone who was eligible for parole, which meant that he was not someone for whom the board could establish a parole release date. See Norris, 331 Or at 208. Because he was not such a person, there was no parole date that applied to his first sentence when the board issued its rehabilitation order, and the court appears to have assumed that no date could or would be set on that sentence until the board issued a rehabilitation order on the second sentence. See id.; Severy v. Board of Parole, 318 Or 172, 178-79, 864 P2d 368 (1993); ORS 144.110(2)(b)(A).2 *348Nevertheless, there had to be a date on which the petitioner could begin serving the second sentence, and the court concluded that the date on which the board issued its rehabilitation order that modified the terms of the first sentence was the appropriate date to use for that purpose.
Given that understanding of Norris, it follows that the majority is wrong to interpret the case as establishing that a person for whom the board issues a rehabilitation order under ORS 163.105 is a person who must be released on parole on the date that the order is issued. Furthermore, even if my understanding of Norris were incorrect, the majority’s application of Norris to this case would still be wrong.
At a minimum, the majority recognizes that its conclusion that Norris establishes that an inmate must be released on parole on an aggravated murder sentence when the board makes a rehabilitation finding under ORS 163.105 conflicts with other aspects of Norris. For example, the court said in Norris that a rehabilitation finding on the second aggravated murder sentence in that case would make the petitioner “eligible” for parole. However, according to the majority, the petitioner would not just be eligible for parole when the board made the finding, he would have to be paroled.
The court also said in Norris:
“Because [the] petitioner is not eligible for parole unless the Board finds him capable of rehabilitation at [the] second hearing, we express no opinion in this case regarding either the applicability of the matrix or whether the Board erred in its findings in aggravation.”
331 Or at 208. That discussion refers to questions that had been presented by the parties in Norris concerning the calculation of the petitioner’s parole release date. Implicit in *349that statement is the assumption that the board would calculate a parole release date for the petitioner after it issued a rehabilitation order on the second sentence. Until it did that, any discussion of the factors by which it would make that calculation was premature. Again, however, according to the majority, there would be no parole release date to establish for the sentences, because the petitioner would have to be released on parole when the board issued its rehabilitation order on the second sentence. Hence, the Norris court was wrong to suggest that the matrix and aggravation issues would not need to be addressed until the board issued a second rehabilitation order. According to the majority, Norris necessarily rendered those issues moot.
Assuming that the apparent conflicts in Norris cannot be reconciled,3 the majority does not explain why it gives effect to one side of the conflict rather than the other. If we must choose, I believe that we should apply Norris in a way that is most consistent with the relevant statutes.4 That application is one that makes people serving aggravated murder sentences that are subject to ORS 163.105 eligible for parole when the board issues a rehabilitation order but does not require them to be paroled when the board issues the order.5
In summary, the majority errs in concluding that Norris requires us to conclude that the board was required to release petitioner on parole on the date that it issued its order that found that he was likely to be rehabilitated within a reasonable time. There is no other ground that petitioner has raised that establishes that the trial court erred in dismissing his claims. Hence, I respectfully dissent from the decision *350to reverse the trial court judgment that dismissed the writ of habeas corpus.
Because 1984 is the year in which petitioner committed the murder for which he was convicted, the statutes that apply to his conviction are the 1983 versions of them. Hence, all citations to statutes in this opinion are to the 1983 versions.
ORS 144.110(2)(b)(A) provides:
“Notwithstanding the provisions of ORS 144.120 and 144.780:
"*****
“(b) The board shall not release a prisoner on parole:
“(A) Who has been convicted of aggravated murder defined as aggravated murder under the provisions of ORS 163.095, except as provided in ORS 163.105[.]”
In the context of consecutive aggravated murder sentences that are subject to ORS 163.105, that provision can be understood to mean that the board cannot release an inmate on parole on either sentence until it has made the rehabilitation finding *348required by ORS 163.105 on both sentences. Whether or not that is the correct understanding, the Norris court appears to have interpreted the statute to work that way, see Norris, 331 Or at 208, which means that, if the petitioner could not be released on parole on the first sentence, then he could never begin serving the second sentence. The court resolved the conundrum by treating the date that the board issued its rehabilitation order on the first sentence as the date on which the petitioner began serving the second sentence.
Of course, they can be. See 194 Or App at 345-48 (Armstrong, J., concurring in part and dissenting in part).
See Aylett v. Mardis, 59 Or App 109, 111-14, 114 n 6, 650 P2d 165, rev den, 294 Or 212 (1982) (court refused to follow erroneous Supreme Court decision in case that conflicted with controlling legal principle).
The incongruity of a contrary understanding of the statutes should be obvious. It makes no sense for ORS 163.105 to focus on whether an inmate is likely to be rehabilitated within a reasonable time but to require the inmate to be immediately released on parole before he has been rehabilitated. If there are conflicting aspects of Norris and we cannot give effect to one aspect of it without rejecting another, then we should reject the aspect that makes no sense. See Aylett, 59 Or App at 111-14, 114 n 6.