Thiersaint v. Commissioner of Correction

****************************************************** The ‘‘officially released’’ date that appears near the beginning of each opinion is the date the opinion will be published in the Connecticut Law Journal or the date it was released as a slip opinion. The operative date for the beginning of all time periods for filing postopinion motions and petitions for certification is the ‘‘officially released’’ date appearing in the opinion. In no event will any such motions be accepted before the ‘‘officially released’’ date. All opinions are subject to modification and technical correction prior to official publication in the Connecti- cut Reports and Connecticut Appellate Reports. In the event of discrepancies between the electronic version of an opinion and the print version appearing in the Connecticut Law Journal and subsequently in the Con- necticut Reports or Connecticut Appellate Reports, the latest print version is to be considered authoritative. The syllabus and procedural history accompanying the opinion as it appears on the Commission on Official Legal Publications Electronic Bulletin Board Service and in the Connecticut Law Journal and bound volumes of official reports are copyrighted by the Secretary of the State, State of Connecticut, and may not be repro- duced and distributed without the express written per- mission of the Commission on Official Legal Publications, Judicial Branch, State of Connecticut. ****************************************************** THIERSAINT v. COMMISSIONER OF CORRECTION—FIRST DISSENT PALMER, J., dissenting. I disagree with the majority’s conclusion that Padilla v. Kentucky, 559 U.S. 356, 130 S. Ct. 1473, 176 L. Ed. 2d 284 (2010), which held that the failure of defense counsel to advise a noncitizen client regarding the immigration consequences of plead- ing guilty establishes the performance prong of a claim of ineffective assistance of counsel under the sixth amendment, announced a new procedural rule within the meaning of Teague v. Lane, 489 U.S. 28, 109 S. Ct. 1060, 103 L. Ed. 2d 334 (1989), and therefore does not apply retroactively in habeas proceedings. Although I agree that we should continue to apply the Teague framework in determining whether a decision that rec- ognizes a constitutional rule of criminal procedure should be applied retroactively in habeas proceedings, I would conclude that Padilla did not announce a new rule because it was merely an application of the well established standard governing ineffective assistance of counsel claims under Strickland v. Washington, 466 U.S. 668, 687, 104 S. Ct. 2052, 80 L. Ed. 2d 674 (1984), and because, under Connecticut law, the obligations of counsel for the petitioner, Emmanuel Thiersant, were consistent with the court’s holding in Padilla. I con- clude, therefore, that the habeas court in the present case properly determined that the petitioner was deprived of his right to the effective assistance of coun- sel. Accordingly, I respectfully dissent.1 The relevant facts and procedural history, which are not disputed, may be summarized briefly as follows. The petitioner is a native of Haiti who moved to the United States in 1994, when he was fourteen years old. Shortly after his arrival, he was critically injured in a motor vehicle accident, and required the amputation of his right leg above the knee. Following the accident, the petitioner was hospitalized for eight months, and was administered a variety of drugs for his injuries. After leaving the hospital, he developed an addiction to crack cocaine. In 2006, the petitioner was arrested and charged with various narcotics offenses after mak- ing two $20 sales of crack cocaine to an undercover police officer. The petitioner was represented by Attor- ney John Imhoff. In 2007, the petitioner pleaded guilty to possession of cocaine with intent to sell in violation of General Statutes § 21a-277 (a), and, based on his conviction for that offense, he was subsequently ordered removed from the United States. Thereafter, the petitioner commenced the present habeas action, alleging that Imhoff had failed to advise him of the immigration consequences of his guilty plea in violation of Padilla v. Kentucky, supra, 559 U.S. 374. The habeas court determined that Padilla, which was decided after the petitioner’s conviction became final, applied retro- actively to the petitioner’s claim. On the merits of the petitioner’s claim, the habeas court agreed with the petitioner that he had been deprived of the effective assistance of counsel. The court first determined that Imhoff had performed defi- ciently because his representation fell below ‘‘an objec- tive standard of reasonableness.’’ Strickland v. Washington, supra, 466 U.S. 688. In support of this conclusion, the habeas court found that Imhoff had failed to advise the petitioner that his conviction under § 21a-277 (a) would constitute an aggravated felony for immigration law purposes and that, as a result of that conviction, he would be subject to mandatory detention by the United States Immigration and Customs Enforce- ment Agency upon the completion of his sentence. In addition, the court found that Imhoff had failed to advise the petitioner that his conviction would render him ineligible for almost all defenses to removal, virtu- ally assuring that he would be ordered removed from the United States and permanently barred from returning. According to the habeas court, those adverse immigration consequences of a conviction under § 21a- 277 (a) were sufficiently clear and definite that Imhoff had a duty to advise the petitioner about those conse- quences prior to his plea. The court further concluded that, at the time Imhoff represented the petitioner, a reasonably competent defense attorney would have provided such advice. Instead, Imhoff ‘‘gave the peti- tioner differing, unspecific and incorrect advice, all of which left room for the petitioner to believe that he could contest his removal,’’ and advised the petitioner to consult an immigration attorney for more specific advice. The court also concluded that reasonably com- petent counsel would have raised the petitioner’s immi- gration status during plea negotiations in an effort to obtain an alternative disposition that would not have resulted in a conviction of an aggravated felony, and Imhoff had failed to do so. In addition, the habeas court found that the petitioner was prejudiced by Imhoff’s deficient performance because, had the petitioner known that his conviction would subject him to near certain and permanent removal from the United States, he would have insisted on going to trial rather than pleading guilty. The habeas court found that this would have been a rational deci- sion based in large part on the conditions the petitioner would face in Haiti upon his return to that country, as well as the petitioner’s strong ties to the United States. The petitioner has been a lawful permanent resident of the United States for almost twenty years and currently lives with his longtime girlfriend, who suffers from sei- zures and often requires his assistance, and their young daughter. As a result of the injuries he suffered in the motor vehicle accident, the petitioner requires the use of either a prosthesis or a wheelchair. The habeas court credited evidence presented by the petitioner regarding the inhumane conditions in Haiti for criminal deportees, especially those who are disabled. Upon return to Haiti as a criminal deportee, the petitioner would be subject to indefinite detention in a cell measuring ten feet by ten feet, containing twenty to sixty other detainees, where temperatures reach 100 degrees and rodents, insects, and disease are rampant. Detainees do not have access to bathroom facilities or medical care, and many detainees become seriously ill or die due to the extreme conditions. Detainees also must rely on family members or fellow detainees for food and water because it is not provided by the government. Because the petitioner has no family in Haiti, he would have to rely on his fellow detainees or go without nourishment. Even if he were released from detention, the petitioner likely would face discrimination and harassment from Haitian citizens because of his disability and his status as a criminal deportee. As the habeas court observed, in light of these extraordinary circumstances, the petitioner’s decision to forgo the plea offer and proceed to trial would have been a reasonable one regardless of the strength of the state’s case against him because ‘‘he would have had nothing to lose’’ by taking his chances at trial. The habeas court therefore granted the petitioner’s petition for a writ of habeas corpus and ordered the petitioner’s conviction vacated so that he could stand trial on the charges. The respondent, the Commissioner of Correc- tion, appealed, claiming that the habeas court had improperly concluded that Padilla applied retroactively to the petitioner’s claim. I As the majority has explained, claims of ineffective assistance of counsel at the plea stage are governed by the standard set forth in Strickland v. Washington, supra, 466 U.S. 687–91, and Hill v. Lockhart, 474 U.S. 52, 106 S. Ct. 366, 88 L. Ed. 2d 203 (1985). Under Strick- land and Hill, a habeas petitioner must prove that ‘‘counsel’s representation fell below an objective stan- dard of reasonableness . . . [and] that there is a rea- sonable probability that, but for counsel’s errors, [the petitioner] would not have pleaded guilty and would have insisted on going to trial.’’2 (Citations omitted; internal quotation marks omitted.) Johnson v. Commis- sioner of Correction, 285 Conn. 556, 575–76, 941 A.2d 248 (2008). In Padilla v. Kentucky, supra, 559 U.S. 374, the United States Supreme Court concluded that advice regarding the immigration consequences of a guilty plea is not categorically removed from the sixth amend- ment’s protections, and that a criminal defense attorney may render constitutionally ineffective assistance by failing to advise a noncitizen client about such conse- quences. Subsequently, in Chaidez v. United States, U.S. , 133 S. Ct. 1103, 185 L. Ed. 2d 149 (2013), the court concluded that Padilla does not apply retroac- tively in federal habeas proceedings, so that a nonciti- zen whose conviction became final before the decision in Padilla was announced is not entitled to relief in federal habeas court. In reaching its conclusion, the court applied the retroactivity framework it previously had set forth in Teague v. Lane, supra, 489 U.S. 288. As the court explained, under Teague, when a decision announces a ‘‘new rule, a person whose conviction is already final may not benefit from the decision in a [federal] habeas or similar proceeding. Only when we apply a settled rule may a person avail herself of the decision on collateral review.’’3 (Footnote omitted; internal quotation marks omitted.) Chaidez v. United States, supra, 1107. The court concluded that Padilla announced a new rule and, consequently, declined to give it retroactive effect. Id., 1113. Of course, if we were bound to follow Chaidez, the respondent in the present case would be entitled to prevail on his appeal because, under Chaidez, Padilla does not apply retroactively. As the United States Supreme Court made clear in Danforth v. Minnesota, 552 U.S. 264, 128 S. Ct. 1029, 169 L. Ed. 2d 859 (2008), however, Teague applies only to federal habeas review, and states are therefore free to give broader retroactive effect to a constitutional rule as a matter of state law. In Danforth, the court explained that Teague’s ‘‘general rule of nonretroactivity [was justified] in part by refer- ence to comity and respect for the finality of state convictions,’’ and that ‘‘[f]ederalism and comity consid- erations are unique to federal habeas review of state convictions. . . . If anything, considerations of comity militate in favor of allowing state courts to grant habeas relief to a broader class of individuals than is required by Teague. And while finality is, of course, implicated in the context of state as well as federal habeas, finality of state convictions is a state interest, not a federal one. It is a matter that [s]tates should be free to evaluate, and weigh the importance of, when prisoners held in state custody are seeking a remedy for a violation of federal rights by their lower courts.’’ (Citation omitted; emphasis in original.) Id., 279–80. Accordingly, in determining whether to give retroac- tive effect to Padilla, we need not follow Teague, and may adopt our own retroactivity test for purposes of state law. As the majority observes, however, we pre- viously have adopted the Teague framework,4 and I agree generally that we should continue to adhere to that framework when determining whether a decision applies retroactively to cases that otherwise have pro- ceeded to final judgment. The Teague framework has significant advantages, in that it gives proper weight to the state’s important interest in ensuring the finality of convictions, is relatively straightforward to apply, and leads to consistent results. As the United States Supreme Court itself has expressly recognized, however, ‘‘[a] decision . . . that a new rule does not apply retroactively under Teague does not imply that there was no right and thus no violation of that right at the time of trial—only that no remedy will be provided in federal habeas courts.’’ Id., 291. Accordingly, when we determine whether to give a decision retroactive effect, we are not determining whether the constitutional right in question existed at the time of the alleged violation, but only whether to afford a remedy for a violation of that right. It necessar- ily follows, then, that when we decide not to give a rule retroactive effect, we are deciding that an undetermined number of constitutional violations will stand uncor- rected. This fact militates in favor of a more flexible or liberal application of the Teague framework when, as here, considerations of federalism and comity do not come into play. Furthermore, critics of the Teague framework have pointed out that the United States Supreme Court has given an exceedingly broad interpretation to the defini- tion of a new rule, thereby restricting significantly the availability of remedies for constitutional violations. As first articulated in Teague, a decision was deemed to announce a new rule if the result was not ‘‘dictated by precedent’’ existing at the time the defendant’s convic- tion became final. (Emphasis omitted.) Teague v. Lane, supra, 489 U.S. 301. Subsequently, the Supreme Court ‘‘has greatly expanded the meaning of what is ‘new’ to include results not ‘apparent to all reasonable jurists’ at the time.’’ Commonwealth v. Sylvain, 466 Mass. 422, 433, 995 N.E.2d 760 (2013), quoting Lambrix v. Sin- gletary, 520 U.S. 518, 527–28, 117 S. Ct. 1517, 137 L. Ed. 2d 771 (1997). Consequently, a decision is considered new for purposes of Teague ‘‘even when it is controlled or governed by prior law and is the most reasonable interpretation of that law, unless no other interpretation is reasonable.’’ Rhoades v. State, 149 Idaho 130, 138, 233 P.3d 61 (2010), cert. denied, U.S. , 131 S. Ct. 1571, 179 L. Ed. 2d 477 (2011), citing Butler v. McKel- lar, 494 U.S. 407, 415, 110 S. Ct. 1212, 108 L. Ed. 2d 347 (1990). Because this standard is so broad, ‘‘decisions defining a constitutional safeguard rarely merit applica- tion on collateral review.’’ Colwell v. State, 118 Nev. 807, 818, 59 P.3d 463 (2002), cert. denied, 540 U.S. 981, 124 S. Ct. 462, 157 L. Ed. 2d 370 (2003). Although the United States Supreme Court has deemed this approach to be appropriate in the context of determining whether to apply a rule retroactively on federal habeas review of state convictions, the same federalism and comity concerns are not implicated when a state determines whether to apply the rule for purposes of its own habeas proceedings. In fact, with respect to the state’s signifi- cant interest in finality, because state collateral review generally takes place sooner than federal collateral review, the adverse effect on that interest is less pro- nounced in the former context than it is in the latter.5 Thus, a number of states have recognized that the restrictive approach to retroactivity mandated by Teague and its progeny provides insufficient protection for criminal defendants seeking to vindicate constitu- tional rights in state habeas proceedings; see, e.g., Rhoades v. State, supra, 138–39; Commonwealth v. Syl- vain, supra, 433–35; Colwell v. State, supra, 818–20; an observation with which I agree. I therefore believe that the decision of whether ‘‘to give retroactive effect to a rule of law [under the Teague framework] should reflect independent judgment, based upon the concerns of this [c]ourt and the uniqueness of our state, our [c]onstitu- tion, and our long-standing jurisprudence.’’ (Internal quotation marks omitted.) Rhoades v. State, supra, 139. Applying the Teague framework to the present case, and with the foregoing considerations in mind, I would conclude that Padilla did not announce a new constitu- tional rule. In reaching this conclusion, I agree with Justice Sotomayor’s dissent in Chaidez, in which she explains that Padilla did not announce a new rule under Teague because it was merely an application of the rule set forth in Strickland. Chaidez v. United States, supra, 133 S. Ct. 1114. Furthermore, Connecticut case law and practice at the time Padilla was decided support rather than foreclose this conclusion and, contrary to the con- cern expressed by the majority, applying Padilla retro- actively would not open the floodgates to petitioners seeking relief for alleged violations of Padilla. As the majority opinion explains, the court in Chaidez acknowledged that, ‘‘[w]here the beginning point of our analysis is a rule of general application, a rule designed for the specific purpose of evaluating a myriad of factual contexts, it will be the infrequent case that yields a result so novel that it forges a new rule, one not dictated by precedent.’’ (Internal quotation marks omitted.) Id., 1107. The court further acknowledged that, because Strickland itself is a rule of general application, ‘‘gar- den-variety applications of the test in Strickland . . . for assessing claims of ineffective assistance of counsel do not produce new rules.’’ Id. According to the court, however, ‘‘Padilla did something more’’ than merely apply Strickland to a new factual context. Id., 1108. Specifically, the court explained, ‘‘[b]efore deciding if failing to provide [advice regarding immigration conse- quences] fell below an objective standard of reasonable- ness, Padilla considered a threshold question: Was advice about deportation categorically removed from the scope of the [s]ixth [a]mendment right to counsel because it involved only a collateral consequence of a conviction, rather than a component of the criminal sentence?’’ (Internal quotation marks omitted.) Id. As the court further explained, it had never decided whether the sixth amendment requires attorneys to advise their clients regarding the collateral conse- quences of a conviction, and state and lower federal courts had been nearly unanimous in concluding that it does not. Id., 1108–1109. According to the court, this meant that its decision in Padilla was not ‘‘ ‘dictated’ ’’ by precedent, and, therefore, that it announced a new rule. (Emphasis omitted.) Id., 1110. In dissent, Justice Sotomayor, joined by Justice Gins- burg, argued that Padilla did not announce a new rule because it merely applied ‘‘the existing rule of Strick- land . . . in a new setting, the same way the [c]ourt has done repeatedly in the past: by surveying the relevant professional norms and concluding that they unequivo- cally required attorneys to provide advice about the immigration consequences of a guilty plea.’’ Id., 1114. Although, as I have indicated, the United States Supreme Court has taken a broad view of what consti- tutes a new rule, it has also stated that ‘‘a case does not announce a new rule [when] it [is] merely an applica- tion of the principle that governed a prior decision to a different set of facts.’’ (Emphasis in original; internal quotation marks omitted.) Id., 1107, quoting in part Teague v. Lane, supra, 489 U.S. 307. Thus, whether a rule is new ‘‘depends in large part on the nature of the rule. If the rule in question is one which of necessity requires a case-by-case examination of the evidence, then we can tolerate a number of specific applications without saying that those applications themselves cre- ate a new rule.’’ Wright v. West, 505 U.S. 277, 308, 112 S. Ct. 2482, 120 L. Ed. 2d 225 (1992) (Kennedy, J., concur- ring). In other words, ‘‘when all [the court does] is apply a general standard to the kind of factual circumstances it was meant to address, [it] will rarely state a new rule for Teague purposes.’’ Chaidez v. United States, supra, 1107. Padilla was the quintessential application of a gen- eral standard to a different set of facts. As Justice Soto- mayor explained in her dissent, the standard for determining deficient performance under Strickland is ‘‘simply reasonableness under prevailing professional norms . . . [which] takes its content from the stan- dards by which lawyers judge their professional obliga- tions . . . and those standards are subject to change.’’ (Citations omitted; internal quotation marks omitted.) Id., 1114. Put another way, the evolving nature of profes- sional norms means that representation that satisfies the sixth amendment one year may not do so the next, but this does not mean that the sixth amendment stan- dard itself changes. Rather, although the Strickland reasonableness standard is pegged to a benchmark that evolves over time, the rule remains the same. ‘‘That is why, despite the many different settings in which it has been applied, [the court had] never found that an application of Strickland resulted in a new rule.’’ Id., 1114–15 (Sotomayor, J., dissenting). In Padilla, the court merely applied the standard set forth in Strickland and concluded that, at the time of the petitioner’s conviction, prevailing professional norms required attorneys to advise their clients of the immigra- tion consequences of pleading guilty, depending on the severity and certainty of those consequences. Padilla v. Kentucky, supra, 559 U.S. 366–69. The court first surveyed the radical changes in immigration law from the early 1900s—when Congress first authorized removal of noncitizens convicted of certain crimes, but also gave state and federal judges the power to make binding recommendations at sentencing to prevent removal—through the Illegal Immigration Reform and Immigrant Responsibility Act of 1996, 8 U.S.C. § 1101 et seq., which made removal ‘‘practically inevitable’’ for noncitizens who commit an eligible offense. See id., 360–64; see also Chaidez v. United States, supra, 133 S. Ct. 1116 (Sotomayor, J., dissenting). The court then explained that professional norms had evolved to respond to these changes in immigration law, such that ‘‘[a]uthorities of every stripe—including the American Bar Association, criminal defense and public defender organizations, authoritative treatises, and state and city bar publications—universally require defense attorneys to advise as to the risk of deportation consequences for non-citizen clients.’’ (Internal quotation marks omitted.) Chaidez v. United States, supra, 1116 (Sotomayor, J., dissenting), quoting Padilla v. Kentucky, supra, 367. Thus, when the court in Padilla recognized a shift in defense counsel’s obligations under the sixth amend- ment, it was because prevailing professional norms had evolved in response to changes in immigration law, not because the court viewed the sixth amendment in a substantively different way. In other words, ‘‘[b]oth before Padilla and after, counsel was obligated to fol- low the relevant professional norms. It was only because those norms reflected changes in immigration law that Padilla reached the result it did, not because the [s]ixth [a]mendment right had changed at all.’’ Chaidez v. United States, supra, 1117 (Sotomayor, J., dissenting). Like the court in Chaidez, the majority in the present case relies heavily on the characterization of Padilla as having answered a ‘‘threshold’’ question about the scope of the sixth amendment. I believe that this view of Padilla fails to account for the fact that the court in Padilla answered that question simply by applying the Strickland standard to the facts of the case. As Justice Sotomayor explained in her dissent in Chaidez, the court in Padilla expressly declined to address the distinction between direct and collateral consequences because ‘‘deportation has a ‘close connection to the criminal process,’ and is ‘uniquely difficult to classify as either a direct or a collateral consequence.’ ’’ Id., 1117, quoting Padilla v. Kentucky, supra, 559 U.S. 366. Prior to Padilla, the court had never decided whether counsel’s duty to provide advice concerning the collat- eral consequences of a conviction is within the scope of the sixth amendment, and it still has not done so because Padilla expressly refused to address that ques- tion. Because the court concluded that immigration consequences are not collateral, the court ultimately could decide that the sixth amendment does not require attorneys to advise their clients regarding the collateral consequences of a conviction without disturbing the holding of Padilla. Thus, the ‘‘chink-free wall between direct and collateral consequences’’; Chaidez v. United States, supra, 1110; remains unbreached. The fact that the court in Padilla first addressed whether the sixth amendment applied to the claim before it was due to the nature of the decision the court was reviewing and the question for which it had granted review, rather than the nature of the sixth amendment inquiry. The Supreme Court of Kentucky had concluded that deportation was a collateral consequence of a con- viction that fell outside the scope of the sixth amend- ment and, therefore, that the appellee was not entitled to an evidentiary hearing on his ineffective assistance claim. Commonwealth v. Padilla, 253 S.W.3d 482, 484–85 (Ky. 2008). As Justice Sotomayor observes in her dissent, the United States Supreme Court rejected that conclusion simply by applying the Strickland stan- dard. Chaidez v. United States, supra, 133 S. Ct. 1114. Thus, although the court had to address whether immi- gration consequences fell outside the scope of the sixth amendment because of the Supreme Court of Ken- tucky’s disposition of the case, it answered that ques- tion by applying Strickland to determine the obligations of defense counsel under the facts of the case, just as it would for any other ineffective assistance of coun- sel claim. That the court in Padilla did not announce a new rule is further supported by the fact that, in addressing a claim under Padilla, a habeas court merely applies the Strickland test as it would for any other ineffective assistance claim. In a standard Strickland-Hill case in which the petitioner claims that his attorney provided ineffective assistance at the plea stage, the petitioner must demonstrate that his attorney’s advice was objec- tively unreasonable, such that it ‘‘ ‘fell below an objec- tive standard of reasonableness’ ’’; Hill v. Lockhart, supra, 474 U.S. 57; and that, ‘‘but for counsel’s errors, he would not have pleaded guilty and would have insisted on going to trial.’’ Id., 59. This is exactly the inquiry that a habeas court makes when addressing a claim under Padilla. See Padilla v. Kentucky, supra, 559 U.S. 366–69 (explaining that performance of criminal defense attorney when advising noncitizen client is to be judged by Strickland reasonableness standard, and counsel’s duty in such cases depends on, inter alia, clarity of potential immigration consequences client faces). As in the present case, a petitioner seeking to prevail on a claim under Padilla must show that a reasonably competent attorney would have advised his client regarding the immigration consequences of plead- ing guilty, and that he would not have pleaded guilty had his attorney provided such advice. Thus, the rule to be applied when a petitioner brings a claim under Padilla is the rule set forth in Strickland and Hill. The habeas court does not apply a ‘‘Padilla rule’’ because the court in Padilla did not establish a rule—it merely applied the rule established by Strickland and Hill. I recognize that the majority in Chaidez relied on the fact that, prior to Padilla, the overwhelming majority of courts to address the question concluded that immi- gration consequences are collateral and therefore out- side the scope of the sixth amendment.6 In Connecticut, however, the obligations of defense counsel recognized in Padilla were already required under state law and prevailing professional norms at the time that decision was issued. Contrary to the majority’s assertions, there was no binding precedent from this court or the Appel- late Court at that time holding that advice regarding immigration consequences was outside the scope of the sixth amendment right to the effective assistance of counsel.7 All of the case law on which the majority relies addresses what is required of the trial court when canvassing a defendant to ensure that a plea is voluntary under the fifth amendment. See State v. Malcolm, 257 Conn. 653, 662, 778 A.2d 134 (2001) (‘‘only substantial compliance [with General Statutes § 54-1j] is required . . . in order to ensure that the plea is voluntary’’ [foot- note omitted]); State v. Andrews, 253 Conn. 497, 500, 513–14, 752 A.2d 49 (2000) (trial court’s failure to advise regarding parole ineligibility did not render plea invol- untary); State v. Irala, 68 Conn. App. 499, 520, 792 A.2d 109 (trial court’s failure to advise regarding specific immigration consequences does not render plea invol- untary), cert. denied, 260 Conn. 923, 797 A.2d 519, cert. denied, 537 U.S. 887, 123 S. Ct. 132, 154 L. Ed. 2d 148 (2002). As the majority recognizes in declining to address the petitioner’s alternative claim that his plea was involuntary in violation of his right to due process, whether a plea is voluntary under the fifth amendment is a distinct issue from whether defense counsel provided ineffective assistance under the sixth amendment.8 See Lafler v. Cooper, U.S. , 132 S. Ct. 1376, 1390, 182 L. Ed. 2d 398 (2012) (‘‘[a]n inquiry into whether the rejection of a plea is knowing and voluntary . . . is not the correct means by which to address a claim of ineffective assistance of counsel’’). Consequently, our recognition that trial courts are not constitutionally required to ‘‘instruct defendants on the intricacies of immigration law’’; State v. Malcolm, supra, 663; did not, as the majority maintains, have any bearing on defense counsel’s obligations under the sixth amendment. Moreover, some Connecticut cases decided prior to Padilla may be read to suggest that the sixth amend- ment does require criminal defense attorneys to advise their clients regarding immigration consequences. For example, in State v. Irala, supra, 68 Conn. App. 500–501, the defendant sought to withdraw her pleas, claiming, inter alia, that her trial counsel was ineffective for failing to advise her regarding the immigration consequences of her conviction. The trial court denied her motions and, in affirming the trial court’s judgments, the Appel- late Court concluded that the defendant’s attorney had not performed deficiently because he had discussed the immigration consequences of the pleas with the defendant. Id., 506, 526–27. Thus, the Appellate Court implicitly accepted that advice regarding immigration consequences did fall within the scope of the sixth amendment, but concluded that counsel in that case met those obligations. Id., 527. At least one habeas court prior to the decision in Padilla found that an attorney had performed defi- ciently by providing inaccurate advice regarding the immigration consequences to his client of pleading guilty. In Durant v. Coughlin, Superior Court, judicial district of Ansonia-Milford at Milford, Docket No. CV- 99-066532 (July 12, 1999), the court determined that the petitioner’s attorney performed deficiently under the sixth amendment by erroneously advising the petitioner that he would not be deported because he was married to a United States citizen. The court determined, how- ever, that the petitioner failed to prove that he was prejudiced by counsel’s erroneous advice. See also Dawkins v. Armstrong, Superior Court, judicial district of New London, Docket No. CV-552015 (May 30, 2001) (addressing petitioner’s claim on merits but concluding that petitioner knew he would likely be deported, and, in any event, that petitioner was not prejudiced). Thus, the majority is incorrect that, at the time Padilla was decided, the courts of this state held that advice regard- ing immigration consequences is outside the scope of the sixth amendment. Although our case law had not yet definitively answered the question that the court resolved in Padilla regarding the scope of the sixth amendment, Connecti- cut law was consistent with the result in that case. As Justice Eveleigh observes, Connecticut frequently has been ahead of our sister states and the federal govern- ment in recognizing the right to counsel, having guaran- teed that right long before it was formally incorporated into our state constitution in 1818. In 1917, Connecticut was the first state to establish a public defender system, decades before the United States Supreme Court, in Gideon v. Wainwright, 372 U.S. 335, 83 S. Ct. 792, 9 L. Ed. 2d 799 (1963), recognized that the federal constitu- tion requires states to provide counsel to indigent crimi- nal defendants. In addition to recognizing the paramount importance of the right to counsel generally, this court has specifically held that state law guarantees the right to the effective assistance of counsel even where it is not required by the federal constitution, ensuring that Connecticut residents are not stripped of important rights without competent representation. See Lozada v. Warden, 223 Conn. 834, 838, 842–43, 613 A.2d 818 (1992) (recognizing right to effective assistance of counsel in habeas proceedings); cf. State v. Anony- mous, 179 Conn. 155, 159–60, 425 A.2d 939 (1979) (statu- tory right to counsel in termination of parental rights proceeding implicitly includes right to effective assis- tance of counsel). The criminal defense bar in Connecticut has long been aware that noncitizen defendants often give great weight to immigration consequences when deciding whether to accept or to reject a guilty plea. As the habeas court found, prevailing professional norms and statutory law at the time of the petitioner’s plea required criminal defense attorneys to advise their clients regard- ing the immigration consequences of pleading guilty. The habeas court credited the testimony of the petition- er’s expert witnesses, Attorney Anthony D. Collins and Attorney Christopher Caldwell, who both testified regarding prevailing professional norms in Connecticut at the time the petitioner pleaded guilty in 2007. Specifi- cally, Collins and Caldwell both testified that a reason- ably competent defense attorney representing a noncitizen in 2007 had a duty to avoid allowing her client to be convicted of an aggravated felony ‘‘at all costs,’’ to explain to her client the immigration conse- quences of such a conviction, and to seek an alternate disposition through plea negotiations. Moreover, as dis- cussed by Justice Eveleigh in greater detail, training available to criminal defense attorneys at that time pro- vided detailed guidance on representing noncitizen defendants in Connecticut, suggesting that the defense bar was well aware of the need to advise such clients regarding the immigration consequences of a con- viction. That prevailing professional norms were consistent with the holding in Padilla is further supported by Con- necticut statutory law, which has long recognized the importance of ensuring that criminal defendants under- stand that pleading guilty may carry immigration conse- quences. The legislature enacted § 54-1j in 1982, requiring trial court judges to advise defendants that pleading guilty may carry the risk of deportation; see Public Acts 1982, No. 82-177; and amended it in 2003 to require that judges provide defendants the opportu- nity to discuss any potential immigration consequences with their counsel before the court may accept a guilty plea. See Public Acts 2003, No. 03-81, § 1. The 2003 amendment is particularly persuasive proof of prevail- ing professional norms at the time of the petitioner’s plea. Even if professional norms had not yet evolved to the point that defense counsel recognized their obli- gation to advise their clients regarding immigration con- sequences prior to 2003, that amendment surely put the defense bar on notice of that duty. Thus, although no Connecticut case had yet recognized that the sixth amendment required defense counsel to advise nonciti- zen defendants regarding the immigration conse- quences of a conviction, our statutory law and the actual practices of the legal community were fully in accord with the dictates of Padilla at least as early as 2003. Finally, the finality concerns raised by the majority that traditionally counsel against applying a decision retroactively on collateral review are not implicated in the present case because the number of noncitizens who would be eligible for relief in Connecticut is lim- ited, and the number who would actually be able to obtain relief is further limited by the Strickland stan- dard. As the petitioner has underscored, a decision to apply Padilla retroactively in state habeas proceedings would only lead to a new trial for those who, like the petitioner: (1) are noncitizens; (2) have pleaded guilty to an offense that would result in their removal; (3) whose convictions became final before 2010; (4) have not yet been deported; (5) were not advised that plead- ing guilty would subject them to removal; (6) pleaded guilty at a time when a reasonably competent attorney would have provided such advice; (7) would not have pleaded guilty had they been so advised; and (8) can prove that such a decision would have been a rational one under the circumstances.9 These requirements will make it exceedingly difficult for a petitioner to prevail under Padilla on collateral review. The Strickland standard itself also would limit the number of petitioners entitled to relief, adequately pro- tecting the finality of convictions and preventing a flood of overturned convictions. In determining whether an attorney’s performance was constitutionally deficient, Strickland expressly instructs courts to ‘‘eliminate the distorting effects of hindsight’’ and to ‘‘evaluate the conduct from counsel’s perspective at the time.’’ Strick- land v. Washington, supra, 466 U.S. 689. This means that whether an attorney’s advice was constitutionally deficient depends on whether the advice conformed to professional norms at the time of the plea. While it is clear that professional norms in Connecticut required defense counsel to advise their clients about immigra- tion consequences at the time the petitioner pleaded guilty in 2007, the claims of noncitizens who pleaded guilty prior to that date would be judged by the profes- sional norms as of the date of their pleas. In addition, under Strickland’s prejudice prong, a defendant must prove that, if he had been advised about the risk of deportation, ‘‘he would not have pleaded guilty and would have insisted on going to trial’’; Hill v. Lockhart, supra, 474 U.S. 59; and that ‘‘a decision to reject the plea bargain would have been rational under the circum- stances.’’ Padilla v. Kentucky, supra, 559 U.S. 372. This difficult task is especially challenging in cases alleging a violation of Padilla because the petitioner must con- vince the habeas court that he would have insisted on a trial even though he likely would face a substantially greater sentence upon conviction, and then would still be subject to deportation after serving that sentence. As the court observed in Padilla, ‘‘[t]he nature of relief secured by a successful collateral challenge to a guilty plea—an opportunity to withdraw the plea and proceed to trial—imposes its own significant limiting principle: [t]hose who collaterally attack their guilty pleas lose the benefit of the bargain obtained as a result of the plea. Thus, a different calculus informs whether it is wise to challenge a guilty plea in a habeas proceeding because, ultimately, the challenge may result in a less favorable outcome for the [petitioner] . . . .’’ (Empha- sis in original.) Id., 372–73. For these reasons, it is highly unlikely that applying Padilla retroactively will lead to an appreciable number of new trials. The refusal of the United States Supreme Court to apply Padilla retroactively in federal habeas proceed- ings is understandable, given the much larger pool of noncitizens to whom such a decision would apply and the comity considerations necessarily implicated by any such decision. Those concerns, however, simply do not apply to our decision whether to apply Padilla retroac- tively. As discussed previously, moreover, prevailing professional norms and statutory law in Connecticut required defense attorneys to advise their clients regarding immigration consequences long before the court in Padilla formally recognized that requirement under the sixth amendment. I would conclude, there- fore, that Padilla did not announce a new rule under Connecticut law, and that the habeas court properly applied Padilla retroactively to the petitioner’s claim. II I now turn to the merits of the petitioner’s ineffective assistance claim. In light of its findings, which are sup- ported by the record, the habeas court properly found that Imhoff’s performance was objectively unreason- able. The habeas court found that, at the time the peti- tioner pleaded guilty, a reasonably competent attorney would have been aware that possession of narcotics with intent to sell in violation of § 21a-277 (a) consti- tutes an aggravated felony. The habeas court further found that a reasonably competent attorney in Imhoff’s position would have advised the petitioner that, upon conviction of an aggravated felony, he would be subject to mandatory removal from the United States, ineligible for virtually all defenses to removal, and permanently barred from returning to the United States. Consistent with Padilla, the habeas court noted that, because the law regarding the immigration consequences of the peti- tioner’s conviction was ‘‘clear and succinct,’’ Imhoff had a duty to provide the petitioner with correct advice regarding those consequences. See Padilla v. Kentucky, supra, 559 U.S. 368–69. The habeas court further found that Imhoff failed to meet that standard because he ‘‘was unaware of the specific consequences of the petitioner’s plea in this case and as a result was unable to, and did not, provide clear and accurate advice.’’ Rather than correctly advising the petitioner that a conviction under § 21a-277 (a) would constitute an aggravated felony, virtually assuring that he would be permanently removed from the United States, Imhoff gave the peti- tioner vague advice that did not impress upon him the grave and certain nature of the immigration conse- quences he faced. As the habeas court also stated, ‘‘[a]dvising the petitioner, who was indigent and incar- cerated, to obtain and consult with an immigration attorney, when he had no ability or resources to do so,’’ fell below the standard expected of reasonably competent counsel under the sixth amendment. The habeas court also found that this is one of the rare cases in which the evidence establishes that, had the petitioner been advised about the immigration con- sequences of his guilty plea, he would have insisted on going to trial, and that such a decision would have been ‘‘rational under the circumstances.’’ Padilla v. Ken- tucky, supra, 559 U.S. 372. The petitioner, who has been a permanent resident of the United States for almost twenty years, has strong family ties to this country in general and to Connecticut in particular, and no family connections in Haiti whatsoever. Although this may be true of many noncitizens—and alone, likely to be insufficient to satisfy the prejudice prong of the Strick- land-Hill test—the petitioner’s case is compelling given the extremely harsh conditions that he would face upon his return to Haiti. As the habeas court found, criminal deportees to Haiti are subject to ‘‘deplorable and inhu- mane conditions’’ that include being held in an over- crowded cell where disease is rampant and temperatures reach 100 degrees, with no bathroom facilities or access to medical care. As deplorable as the conditions are in Haiti for criminal deportees gener- ally, the petitioner likely would be treated even more harshly because of his disability, as ‘‘disabled persons . . . are treated as outcasts in prison and in society generally’’ and ‘‘[a]n amputee, like the [petitioner], would not be provided a wheelchair or prosthesis in the holding cells, or any medical treatment or medica- tions.’’ It is no wonder, then, that the habeas court credited the petitioner’s testimony that ‘‘he would have risked spending significantly more time in jail in this country rather than be deported to Haiti.’’ Although it ordinarily is well-nigh impossible for a petitioner to prove that it would have been objectively reasonable to forgo a favorable plea agreement that offers a dramatically reduced jail sentence and insist on a trial when, as in the present case, the state had overwhelming evidence of guilt—especially in view of the fact that the petitioner is subject to mandatory deportation after serving his sentence—the petitioner’s case presents extraordinary circumstances that justify affording him relief. I would conclude, therefore, that the habeas court properly determined that the peti- tioner was deprived of his right to the effective assis- tance of counsel. Accordingly, I respectfully dissent. 1 I therefore agree with Justice Eveleigh’s conclusion that Padilla applies retroactively in state habeas proceedings, and I also agree with a portion of his analysis and reasoning. I am concerned, however, that the approach he advocates would, in practice, lead to near universal retroactivity for all constitutional rules, and that a new trial will be required in every such case, no matter when the conviction was obtained. I therefore am unable to agree that we should apply a constitutional rule retroactively whenever ‘‘the fundamental fairness of a trial or plea is seriously diminished without the rule . . . .’’ Although this test seems sensible in theory, the concept of ‘‘fundamental fairness’’ is so amorphous that virtually all constitutional rules of criminal procedure pertaining to a criminal trial or plea may be said to implicate ‘‘fundamental fairness’’ in one way or another. Insofar as the vast majority of such rules would be subject to retroactive applicability under Justice Eveleigh’s test, I do not believe that the test takes sufficient account of the state’s significant interest in finality. It bears emphasis, moreover, that new substantive rules invariably apply retroactively because they ‘‘nec- essarily carry a significant risk that a defendant stands convicted of an act that the law does not make criminal or faces a punishment that the law cannot impose upon him,’’ whereas only procedural rules are subject to the Teague retroactivity test because ‘‘[t]hey do not produce a class of persons convicted of conduct the law does not make criminal, but merely raise the possibility that someone convicted with use of the invalidated procedure might have been acquitted otherwise.’’ (Internal quotation marks omitted.) Schriro v. Summerlin, 542 U.S. 348, 352, 124 S. Ct. 2519, 159 L. Ed. 2d 442 (2004). As I explain hereinafter, I believe that we should continue to follow the Teague framework, including the important exceptions delineated by the United States Supreme Court; see footnote 3 of this dissenting opinion; because that framework gives due weight to the state’s legitimate interests and leads to consistent results. As I further explain, however, I would conclude that Padilla applies retroactively in habeas proceedings in this state because it did not announce a new rule. 2 The court in Strickland set forth the general standard for establishing an ineffective assistance of counsel claim, requiring a petitioner to prove both that trial counsel performed deficiently and that, ‘‘but for counsel’s unprofessional errors, the result of the proceeding would have been differ- ent.’’ Strickland v. Washington, supra, 466 U.S. 694. In Hill, the court held that the Strickland standard applies to claims of ineffective assistance at the plea stage, but explained that the prejudice prong requires a petitioner to prove that, ‘‘but for counsel’s errors, he would not have pleaded guilty and would have insisted on going to trial.’’ Hill v. Lockhart, supra, 474 U.S. 59. 3 The court in Teague recognized two exceptions to the general rule of nonretroactivity for new constitutional rules. See Teague v. Lane, supra, 489 U.S. 307 (‘‘First, a new rule should be applied retroactively if it places certain kinds of primary, private individual conduct beyond the power of the criminal law-making authority to proscribe. . . . Second, a new rule should be applied retroactively if it requires the observance of those proce- dures that . . . are implicit in the concept of ordered liberty.’’ [Citation omitted; internal quotation marks omitted.]). Because I would conclude that Padilla did not announce a new rule, those exceptions are not relevant to my analysis. 4 I note, however, that we have never addressed the question of whether to give retroactive effect to a particular decision where, as in the present case, the United States Supreme Court has previously ruled that the decision does not apply retroactively on federal habeas review. In fact, we have only actually conducted a Teague analysis in one case, Duperry v. Solnit, 261 Conn. 309, 318–24, 803 A.2d 287 (2002), wherein we reversed the judgment of the habeas court on the ground that it announced a new rule in a collateral proceeding. In the remaining cases on which the majority relies, we did not actually apply the Teague framework at all. See State v. Payne, 303 Conn. 538, 549–50, 550 n.10, 34 A.3d 370 (2012) (adopting new approach to joinder of criminal trials under rules of practice and noting that it will not apply retroactively on collateral review); Johnson v. Warden, 218 Conn. 791, 796– 98, 591 A.2d 407 (1991) (concluding that United States Supreme Court deci- sion on which petitioner relied did not apply retroactively because it had resolved question of statutory interpretation and did not announce constitu- tional rule). I also note that I agree with the majority that nothing we said in Luurtsema v. Commissioner of Correction, 299 Conn. 740, 12 A.3d 817 (2011), suggests that we should no longer follow the Teague framework in determining whether a constitutional rule applies retroactively. 5 As the Nevada Supreme Court recently put it, ‘‘[t]he policy concerns behind Teague are partly germane to collateral review by this and other state courts and partly not. We share the concern that the finality of convic- tions not be unduly disturbed, but the need to prevent excessive interference by federal habeas courts has no application to habeas review by state courts themselves. And even the effect on finality is not as extreme when a state appellate court, as opposed to a federal court, decides to apply a rule retroactively: first, the decision affects only cases within that state, and second, most state collateral review occurs much sooner than federal collat- eral review. In addition, we are concerned with encouraging the [trial] courts of this state to strive for perspicacious, reasonable application of constitutional principles in cases where no precedent appears to be squarely on point.’’ Colwell v. State, supra, 118 Nev. 818. 6 I note, however, that ‘‘[t]he standard for determining when a case estab- lishes a new rule is objective, and the mere existence of conflicting authority does not necessarily mean a rule is new.’’ (Internal quotation marks omitted.) Chaidez v. United States, supra, 133 S. Ct. 1120 (Sotomayor, J., dissenting). This is especially true with cases involving a governing standard that, like Strickland, will necessarily evolve to reflect current practices. Thus, in the present case, the ‘‘earlier decisions show nothing more than that the underlying professional norms had not yet evolved to require attorneys to provide advice about deportation consequences.’’ Id., 1118 (Sotomayor, J., dissenting). 7 In arguing to the contrary, the respondent relies on State v. Aquino, 89 Conn. App. 395, 406–407, 873 A.2d 1075 (2005), in which the Appellate Court concluded that deportation was a collateral consequence outside the scope of the sixth amendment. Subsequently, however, following our grant of certification, we reversed the judgment of the Appellate Court and concluded that that court did not have jurisdiction to address the claim and should have dismissed the appeal as moot. State v. Aquino, 279 Conn. 293, 298–99, 901 A.2d 1194 (2006). Because the Appellate Court did not have jurisdiction over the appeal in Aquino, its discussion of the merits was dicta and not binding on habeas courts at the time Padilla was decided. Cf. State v. Singleton, 274 Conn. 426, 440, 876 A.2d 1 (2005) (‘‘when a court dismisses a case for lack of subject matter jurisdiction, any further discussion of the merits of that case is dicta’’). 8 The majority also notes that, in Chaidez, the court included Connecticut among the jurisdictions to have determined that advice concerning immigra- tion consequences does not fall within the scope of the right to the effective assistance of counsel. See Chaidez v. United States, supra, 133 S. Ct. 1109 n.8. The case on which the court in Chaidez relied for that proposition, however, Niver v. Commissioner of Correction, 101 Conn. App. 1, 919 A.2d 1073 (2007), held no such thing. In Niver, the Appellate Court affirmed the habeas court’s denial of a claim of ineffective assistance of counsel based on counsel’s failure to provide adequate advice regarding the immigration consequences of a plea, and, in setting forth the governing law, stated that ‘‘[t]he impact of a plea’s immigration consequences . . . is not of constitu- tional magnitude . . . .’’ (Internal quotation marks omitted.) Id., 4. The court did not resolve the case on those grounds, however, and went on to conclude that counsel’s performance was not deficient because he ‘‘specifi- cally discussed the potential immigration consequences of [the petitioner’s] plea’’; id., 5; and that, in any event, the petitioner failed to prove that she was prejudiced. Id., 5–6. Thus, Niver cannot fairly be read to stand for the proposition for which it was cited in Chaidez. 9 Although the majority cites four cases in which Padilla claims were rejected in the eight months following Chaidez as evidence that ‘‘the petition- er’s claim that retroactive application of Padilla in Connecticut will have a limited effect is belied by the facts,’’ the majority fails to acknowledge that, in three of those cases, the habeas court denied the claim on the merits. See Gonzalez v. Commissioner of Correction, 145 Conn. App. 28, 30, 74 A.3d 509 (noting habeas court determined that counsel adequately advised petitioner regarding immigration consequences), cert. denied, 310 Conn. 929, 78 A.3d 145 (2013); Saksena v. Commissioner of Correction, Superior Court, judicial district of Tolland, Docket No. CV-08-4002306-S (June 2, 2011) (same), aff’d, 145 Conn. App. 152, 158–59, 76 A.3d 192, cert. denied, 310 Conn. 940, 79 A.3d 892 (2013); Alcena v. Warden, Superior Court, judicial district of Tolland, Docket No. CV-10-4003448-S (May 5, 2011) (same; also determined that petitioner did not establish prejudice). The court in the fourth case did not reach the merits, rejecting the claim on the ground that Padilla did not apply retroactively. Gjini v. Warden, Superior Court, judicial district of Tolland, Docket No. CV-10-4003834-S (March 6, 2013). These cases do not support the majority’s concerns that applying Padilla retroactively would lead to a flood of new trials; in fact, they demonstrate that the Strickland standard adequately protects the state’s interest in the finality of convictions.