Lead Opinion
{¶ 1} In this сase, we must again consider when a trial court has the authority to correct a sentence when one of the sanctions originally imposed by the trial court is void. We hold that a trial court cannot add a term of postrelease control as a sanction for a particular offense after the defendant has already served the prison term for that offense, even if the defendant remains in prison for other offenses.
FACTS AND PROCEDURAL HISTORY
{¶ 2} On July 29, 1999, the Wyandot County Court of Common Pleas entered a judgment finding appellant, Henry Allen Holdcroft, guilty of aggravated arson and arson, pursuant to a jury verdict. In September 1999, the trial court imposed a prison term of ten years for Holdcroft’s aggrаvated-arson offense and a prison term of five years for Holdcroft’s arson offense. The trial court ordered
{¶ 3} Holdcroft completed his prison term for aggravated arson in 2009 and began serving the term of imprisonment for arson at that time. On January 26, 2010, the trial court held a new sentencing hearing to corrеct its errors related to postrelease control. Holdcroft argued in part that he had served 10/6 years in prison and that his aggravated-arson term of incarceration had expired. But the trial court reimposed a prison term of ten years for aggravated arson and a prison term of five years for arson, and it ordered the arson prison term to be served consecutively to the aggravated-arson prison term. The trial court also imposed a mandatory term of five years of postrelease control for the aggravated-arson offense and a discretionary postrelease-control term of up to three years for the arson offense.
{¶ 4} Holdcroft appealed the sentencing judgment, asserting that the trial court lacked jurisdiction to impose postrelease control related to his aggravated-arson offense because he had already served the prison sentence for that offense. In a divided decision, the Third District Court of Appeals disagreed. The appellate court held that a trial court may resentence a defendant for the purpose of correctly imposing postrelease control, so long as the defendant is still serving a prison term for any of the other offenses included in the same judgment entry of sentence. The court certified that its judgment conflicted with thе judgment in State v. Dresser, 8th Dist. Cuyahoga No. 92105,
Does a trial court have jurisdiction to resentence a defendant for the purpose of imposing mandatory post-release control regarding a particular conviction, when the defendant has served the stated prison term regarding that conviction, but has yet to serve the entirety of his aggregate prison sentence, when all of the convictions which led to the aggregate sentence resulted from a single indictment?
Holdcroft filed a notice of certified conflict and a discretionary appeal in this court. We determined that a conflict exists and accepted thе discretionary appeal, and we consolidated the cases.
ANALYSIS
{¶ 5} This court has consistently and repeatedly held that a trial court loses jurisdiction to resentence a defendant for the purpose of imposing postrelease control once the defendant has served his entire sentence of incarceration. Hernandez v. Kelly,
{¶ 6} Resolution of this issue is helped significantly by examining the definitions of the words “sentence,” “sanction,” and “conviction” and by reviewing State v. Saxon,
{¶ 7} Although we have not previously so stated, generally speaking, our recent cases in this area have dealt with void sanctions, rather than sentences that were void ab initio. For example, in Fischer,
{¶ 8} The Fischer rule does not apply to most sentencing challenges. For example, challenges to a trial court’s compliance with R.C. 2929.11 and 2929.12, which may be brought by either the defendant or the state, must still be presented in a timely direct appeal under R.C. 2953.08. R.C. 2953.08(E); see generally State v. Kalish,
{¶ 10} Given the posture of this case, it is clear that if the trial court was able to correct anything related to Holdcroft’s aggravated-arson sentence, it was able to correct the void postrelease-control sanction that was part of that sentence. But the broader question remains: what effect, if any, does the fact that Holdcroft had completed the valid prison sanction for his aggravated-arson sentence have on the trial court’s ability to impose a postrelease-control sanction for that offense? We conclude that once Holdcroft completed his prison term for aggravated arson, the trial court lost the authority to resentence him for that offense.
{¶ 11} As we noted above, we have consistently held that once an offender has been released from prison, he cannot be subjected to another sentencing to correct the trial court’s flawed imposition of postrelease control. E.g., Bloomer,
{¶ 12} Although both Simpkins and Holdcroft were incarcerated when they were resentenced, Holdcroft had served the entirety of his prison sanction for aggravated arson at the time he was resentenced. We conclude that Holdcroft had a legitimate expectation of finality in the sentence that he had fully served and that his situation is therefore more analogous to Hernandez than to Simpkins.
{¶ 14} To be consistent in sentencing, there must be a distinction between the rule of Fischer and the rule of Hernandez. Fischer’s sanction-correction rule is based on principles of res judicata — while a void sanction may be modified, a valid sanction generally cannot. Fischer,
{¶ 15} A defendant’s expectation of the finality of his sentence increases as time passes. The defendants in Fischer and Simpkins, whose appeal time had passed, had a greater expectation of the finality of their sentences than the defendant in Saxon, whose case remained on direct review. And the defendants in Bezak,
{¶ 16} This court has examined the effect that a defendant’s legitimate expectation of finality has on the court’s authority to modify a sentence, and those cases lend further support to our conclusion here. In State v. Roberts,
{¶ 17} Finally, in Simpkins,
{¶ 18} Together, all of these cases provide a clear demonstration of the role that a defendant’s legitimate expectation of finality plays in constraining a court’s authority to review a sentence, and three principles provide a framework for future reference. First, when a sentence is subject to direct review, it may be modified; second, when the prison-sanction portion of a sentence that also includes a void sanction has not been completely served, the void sanction may be modified; and third, when the entirety of a prison sanction has been served, the defendant’s interest in finality in his sentence becomes paramount, and his sentence for that crime may no longer be modified. Put another way, either the defendant or the state may challenge any aspect of a sentence so long as a timely appeal is filed. See, e.g., R.C. 2953.08 and DiFrancesco at 132. But once the time for filing an appeal has run, Ohio courts are limited to correcting a void sanction. Fischer at ¶ 27. And once the prison-sanction portion of a sentence for a crime has been fully served, the structure of Ohio felony-sentencing law and the defendant’s legitimate expectation in finality in his sentence prevent a court from further modifying the sentence for that crime in any way. A trial court does not have the authority to resentence a defendant for the purpose of adding a term of postrelease control as a sanction for a particular offense after the defendant has already served the prison term for that offense. Although it is true that some other sanctions (such as restitution) may yet be outstanding, a sentence served is a sentence completed.
{¶ 19} Neither this court’s jurisprudence nor Ohio’s criminal-sentencing statutes allow a trial court to resentence a defendant for an offense when the defendant has already completed the prison sanction for that offense. It is irrelevant whether the defendant is still in prison for other offenses. Because Holdcroft was no longer serving a prison sanction for the offense of aggravated arson, the trial court was not. authorized to impose a mandatory five years of postrelease control for that offense. We therefore reverse the judgment of the court of appeals and remand this cause to the trial court with instructions to vacate the imposition of postrelease control for Holdcroft’s aggravated-arson offense.
Judgment reversed and cause remanded.
Concurrence Opinion
concurring in judgment only.
{¶ 20} Unfortunately, the majority’s discussion of “void sanctions” continues the confusion resulting from this court’s opinions regarding errors in the imposition оf postrelease control, precedent that I have advocated against consistently, most recently in my dissent in In re J.S.,
Enactment of R.C. 2929.191
{¶ 21} To put this case in context, it is helpful to look back to a case that was decided when this court was of one mind with regard to postrelease-control sentencing errors. Hernandez v. Kelly,
As part of the General Assembly’s goal of achieving “truth in sentencing,” the new felony-sentencing law was intended to ensure that all persons with*535 an interest in a sentencing decision would know precisely the sentence a defendant is to receive upon conviction for committing a felony. The goal is that when the prosecutor, the defendant, and victims leave the courtroom following a sentencing hearing, they know precisely the nature and duration of the restrictions that have been imposed by the trial court on the defendant’s personal liberty. Confidence in and respect for the criminal-justice system flow from a belief that courts and officers of the courts perform their duties pursuant to established law.
Id. at ¶ 31.
{¶ 22} Hernandez was not properly notified of mandatory postrelease control at his initial sentencing hearing, although he had been told that “he was ‘being sent to prison and placed on post-release control by the Parole Board for a period of up to five years.’ ” Id. at ¶ 2. He appealed and was resentenced, but postrelease control was not mentioned during his resentencing hearing, nor was it included in the judgment of conviction. Id. at ¶ 4. Nevertheless, upon Hernandez’s release from prison, the Adult Parole Authority determined that he was subject to five years of postrelease control, and after he violated several postrelease сonditions, it returned him to prison. Id. at ¶ 5-6. Hernandez filed a writ of habeas corpus, seeking release from prison and from postrelease control. Id. at ¶ 7. We granted the writ, unanimously holding:
The Adult Parole Authority was not authorized to put Hernandez on postrelease control and sanction him for violating the terms of that control in the absence of appropriate notification of postrelease control by the trial court and incorporation of postrelease control in its sentencing entry. In that his journalized sentence has expired, Hernandez is entitled to the writ and release from prison and from further postrelease control.
(Emphasis added.) Id. at ¶ 32.
{¶ 23} After our decision in Hernandez, the General Assembly enacted R.C. 2929.191, which authorizes a trial court to correct a sentencing error related to the imposition of postrelease control, provided that the correction is made after the offender is given a hearing and while the offender is still serving the prison term for the relevant offense. 2006 Am.Sub.H.B. No. 137 (“H.B. 137”). Before the enactment of H.B. 137, a trial court had no special statutory power to correct a sentence that contained a postrelease-control error.
Refusal to Apply R.C. 2929.191 Retroactively
{¶ 24} Despite the clear intent of the General Assembly that R.C. 2929.191 was to apply retroactively, a majority of this court initially refused to so apply it. State v. Singleton,
{¶ 25} In discussing the retrospective application of R.C. 2929.191, the court stated:
R.C. 2929.191 purports to authorize application of the remedial procedure set forth therein to add postrelease control to sentences imposed before its effective date. We recognize the General Assembly’s authority to alter our caselaw’s characterization of a sentence lacking postrelease control as a nullity and to provide a mechanism to correct the procedural defect by adding postrelease control at any time before the defendant is released from prison. However, for sentences imposed prior to the effective date of the statute, there is no existing judgment for a sentencing court to correct. H.B. 137 cannot retrospectively alter the character of sentencing entries issued prior to its effective date that were nullities at their inception, in order to render them valid judgments subject to*537 correction. Therefore, for criminal sentences imposed prior to July 11, 2006, in which a trial court failed to properly impose postrelease control, the de novo sentencing procedure detailed in decisions of the Supreme Court of Ohio should be followed to properly sentence an offender.
Id. at ¶ 26.
{¶ 26} The court did hold, however, that “[f]or criminal sentences imposed on and after July 11, 2006, in which a trial court fаiled to properly impose postrelease control, trial courts shall apply the procedures set forth in R.C. 2929.191.” Id. at paragraph two of the syllabus.
Revival of R.C. 2929.191's Retrospective Application
{¶ 27} In Singleton, this court refused to apply R.C. 2929.191 retrospectively by stating that a sentencing entry omitting postrelease control was a void judgment and was thus a nullity incapable of being corrected. A year later, however, this court retreated from Singleton’s holding that a de novo sentencing hearing was required for postrelease-control sentencing errors in sentences imposed before July 11, 2006. State v. Fischer,
{¶ 28} Therefore, I believe that the resolution of this case lies in the retroactive application of R.C. 2929.191.
Application of R.C. 2929.191
{¶ 29} R.C. 2929.191(A)(1) states:
If, prior to July 11, 2006, a court imposed a sentence including a prison term of a type described in division (B)(2)(c) of section 2929.19 of the Revised Code and failed to notify the offender pursuant to that division that the offender will be supervised under section 2967.28 of the Revised Code after the offender leaves prison or to include a statement to that effect in the judgment of conviction entered on the journal or in the sentence pursuant to division (D)(1) of section 2929.14 of the Revised Code, at any time before the offender is released from imprisonment under that term and at a hearing conducted in accordance with division (C) of this*538 section, the court may prepare and issue a correction to the judgment of conviction that includes in the judgment of conviction the statement that the offender will be supervised under section 2967.28 of the Revised Code after the offender leaves prison.
(Emphasis added.)
{¶ 30} R.C. 2929.191(A)(2) discusses the method and effect of a correction:
If a court prepаres and issues a correction to a judgment of conviction as described in division (A)(1) of this section before the offender is released from imprisonment under the prison term the court imposed prior to July 11, 2006, the court shall place upon the journal of the court an entry nunc pro tunc to record the correction to the judgment of conviction and shall provide a copy of the entry to the offender * * *. * * * The court’s placement upon the journal of the entry nunc pro tunc before the offender is released from imprisonment under the term shall be considered, and shall have the same effect, as if the court at the time of original sentencing had included the statement in the sentence and the judgment of conviction entered on the journal and had notified the offender that the offender will be so supervised * * *.
(Emphasis added.)
{¶ 31} In other words, the General Assembly has enacted a procedure whereby postrelease control may be properly authorized and given effect, even though initial notification was inadequate, if the offender has not been released from prison “under the prison term the court imposed.” The statute refers to completion of the “prison term the court imposed” as a time limitation for the correction to be made.
Sentencing Error Is Not Jurisdictional
{¶ 32} The statutory definitions of “sentence” as “the sanction or combination of sanctiоns imposed * * * on an offender who is convicted of or pleads guilty to an offense” (R.C. 2929.01(EE)) and “sanction” as “any penalty imposed upon an offender who is convicted of or pleads guilty to an offense, as punishment for the offense” (R.C. 2929.01(DD)) make clear that these words relate to a specific offense, rather than all counts within an indictment. I agree with the majority’s analysis in regard to these definitions and its references to State v. Saxon,
{¶ 33} But the majority is incorrect in saying that this court has been consistent in holding that a trial court “loses jurisdiction” to impose postrelease
{¶ 34} Holdcroft’s incorrect sentence could have been corrected on direct appeal. Both Holdcroft and the state did in fact appeal from the trial court’s judgment, but neither party raised the postrelease-control issue on appeal. Nevertheless, by statute, the General Assembly has extended the time for correcting this type of sentencing error pursuant to R.C. 2929.191, which allows a court to correct the error provided that it is done before the prisoner is released from the prison term for the offense. The state’s argument that all of an offender’s prison terms that resulted from a single indictment should be aggregatеd to allow a longer period for sentence modification is not persuasive. The Expectation of Finality
{¶ 35} I agree with the majority that Holdcroft had a legitimate expectation in the finality of his sentence for aggravated arson, but for different reasons. This court has previously stated that “there can be no reasonable, legitimate expectation of finality in [a void sentence].” State v. Simpkins,
{¶ 36} We have incorporated DiFranceseo’s reasoning, stating, “[W]hen the legislature has provided the government with a statutory right of appeal, ‘[t]he defendant * * * is charged with knowledge of the statute and its appeal prоvisions, and has no expectation of finality in his sentence until the appeal is concluded or the time to appeal has expired.’ ” (Emphasis added.) State v. Roberts,
{¶ 37} More than ten years after Holdcroft was originally sentenced, the state filed а motion under R.C. 2929.191 for the court to modify his sentence for aggravated arson to include the postrelease-control sanctions that had been omitted by the court. Holdcroft’s appeal time had run, and moreover, he had completed his ten-year prison term for aggravated arson. He therefore had a legitimate expectation in the finality of that sentence, and no additional punishment could be imposed without running afoul of the Double Jeopardy Clause as well as R.C. 2929.191.
Conclusion
{¶ 38} Postrelease-control sentencing errors may be corrected on direct appeal just as any other sentencing error may be corrected. In addition, the General Assembly has provided an alternative procedure under R.C. 2929.191 that expands the time for correction of errors in imposition of postrelease control until a prison term has been completed for the relevant offense. In this case, the error in imposing postrelease control was not appealed by the state. The state also did not seek to have the postrelease-control error corrected pursuant to R.C. 2929.191 until after Holdcroft had completed his prison term for aggravated arson. Because I disagree with the reasoning expressed in the majority opinion and would rely solely on R.C. 2929.191, I concur only in the judgment that vacates the imposition of five-years’ postrelease control on Holdcroft for his offense of aggravated arson.
Notes
. H.B. 137 also amended R.C. 2929.19 and 2967.28. Language added to these two sections provides that on or after July 11, 2006, a trial court’s failure to include postrelease control at sentencing or in
. The language allowing the correction of an illegal sentence at any time was eliminated from Fed.R.Crim.P. 35(a) by Pub.L. No. 98-473, 98 Stat.2015, eff. Nov. 1, 1987.
