DUPERRY v. Kirk

570 F. Supp. 2d 348, 2008 U.S. Dist. LEXIS 60806, 2008 WL 3288114

Opinion

ORDER ON MOTION TO STAY PENDING APPEAL

STEFAN R. UNDERHILL, District Judge.

On July 3, 2008, I granted petitioner Adam DuPerry’s petition for a writ for habeas corpus pursuant to 28 U.S.C. § 2254, concluding that his not guilty by reason of insanity (“NGRI”) plea was con stitutionally invalid because it was not entered knowingly, intelligently, or voluntarily. As a remedy, I vacated his NGRI plea and, because the order of confinement based on that plea no longer provided a valid legal basis for his confinement, I ordered his immediate release. In addition, I barred the state from re-prosecuting DuPerry unless it brought him to trial within forty-five days of that order. The respondent (hereinafter “the State”) complied with that order and released DuPerry, ending his twenty-plus years of involuntary confinement at the Whiting Forensic Division of Connecticut Valley Hospital. The State has now moved for a stay, pending appeal to the United States Court of Appeals for the Second Circuit, of the portion of the order barring the re-prosecution of DuPerry unless he is brought to trial within forty-five days (the “retrial deadline”). As explained more fully below, most of the reasons set forth by the State in support of its motion to stay are without merit. Nevertheless, because DuPerry has been released and because a retrial may be unnecessary, I will grant a stay of the portion of the order establishing a retrial deadline in the interest of preserving judicial resources. 1

I. Discussion

The State has moved for a stay of the retrial deadline, arguing that a stay is in the best interests of the state court system and the public at large and that appeal has a strong likelihood of success on the merits. I will discuss each point in turn.

A

The State first suggests that I deviated from the normal form of relief ordered in habeas proceedings by issuing an immediate release order rather than a “conditional release order.” Although it is true that many courts grant conditional release orders, the traditional form of habeas relief was unconditional, immediate release. See Randy Hertz & James S. Liebman, Federal Habeas Practice & Procedure §§ 33.1, 36.4d (5th ed.2005) (hereafter “Hertz & Liebman”); Rosa v. McCray, 2004 WL 2827638, at *7 (S.D.N.Y.2004) (noting that traditional habeas relief was an unconditional order of release). I did not grant unconditional release in the traditional sense, which would have barred the state from re-prosecuting DuPerry altogether. Rather, considering the exceptional circumstances of this case, I granted a form of conditional release that ordered DuPerry to be released from custody immediately, but which still permitted the State to seek his retrial within a specified period of time.

Furthermore, conditional release orders are most commonly used where the petitioner is seeking a type of relief that would not affect the underlying fact of the petitioner’s guilt or afford the petitioner immediate release, such as modification of the length or a condition of confinement, a declaration that one of two sentences being served consecutively is unconstitutional, or specific performance of a plea bargain. See Hertz & Liebman, § 33.4. DuPerry was not seeking that type of procedural relief, but rather challenged the validity of his NGRI plea, the basis for his confinement. Therefore, having vacated DuPerry’s NGRI plea as constitutionally invalid, immediate release was appropriate in this case because there was no longer a legal basis to keep him involuntarily confined. Additionally, immediate release was not an unreasonable remedy considering the significant period of time that DuPerry had been confined pursuant to an invalid waiver of his constitutional rights, which far exceeded the term of confinement in the State’s proffered plea bargain. As the United States Supreme Court stated in Hilton v. Braunskill, 481 U.S. 770, 775, 107 S.Ct. 2113, 95 L.Ed.2d 724 (1987), “[fjederal habeas corpus practice, as reflected by the decisions of this Court, indicates that a court has broad discretion in conditioning a judgment granting habeas relief.” Considering the unusual facts of this case, the relief granted can hardly be considered so “severe” that it necessitates a stay of its terms.

The State also raises the issue of the “public interest” in seeking a stay of the retrial deadline, citing DuPerry’s alleged conduct in connection with the underlying charges and a thoroughly unsupported claim that he presently “suffers from mental disease.” No facts about Du-Perry’s current mental state have ever been presented to this court, nor can I infer from DuPerry’s alleged conduct in 1988 that he remains a danger to the public at large. Even accepting as true the assertions made by the State, it is unclear how those facts support a stay of the retrial deadline. Moreover, considering that DuPerry was involuntarily confined for a period of over twenty years, far more time than what he would have served under the terms of the State’s plea bargain, it remains to be seen what interest any further term of incarceration would serve. Therefore, it is not appropriate for the State to argue, without any factual basis that DuPerry remains mentally ill or is still a threat to society, that this court should rely on those statements when considering whether to stay the retrial deadline.

B

As its second reason in support of a stay of the retrial deadline, the State argues that it is likely to succeed on the merits on the issue of the standard of review. The State insists that I incorrectly “refused” to apply the more deferential standard of review required by 28 U.S.C. § 2254(d), 2 which obligates a federal habeas court to give deference to a state court’s adjudication of federal constitutional claims if the state court reached the merits of those claims. However, that deferential standard does not apply, and the federal habeas court many consider the petitioner’s federal constitutional claims de novo, where the state court did not reach the merits of a particular claim. Torres v. Berbary, 340 F.3d 63, 68 (2d Cir.2003).

Therefore, in order for the State to succeed on the merits of this claim on appeal, it must demonstrate that the Connecticut Supreme Court actually reached the merits of DuPerry’s claim in Count One, the only claim I considered in granting his petition for habeas corpus. To support its claim, the State once again makes the argument that Count One and Count Two of DuPerry’s federal habeas petition actually raise the same claim. In Count One of his petition, DuPerry sought relief on the ground that his NGRI plea was not knowing, intelligent, and voluntary, and thus constitutionally invalid. DuPerry argued that, like a guilty plea, a NGRI plea acts as a waiver of fundamental constitutional rights and subjects the defendant to the possibility of confinement and, therefore, must be knowing, intelligent, and volun tary to be valid. In Count Two, DuPerry sought relief on the ground that the trial court failed to canvass him to ensure that his NGRI plea was made knowingly, intelligently, and voluntarily. The State once more insists that those two claims are actually the same, i.e., that DuPerry’s NGRI plea was not knowing, voluntary, and intelligent because the trial court failed to properly canvass him. However, the State’s argument would require holding that the Connecticut Supreme Court was wrong in Duperry v. Solnit (“Duperry I”), 261 Conn. 309, 803 A.2d 287 (2002), when it treated the two issues as distinct. That court reached only the merits of Du-Perry’s “failure to canvass” claim, Count Two in his federal habeas petition, and expressly declined to reach the issue whether the constitutional principle that requires guilty pleas be knowing, intelligent, and voluntary applies with equal force to NGRI pleas.

In Duperry I, the Connecticut Supreme Court reversed the lower court’s grant of habeas corpus on the ground that it had established and applied a new rule of constitutional criminal procedure in contravention of Teague v. Lane, 489 U.S. 288, 109 S.Ct. 1060, 103 L.Ed.2d 334 (1989), which prohibits courts from establishing or applying new constitutional rules in collateral proceedings, such as habeas corpus review. 261 Conn, at 324, 803 A.2d 287. Specifically, the court determined that the lower court “departed] from prior constitutional jurisprudence” when it concluded that the trial judge should have canvassed DuPerry to ensure he was pleading NGRI knowingly and voluntarily. Id. According to the court’s analysis, there was no established constitutional rule of criminal procedure that required trial judges to canvass defendants pleading NGRI at the time of DuPerry’s trial and neither of the two exceptions to Teague were applicable to DuPerry’s case. Id. The court thus concluded the Teague issue was dispositive and reversed the lower court on that ground. Id. at 324, 326, 803 A.2d 287. Significantly, the court expressly declined to reach the issue whether the constitutional principle that a guilty plea must be knowing, intelligent, and voluntary to pass constitutional muster applied equally to DuPerry’s NGRI plea. “[W]e do not address whether the substance of the habeas court’s holding was proper, i.e., whether the principle of Boykin v. Alabama, 395 U.S. 238, 242-44, 89 S.Ct. 1709, 23 L.Ed.2d 274 (1969), that due process requires a plea of guilty to be knowing and voluntary, applies equally to a plea of not guilty by reason of mental disease or defect.” 261 Conn, at 326 n. 7, 803 A.2d 287.

Whether the constitutional principle that requires guilty pleas be knowing, intelligent, and voluntary applies equally to NGRI pleas is precisely the issue that I addressed in my ruling. Therefore, because the Connecticut Supreme Court explicitly declined to reach that issue on the merits, I was not subject to section 2254(d)’s deferential standard of review. To claim that the deferential standard of review did apply is to ignore that the Connecticut Supreme Court considered that those issues were distinct. Briefly, it bears noting how the two claims are distinguishable. The Connecticut Supreme Court’s conclusion that the trial judge would not have felt constitutionally compelled to canvass DuPerry on his NGRI plea does not foreclose the possibility that, because NGRI pleas in non-adversarial proceedings operate as waivers of constitutional rights and subject defendants to the possibility of confinement in the same manner as guilty pleas, such pleas must be knowing, intelligent, and voluntary to be constitutionally valid. Whether or not a canvass was conducted does not conclusively establish compliance with the under lying constitutional principle that waivers of constitutional rights must be knowing, intelligent, and voluntary. Rather, conducting a plea canvass provides a record from which courts may evaluate whether the defendant’s waiver indeed met the underlying constitutional standard. The Connecticut Supreme Court’s determination in Duperry I that DuPerry could not benefit from the new canvass requirement does not preclude DuPerry from challenging the legality of his NGRI plea on the ground that the fundamental constitutional principle that his plea must be knowing, intelligent and voluntary had been violated. Therefore, the State has not shown that it is likely to succeed on this claim on appeal.

C

Next the State argues that I erred in giving the lower state court’s factual findings a presumption of correctness because they were actually legal conclusions, not factual findings. It further insists that the lower state court’s “legal conclusions” were based on the lack of a canvass only. There are several problems with the State’s analysis.

First, a determination that a plea was not made knowingly, intelligently, and voluntarily can be either a factual finding or a legal conclusion. In this particular case, Judge Spallone’s written ruling clearly demonstrates that he was making a factual finding, not a legal conclusion. First, there are many references to “the record,” indicating that Judge Spallone was making factual findings, e.g., “[tjhe record is devoid of any indication ....;” “[n]o where, in the record is any indication that the petitioner ‘knowingly’ or ‘voluntarily’ relinquished his right to a jury trial or that he was made aware of the resultant consequences of a plea of not guilty by reason of mental disease or defect;” [ujnder the circumstances presented by the record in this case, this court finds that the petitioner did not knowingly and voluntarily waive his constitutional right to a jury trial; nor did he understand the consequences of his plea of not guilty by reason of mental disease or defect, (emphasis added). In short, Judge Spallone found that DuPerry’s plea actually was not knowing, intelligent, or voluntary.

Next, there is no reason to believe that the factual finding that DuPerry actually did not understand the consequences of his decision to plead NGRI was made solely on the basis that the trial judge did not canvass DuPerry on his NGRI plea. Although Judge Spallone outlined the scope of what an adequate canvass should cover and found that the record did not reveal that there was such a canvass, his ultimate factual finding was that DuPerry did not plead NGRI knowingly, intelligently, or voluntarily. As discussed above, a judge is free to find on the basis of a silent record that the plea was not knowing, intelligent, or voluntary. It is not the canvass itself that renders a plea knowing, intelligent, and voluntary; rather the canvass produces a record from which a court may determine whether a defendant’s plea was knowing, intelligent, or voluntary.

Furthermore, even if Judge Spallone was making a legal conclusion, it would make no difference to the State’s possibility of success on the merits of its claim about the standard of review that I used in my ruling. Any such legal conclusion on the merits of the claim contained in Count One of DuPerry’s federal habeas petition represents the same conclusion that I reached, thus permitting my ruling to stand even under a more deferential standard of review. As discussed above, the Connecticut Supreme Court did not reach the merits of the claim in Count One, which left untouched any legal conclusion reached by Judge Spallone on the merits of that claim.

Even if, as the State claims, Judge Spallone reached a legal conclusion on the merits of DuPerry’s claim that the legal basis for his confinement was unconstitutional because the trial judge failed to canvass him on his NGRI plea, the State’s appeal lacks force. First, that is the claim that the Connecticut Supreme Court addressed on the merits, thus superseding Judge Spallone’s legal conclusion on that claim. Second, it still does not mean that I erred by applying a de novo standard of review to Count One of DuPerry’s petition. My conclusion that DuPerry’s NGRI was not constitutionally valid had nothing to do with whether the trial judge conducted a canvass or not. As explained above, I was considering a completely distinct claim from the one addressed by the Connecticut Supreme Court and therefore properly applied a de novo standard of review.

The State also suggests that the Connecticut Supreme Court made “factual findings” and that my review of those facts was in contravention of those findings. It must be noted that the Connecticut Supreme Court does not actually “make” factual findings; that is the role of the lower court that actually conducted the evidentiary hearing. See Smith v. Muellner, 283 Conn. 510, 535, 932 A.2d 382 (2007) (“On appeal, we do not retry the facts or pass on the credibility of witnesses____We afford great weight to the trial court’s findings because of its function to weigh the evidence and determine credibility.”).

The following is the passage selected by the State as evidence of the Connecticut Supreme Court’s “factual findings”:

In the present case, the habeas court found that [the petitioner’s attorney Dennis O’Toole] had provided the petitioner with all the information available to him regarding the petitioner’s plea options. The habeas court further found that the petitioner’s belief that he would spend less time in confinement by pleading not guilty and pursuing a mental health defense rather than by pleading guilty was based on the petitioner’s own conjecture and not on the advice of his attorney. The petitioner has made no argument that the habeas court’s factual determinations are clearly erroneous. We must assume, therefore, that O’Toole provided the petitioner with all the information available to him regarding the petitioner’s plea options.

Duperry I, 261 Conn, at 336, 803 A.2d 287 (emphasis added to the State’s selected passage). 3 A fair reading of this passage reveals that the Connecticut Supreme Court was referencing the lower court’s factual findings, not making its own; in addition, the quoted passage is actually in reference to the findings of the first habeas court on the issue of DuPerry’s ineffective assistance of counsel claim, not the second habeas court that directly addressed the issue whether the record showed that DuPerry pled NGRI knowingly, intelligently, and voluntarily.

D

The State finally claims that it has a likelihood of success on the merits because I never addressed how DuPerry could challenge his “acquittal” in a federal habeas proceeding. The State maintains that because DuPerry pled “not guilty by reason of insanity” he was technically acquitted and therefore cannot challenge that acquittal in a federal habeas proceeding. In support of that proposition, the State cites Archuleta v. Hedrick, 365 F.3d 644, 648 (8th Cir.2004). In Archuleta, the Eighth Circuit considered the pro se habeas petition of a federal inmate committed to the custody of the Federal Medical Center in Springfield, Missouri (“FMC Springfield”). Id. at 646. The petitioner had been found not guilty by reason of insanity of assaulting a federal official and had been committed to the custody of FMC Springfield following a period of conditional release. Id. The Eighth Circuit determined that, although inartfully pled, the petitioner was asserting, in part, that “he may establish his right to discharge by relitigating the initial finding that he was not guilty by reason of insanity.” Id. at 648. In other words, if permitted to relitigate the issue of his insanity he could prove he was not not guilty by reason of insanity. The court then went on to state that the petitioner “may not collaterally attack his decision to assert a successful insanity defense,” id., citing Curry v. Overholser, 287 F.2d 137, 139-40 (D.C.Cir. 1960), which rejected the petitioner’s challenge that the statute that permitted him to plead not guilty by reason of insanity in the first place was unconstitutional. In Archuleta, the Eighth Circuit went on to permit the petitioner to raise several other habeas claims attacking the legality of his confinement. 365 F.3d at 648-49.

Therefore, at most, the cases cited by the State preclude petitioners in the custody of a mental institution as a consequence of pleading NGRI from seeking to relitigate their eligibility to plead NGRI in the first place. DuPerry, however, is not using his habeas petition to relitigate the question of whether he should have been permitted to plead NGRI or that the statute that permitted him to plead NGRI was unconstitutional. Rather, he is challenging the legal basis for his confinement on the ground that he waived several constitutional rights without a full understanding of the consequences because his NGRI was not knowing, intelligent, or voluntary and, thus, was constitutionally invalid. To the extent that the Eighth Circuit’s decision can be interpreted as preventing inmates of mental institutions from using habeas corpus as a procedural mechanism to challenge the legality of their continued confinement — a reading that I do not endorse — I am happy to part ways with that decision. DuPerry is not challenging the fact that he was “acquitted,” he is challenging the constitutionality of his NGRI plea, and in turn, the legality of the commitment order that provided the legal basis for his involuntary confinement. Therefore, the State has not demonstrated a strong likelihood of success on the merits of its claim that NGRI acquittees cannot use habeas to challenge the legality of their involuntary confinement.

II. Conclusion

As noted above, most of the arguments asserted by the State in favor of a stay of the retrial deadline lack merit. I do acknowledge, however, that a stay of the requirement that the State bring Du-Perry to trial within forty-five days or be barred from further prosecution would serve the interest of preserving finite judicial resources. Regardless of the outcome of the State’s appeal, a retrial may be unnecessary 4 and therefore, I order the retrial deadline stayed, pending the outcome of the State’s appeal to the Second Circuit.

For the foregoing reasons, the State’s Motion to Stay That Part of Judgment Pertaining to Retrial of Petitioner (doc.# 52) is GRANTED.

It is so ordered.

1 . I acknowledge that I am ruling on the State’s motion to stay before DuPerry has had the opportunity to file his response. Because the retrial deadline is fast approaching, however, a quick ruling on the State’s motion was necessary to avoid possible prejudice. If Du-Perry feels that he is prejudiced by this order, he may file a motion for reconsideration.

2 . This is the standard of review the State refers to as the "AEDPA Deferential Standard of Review.” The current section 2254(d) was codified by the Antiterrorism and Effective Death Penalty Act of 1996, commonly known as AEDPA.

3 . I must further note that this passage has been taken out of context — it was not made in relation to whether DuPerry’s plea was knowing, intelligent, or voluntary, or even whether a canvass was constitutionally compelled, but rather was made in relation to DuPerry's ineffective assistance of counsel claim. The court concluded in the immediately following sentence that “[i]n light of this, it cannot be argued reasonably that O’Toole’s representation of the petitioner here fell below an objective standard of reasonableness.” Id. Furthermore, the factual findings the court is referring to are those of Judge Hodgson, who presided over DuPerry’s first habeas proceeding. Judge Hodgson was never asked to consider whether DuPerry's plea was knowing, intelligent, and voluntary, only his claims that he was no longer mentally ill and had received ineffective assistance of counsel. See Second Amended Petition of Aug. 20, 1997, at A9, and Judge Hodgson’s ruling, at A121-27.

4 . If the State prevails on its appeal, I will need to address the other arguments raised by DuPerry's petition, which I denied without prejudice as moot. If the State fails on its appeal, DuPerry should be afforded the opportunity to argue that a retrial is prohibited, an issue not decided in the order granting his petition.