dissenting.
The majority announces today that this case is moot because, in its view, no collateral consequences flow from respondents’ parole revocations, which were based on findings that respondents had violated the conditions of parole terms declared void by the courts below. I dissent from this holding because I believe it is contrary to this Court’s precedents and because it ignores the fact that the State of Illinois does attach collateral consequences to parole revocations, a fact recognized both in the State’s brief to the Court of Appeals on the issue of mootness and in state-court decisions in analogous cases.
HH
The majority recognizes that m habeas corpus challenges to criminal convictions, the case “is moot only if it is shown that there is no possibility that any collateral legal consequences will be imposed on the basis of the challenged conviction.” Sibron v. New York, 392 U. S. 40, 57 (1968). This Court has consistently refused to canvass state law to ascertain “the actual existence of specific collateral consequences,” and has presumed that such consequences exist. Id., at 55 *635(discussing United States v. Morgan, 346 U. S. 502 (1954), and Pollard v. United States, 352 U. S. 354 (1957)). See also Carafas v. LaVallee, 391 U. S. 234, 237-238 (1968).
Today, the majority finds the Carafas doctrine inapplicable, arguing that because respondents did not seek to set aside their convictions, their situation is analogous to that of a defendant who seeks habeas corpus review to correct a sentence already served. See North Carolina v. Rice, 404 U. S. 244 (1971) (per curiam). Had respondents served the allegedly void mandatory parole term without incident, I might agree that North Carolina v. Rice controls and join the majority’s conclusion that the consequence of the constitutional violation “expired of its own accord.” Ante, at 631. Here, however, respondents were found to have violated the conditions of their parole. Therefore, unlike the situation in North Carolina v. Rice, respondents seek more than a mere reduction in sentence after the sentence has been completed: they seek to have the parole term declared void, or expunged, in order to avoid the future consequences that attach to parole violations. If collateral consequences do attach to parole violations, both the State and respondents have a live interest in this Court’s review of the lower courts’ holdings that the alleged constitutional violations rendered the guilty pleas void and that respondents were entitled to specific performance of the pleas, in the form of a declaration that the mandatory parole terms were void and should be expunged.
The existence of a five controversy in this case turns on whether collateral consequences attach to parole violations. Because this determination involves a difficult question of state law, I believe that the doctrine of Sibron and Carafas should be applied. This doctrine avoids placing a federal court in the awkward position of determining questions of state law not directly before it. By presuming the existence of collateral consequences, federal courts are not required to predict the manner in which a State may use convictions or *636parole violations in future proceedings. An erroneous determination that collateral consequences do not attach not only injures the individuals challenging the constitutionality of the guilty pleas, but also hinders the State’s ability to use these violations in future proceedings. Today’s opinion is an unfortunate example of such an erroneous interpretation.
II
The majority’s decision is apparently based on a cursory examination of Illinois statutes. Finding no statutory civil disabilities, the majority glibly dismisses nonstatutory consequences as “discretionary decisions” that would remain whether or not the parole terms were declared void or expunged. Ante, at 632-633.1 This reasoning has no basis in *637Illinois law and appears to derive from nothing more than judicial intuition.
Several collateral consequences attach to parole violations under Illinois law.2 First, a sentencing judge may consider parole violations in aggravation of sentence. The majority makes the unwarranted assumption that declaring void the parole term upon which a violation is based has no effect because a sentencing judge would consider the conduct underlying the violation, and not the violation itself, in deciding whether to enhance a sentence. However, as the majority recognizes, there is no way for this Court to determine the basis for respondents’ parole revocation. Under Illinois law, the Prisoner Review Board is given substantial discretion in setting conditions of parole. See Ill. Rev. Stat., ch. 38, ¶ 1003-3-7 (Supp. 1980).3 Conditions of parole may prohibit *638conduct that is otherwise innocent and may affirmatively require the parolee to engage in specified work or rehabilitation programs. Parole may be revoked upon a finding that the parolee has violated any of these parole conditions. See Ill. Rev. Stat., ch. 38, ¶ 1003-3-9 (Supp. 1980); Illinois Prisoner Review Board, Rules Governing Parole 9-10, 13-16 (1979), 3 Ill. Register 144, 162-166 (1979). Therefore, conduct giving rise to a parole violation may be completely innocuous but for the fact that it was prohibited or required as a condition of parole, and it may be entirely irrelevant to a sentencing decision once the parole term is declared void.
Moreover, it is not clear under Illinois law whether a sentencing judge would consider the conduct underlying a parole violation, even if the conduct is not otherwise innocent, where the parole term itself is declared void. In a similar context, the Illinois appellate courts have held that trial courts may not consider a reversed conviction in aggravation of sentence, even where the court, in remanding for a new trial, noted that the evidence was sufficient to support the verdict beyond a reasonable doubt and the matter was never retried. See, e. g., People v. Chellew, 20 Ill. App. 3d 963, 313 N. E. 2d 284 (1974). Cf. People v. Wunnenberg, 87 Ill. App. 3d 32, 34, 409 N. E. 2d 101, 103 (1980). The Illinois courts have also held that review of probation revocation is not rendered moot merely because the defendant has served his entire sentence. See People v. Halterman, 45 Ill. App. 3d 605, 608, 359 N. E. 2d 1223, 1225 (1977) (challenge to probation revocation not moot because “the fact that the defendant has had his probation revoked might be submitted to another judge for his consideration in sentencing the defendant if he has the misfortune of again being convicted of some crime”). These cases do not conclusively demonstrate that a judge would not consider the conduct underlying the violation *639of a void parole term in aggravation of sentence. However, they cast serious doubt on the validity of the majority’s assumption to the contrary. Furthermore, the State argued to the Court of Appeals that the case was not moot because the State “is deeply interested in whether or not it can use the parole violation status of [respondents] for sentencing purposes should they ever again come into contact with the criminal justice system.” Additional Memorandum for Appellants in Nos. 78-1321, 78-1322, 78-1323, 78-1380 (CA7), p. 5 (Mem. to Court of Appeals). This argument at least implies that the State would not use this status for sentencing purposes after a court had declared the parole terms void.
Second, the majority completely overlooks an important collateral consequence that attaches to parole violations should the respondents ever have the misfortune of returning to prison. In rules promulgated by the Prisoner Review Board pursuant to Ill. Rev. Stat., ch. 38, ¶¶ 1003-3-1, 1003-3-2 (Supp. 1980), the State of Illinois has set forth fairly specific criteria upon which parole may be denied. See Illinois Prisoner Review Board, Rules Governing Parole (1979), 3 Ill. Register 144-169 (1979). The Rules provide in relevant part:
“V. BASIS FOR DENYING PAROLE
In accordance with statute, the Board shall not parole a candidate if it determines that:
“A. There is a substantial risk that the candidate will not conform to reasonable conditions of parole based on one or more of the following factors:
“1. Existence of prior adult felony convictions (mitigating as well as aggravating factors to be considered).
“2. An apparent pattern of aggressive or assaultive behavior (misdemeanor offenses also considered).
“3. Prior adult parole or probation violations within five years prior to the present offense.
“4. Refusal to be supervised on parole.
*640“5. No means of financial support or no place of residence. (Continuance not to exceed six months to seek resolution of problem.)
“6. A psychiatric examination determines the candidate is not likely to conform.” Illinois Prisoner Review Board, Rules Governing Parole 6 (1979), 3 Ill. Register 153 (1979) (emphasis added).
Under these rules, parole may be denied simply on the basis of a prior parole violation; the conduct underlying the parole violation is apparently irrelevant unless it falls within one of the other criteria listed in that section. We have no reason to assume that the conduct underlying respondents’ violations would fall within one of the other factors, or that the Prisoner Review Board would deny parole based on a parole violation notwithstanding the fact that the parole term had been declared void. In fact, the State argued to the Court of Appeals that the case was not moot because respondents “still have a substantial stake in ensuring that their parole terms are, indeed, expunged,” because the parole violations would be burdensome if respondents were ever again considered for parole. Mem. to Court of Appeals 5. See also United States ex rel. Howell v. Wolff, No. 78 C 951 (ND Ill. Aug. 9, 1978) (unpublished opinion of Judge Leighton, reprinted in App. to Mem. to Court of Appeals) (finding case not moot due to potential burden on future'parole decision from parole-violation status).
KH 1 — 4
Today s decision, in which the majority undertakes a cursory and misleading examination of state law, starkly demonstrates the wisdom of applying the doctrine of Carafas and Sibron to the determination whether a State attaches collateral consequences to parole violations. I would apply that doctrine, presume the existence of collateral consequences, and reach the merits of this case. Even if the doctrine of *641Carafas and Sibron does not apply, an examination of state law reveals that the majority is wrong in concluding that actual collateral consequences do not attach under state law; there are sufficient collateral consequences flowing from parole-violation status that both the State and the respondents have a live interest in this Court’s resolution of the constitutional question. Therefore, I dissent from the majority’s conclusion that this case is moot.
The majority makes a cryptic reference to the fact that respondents did not request the District Court to expunge or make any change in their records. Ante, at 633, n. 14. The failure to make this request is easily explained on several grounds and is irrelevant to the question whether this case is moot. The respondents did request that the District Court “expunge” the parole terms on which the violations were based. This “expungement” would have the effect of removing respondents’ parole-violation status and would relieve respondents of the collateral consequences flowing from this status. Any further “expungement” that respondents might obtain should be requested in future state proceedings. The State of Illinois has a very limited expungement procedure that would not cover the expungement to which the majority apparently refers. See Ill. Rev. Stat., ch. 38, ¶ 206-5 (Supp. 1980) (person, not convicted of any previous criminal offense, who is acquitted or released without conviction may petition the court for expungement of arrest records).
Furthermore, the State of Illinois has no procedure to expunge convictions that are later reversed or vacated on appeal, but this fact, or the failure of a habeas petitioner to request that a federal district court accord him relief that is unavailable under state law, would hardly render moot a ha-beas petition to set aside a conviction unconstitutionally obtained. The Illinois courts may not consider a reversed conviction in aggravation of sentence, despite the fact that the records of this conviction have not been officially “expunged.” See, e. g., People v. Wunnenberg, 87 Ill. App. 3d 32, 409 N. E. 2d 101 (1980); People v. Chellew, 20 Ill. App. 3d 963, 313 N. E. 2d 284 (1974).
Of course, the existence of express statutory civil disabilities is not a prerequisite to holding that a habeas challenge to a criminal conviction is not moot. See, e. g., Sibron v. New York, 892 U. S. 40, 54-57 (1968) (discussing Fiswick v. United States, 392 U. S. 211 (1946)); United States v. Morgan, 346 U. S. 502 (1954); Pollard v. United States, 352 U. S. 354 (1957)).
Paragraph 1003-3-7 provides:
“(a) The conditions of parole or mandatory supervised release shall be such as the Prisoner Keview Board deems necessary to assist the subject in leading a law-abiding life. The conditions of every parole and mandatory supervised release are that the subject:
“(1) not violate any criminal statute . . . ; and
“(2) refrain from possessing a firearm or other dangerous weapon.
“(b) The Board may in addition to other conditions require that the subject:
“(1) work or pursue a course of study or vocational training;
“(2) undergo medical or psychiatric treatment, or treatment for drug addiction or alcoholism;
“(3) attend or reside in a facility established for the instruction or residence of persons on probation or parole;
“(4) support his dependents;
“(5) report to an agent of the Department of Corrections;
“(6) permit the agent to visit him at his home or elsewhere to the extent necessary to discharge his duties . . .” (emphasis added).
*638See also Illinois Prisoner Review Board, Rules Governing Parole 9-12 (1979), 3 Ill. Register 158-160 (1979).