West v. People

Annotate this Case
Justia Opinion Summary

During 2002 and 2003, a Colorado state public defender with the Mesa County Regional Office represented defendant Thomas West after the victim's mother, D.S., informed police that she discovered West lying in bed next to her six-year-old daughter with his genitals exposed. D.S. and her ex-husband, D.E.S., both testified at trial for the prosecution. Colorado public defenders had represented D.S. approximately 23 times over the eight years preceding West's case. Although West's trial counsel had never himself represented D.S., the Mesa County Office represented her seven times between 1998 and 2001. D.S. was also a client of the El Paso County Regional Office, where she had an open case during West's trial at issue here. In addition, the Mesa County Office represented D.S.'s ex-husband, D.E.S., five times between 1999 and 2002. West's trial counsel had filed an entry of appearance in one of these cases, although the prosecution dismissed that case four days after that entry of appearance. West's trial counsel did not inform West or the trial court about these possible conflicts of interest. There was no record regarding the conflict at trial. The jury convicted West of sexual assault on a child. Following his trial, West filed a Crim. P. 35(c) motion, alleging that his trial counsel labored under a conflict of interest. The trial court found no conflict and denied the motion. The court of appeals reversed, holding that an actual conflict of interest arose from the public defender's dual role as prior and current counsel for D.S. and as prior counsel for D.E.S. The issue this case presented for the Colorado Supreme Court's review centered on the analytical framework a trial court should use to resolve a criminal defendant's post-conviction claim of ineffective assistance of counsel based on alleged conflicts of interest arising from concurrent or successive representation of witnesses against the defendant. The Court shared petitioners' contention that, under "Colorado v. Castro," (657 P.2d 932 (Colo. 1983)), they should not be required to demonstrate a separate "adverse effect" in addition to a conflict of interest in order to receive new trials. In order to prevail on an ineffective assistance of counsel claim predicated on trial counsel's alleged conflict of interest arising from concurrent or successive representation of trial witnesses against a defendant, the Court held that a defendant must show by a preponderance of the evidence both a conflict of interest and an adverse effect resulting from that conflict.

Download PDF
1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 Opinions of the Colorado Supreme Court are available to the public and can be accessed through the Court’s homepage at http://www.courts.state.co.us. Opinions are also posted on the Colorado Bar Association homepage at www.cobar.org. ADVANCE SHEET HEADNOTE January 20, 2015 2015 CO 5 Nos. 12SC301, West v. People & 13SC209, Cano v. People — Conflicts of Interest — Post-conviction and Extraordinary Relief — Ineffective Assistance of Counsel. In these appeals, the defendants alleged that their trial counsel labored under 6 conflicts of interest because counsel concurrently or successively represented trial 7 witnesses against them. The court of appeals remanded both cases to the trial courts to 8 determine whether, under Cuyler v. Sullivan, 446 U.S. 335 (1980), the defendants’ 9 attorneys labored under an “actual conflict.” The defendants separately petitioned for 0 review of the court of appeals’ judgments, asking the court to clarify whether the 1 Sullivan standard requires a defendant to demonstrate, in addition to a conflict of 2 interest, that an “adverse effect” arose from the conflict. 3 In People v. Castro, 657 P.2d 932 (Colo. 1983), the supreme court held that an 4 adverse effect was inherent in a “real and substantial” conflict of interest and thus a 5 separate showing was unnecessary. In this consolidated opinion, the supreme court 6 overrules Castro because the United States Supreme Court recently held that an “actual 7 conflict,” under the Sullivan standard, requires a defendant to show both a conflict of 8 interest and an adverse effect on his attorney’s performance. 1 The supreme court holds that to show an “adverse effect,” a defendant must (1) 2 identify a plausible alternative defense strategy or tactic that trial counsel could have 3 pursued, (2) show that the alternative strategy or tactic was objectively reasonable 4 under the facts known to counsel at the time of the strategic decision, and (3) establish 5 that counsel’s failure to pursue the strategy or tactic was linked to the actual conflict. 6 The supreme court therefore affirms the court of appeals’ judgments in part and 7 instructs the trial courts to consider whether, under this framework, the defendants 8 received ineffective assistance of counsel by virtue of their attorneys’ alleged conflicts 9 and are therefore entitled to new trials. 2 1 2 The Supreme Court of the State of Colorado 2 East 14th Avenue • Denver, Colorado 80203 3 2015 CO 5 4 5 6 Supreme Court Case No. 12SC301 Certiorari to the Court of Appeals Court of Appeals Case No. 10CA883 7 Petitioner: 8 Thomas Ray West, 9 v. 0 Respondent: 1 The People of the State of Colorado. 2 * * * * * 3 4 5 Supreme Court Case No. 13SC209 Certiorari to the Court of Appeals Court of Appeals Case No. 10CA2659 6 Petitioner/Cross-Respondent: 7 Raymond Albert Cano, 8 v. 9 Respondent/Cross-Petitioner: 0 The People of the State of Colorado. 1 2 3 Judgments Affirmed en banc January 20, 2015 4 5 6 7 8 9 0 1 2 Attorney for Petitioner Thomas West: William J. Fritsche, P.C. William J. Fristche Fort Collins, Colorado Attorneys for Respondent: Cynthia H. Coffman, Attorney General Ryan A. Crane, Assistant Attorney General Denver, Colorado 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 Attorney for Amicus Curiae Colorado Criminal Defense Bar: Reppucci Law Firm, P.C. Jonathan D. Reppucci Denver, Colorado Attorney for Petitioner/Cross-Respondent Raymond Cano: Law Office of Gregory Lansky, LLC Gregory Lansky Aurora, Colorado Attorneys for Respondent/Cross-Petitioner: Cynthia H. Coffman, Attorney General Christine C. Brady, Senior Assistant Attorney General Denver, Colorado JUSTICE HOOD delivered the Opinion of the Court. JUSTICE COATS dissents, and JUSTICE EID and JUSTICE MÁRQUEZ join in the dissent. 2 ¶1 What analytical framework should a trial court use to resolve a criminal defendant’s post-conviction claim of ineffective assistance of counsel based on alleged conflicts of interest arising from concurrent or successive representation of witnesses against the defendant? We granted certiorari to address this issue and petitioners’ shared contention that, under People v. Castro, 657 P.2d 932 (Colo. 1983), they should not be required to demonstrate a separate “adverse effect” in addition to a conflict of interest in order to receive new trials. ¶2 While Castro said that an attorney who labors under a real and substantial conflict of interest “cannot avoid being adversely affected,” id. at 944–45, that language cannot be reconciled with the holding in Mickens v. Taylor, 535 U.S. 162 (2002). In Mickens, the Supreme Court held that in order to demonstrate a Sixth Amendment violation, a defendant must establish that the conflict of interest adversely affected his counsel’s performance. Based on the Supreme Court’s holding in Mickens, we now overrule Castro. ¶3 In order to prevail on an ineffective assistance of counsel claim predicated on trial counsel’s alleged conflict of interest arising from concurrent or successive representation of trial witnesses against a defendant, we hold that a defendant must show by a preponderance of the evidence both a conflict of interest and an adverse effect resulting from that conflict. To show an adverse effect, a defendant must (1) identify a plausible alternative defense strategy or tactic that trial counsel could have pursued, (2) show that the alternative strategy or tactic was objectively reasonable under the facts known to counsel at the time of the strategic decision, and (3) establish 3 that counsel’s failure to pursue the strategy or tactic was linked to the actual conflict. A defendant may prove the link under the third prong by showing that the alternative strategy or tactic was inherently in conflict with counsel’s other loyalties or interests or by showing that the alternative strategy or tactic was not undertaken due to those other loyalties or interests.1 I. Facts and Procedural History A. West During 2002 and 2003, a Colorado state public defender with the Mesa County ¶4 Regional Office (hereinafter “the Mesa public defender” or “West’s trial counsel”) represented Thomas West after the victim’s mother, D.S., informed police that she discovered West lying in bed next to her six-year-old daughter with his genitals exposed. D.S. and her ex-husband, D.E.S., both testified at trial for the prosecution. Colorado public defenders had represented D.S. approximately 23 times over the ¶5 eight years preceding West’s case. Although West’s trial counsel had never himself 1 We granted certiorari to review the following three issues: 1. Whether a defendant who has demonstrated ineffective assistance of counsel in a Crim. P. 35(c) proceeding because of a conflict of interest must separately demonstrate an “adverse effect” arising from that conflict. [West and Cano] 2. Whether a defendant must identify something that counsel chose to do or not to do, as to which he had conflicting duties, and show that the course taken was influenced by the conflict in order to demonstrate an “adverse effect” arising from that conflict. [Cano] 3. Whether the court of appeals erred in concluding that the Rules of Professional Conduct in effect at the time of petitioner’s trial applied on collateral review when the pertinent rules were modified prior to defendant’s evidentiary hearing on his post-conviction motion. [Cano] 4 represented D.S., the Mesa County Office represented her seven times between 1998 and 2001. D.S. was also a client of the El Paso County Regional Office, where she had an open case during West’s trial at issue here. In addition, the Mesa County Office represented D.S.’s ex-husband, D.E.S., five ¶6 times between 1999 and 2002. West’s trial counsel had filed an entry of appearance in one of these cases, although the prosecution dismissed that case four days after that entry of appearance. West’s trial counsel did not inform West or the trial court about these possible ¶7 conflicts of interest. There was no record regarding the conflict at trial. The jury convicted West of sexual assault on a child. Following his trial, West filed a Crim. P. 35(c) motion, alleging that his trial ¶8 counsel labored under a conflict of interest. The trial court found no conflict and denied the motion. The court of appeals reversed, holding that an actual conflict of interest arose from the public defender’s dual role as prior and current counsel for D.S. and as prior counsel for D.E.S.2 The court of appeals remanded the case to the trial court to determine whether, under Cuyler v. Sullivan, 446 U.S. 335 (1980), the alleged conflicts adversely affected the Mesa public defender’s performance in West’s trial. B. Cano Colorado state public defenders with the Adams County Regional Office ¶9 represented Raymond Cano during his murder trial for a gang-related stabbing. At the People v. West, No. 10CA883 (Colo. App. Nov. 23, 2011) (not published under C.A.R. 35(f)). 2 5 same time, an attorney from that office entered an appearance for Sergio Aguilar, a prosecution witness against Cano, after Aguilar was arrested on an unrelated attempted murder charge. The Adams County Office thus represented both Cano and Aguilar in different matters during Cano’s trial. Although Aguilar did not testify at the trial, evidence indicated that he could have been an alternate suspect in the stabbing alleged in Cano’s case. Cano’s attorneys did not pursue this theory. There was no record at trial of the conflict. A jury found Cano guilty of first degree murder. ¶10 Cano, like West, filed a Crim. P. 35(c) motion for post-conviction relief, raising the conflict of interest issue. The trial court denied the motion, finding that no conflict existed under Strickland v. Washington, 466 U.S. 668 (1984), and that even if a conflict existed, no prejudice to Cano’s defense resulted from it. The court of appeals held that the trial court erred by applying Strickland and reversed and remanded for consideration under the Sullivan standard.3 II. Standard of Review ¶11 In a Crim. P. 35(c) proceeding, the trial court is the trier of fact and determines the weight and credibility of witness testimony. Dunlap v. People, 173 P.3d 1054, 1061– 62 (Colo. 2007); Kailey v. Colo. State Dep’t of Corr., 807 P.2d 563, 567 (Colo. 1991). We defer to a post-conviction court’s findings of fact when they are supported by the evidence but review conclusions of law de novo. People v. Romero, 953 P.2d 550, 555 (Colo. 1998); People v. Quezada, 731 P.2d 730, 732–33 (Colo. 1987). People v. Cano, No. 10CA2659 (Colo. App. Jan. 31, 2013) (not published under C.A.R. 35(f)). 3 6 III. Analysis ¶12 After exploring the evolution of Sixth Amendment doctrine and the resulting right to effective, conflict-free counsel, we turn to the tension between our decision in Castro and the Supreme Court’s holding in Mickens. We update our precedent to comport with Mickens and then turn to what measure of prejudice must be shown for a defendant to obtain post-conviction relief in this setting. ¶13 We hold that a defendant must show an “adverse effect” resulting from a conflict of interest. We then confront what “adverse effect” means and what analytical framework trial courts should use to evaluate when it exists. We explore various tests that have cropped up in the wake of Mickens, ultimately settling on a three-part test that we believe strikes an appropriate balance between a criminal defendant’s right to have effective representation, in the form of conflict-free counsel, and the public’s interest in the finality of verdicts. A. The Right to the Assistance of Conflict-Free Counsel ¶14 “In all criminal prosecutions, the accused shall enjoy the right . . . to have the Assistance of Counsel for his defen[s]e.” U.S. Const. amend. VI; see also Colo. Const. art. II, § 16 (“In criminal prosecutions the accused shall have the right to appear and defend in person and by counsel.”). From this bedrock right, the Supreme Court has generated equally familiar corollaries, such as the right in some cases to appointed counsel and the right in all cases to the effective assistance of counsel. Strickland, 466 U.S. at 685–86. 7 The overarching objective, of course, is to secure a criminal defendant’s right to a ¶15 fair trial. “The right to counsel plays a crucial role in the adversarial system embodied in the Sixth Amendment, since access to counsel’s skill and knowledge is necessary to accord defendants the ample opportunity to meet the case of the prosecution.” Id. at 685 (internal quotation marks omitted). Indeed, “[o]f all the rights that an accused person has, the right to be represented by counsel is by far the most pervasive for it affects his ability to assert any other rights he may have.” United States v. Cronic, 466 U.S. 648, 654 (1984). Because of the crucial role of impartial and zealous counsel in securing due process, the right to effective assistance of counsel includes the right to conflict-free counsel. See, e.g., Mickens, 535 U.S. at 166; Wood v. Georgia, 450 U.S. 261, 271 (1981); People v. Martinez, 869 P.2d 519, 524 (Colo. 1994). “That a person who happens to be a lawyer is present at trial alongside the accused . . . is not enough to satisfy the constitutional command.” Strickland, 466 U.S. at 685. ¶16 A conflict often exists when one attorney simultaneously represents two or more codefendants, Holloway v. Arkansas, 435 U.S. 475, 490 (1978), and may arise when one attorney simultaneously represents a defendant and a witness in that defendant’s trial. See Allen v. Dist. Court, 184 Colo. 202, 205, 519 P.2d 351, 353 (1974) (“It is of the utmost importance that an attorney’s loyalty to his client not be diminished, fettered, or threatened in any manner by his loyalty to another client.”).4 See also Colo. RPC 1.7 cmt. 6 (“Loyalty to a current client prohibits undertaking representation directly adverse to that client without that client’s informed consent. Thus, absent consent, a lawyer may not act as an advocate in one matter against a person the lawyer represents in some other matter, even when the matters are wholly 4 8 ¶17 A similar conflict may arise when an attorney has previously represented a trial witness. This “successive representation” may restrict the attorney’s present representation of the defendant “because of the [attorney’s] duty to maintain the confidentiality of information” that he received in his prior representation of the trial witness. Rodriguez v. Dist. Court, 719 P.2d 699, 704 (Colo. 1986). Because this duty of confidentiality survives the termination of an attorney-client relationship, it “creates the possibility that the attorney will be hindered in cross-examining the witness, which thus impedes the attorney’s ability to zealously represent the current client.” Dunlap, 173 P.3d at 1070 (citing Rodriguez, 719 P.2d at 704).5 B. The Road to Mickens ¶18 A defendant seeking post-conviction relief based on ineffective assistance of counsel resulting from an attorney’s alleged conflict “must demonstrate that an actual conflict of interest adversely affected his lawyer’s performance.” Sullivan, 446 U.S. at 348. To understand what these words mean, we examine the context in which they emerged and how the Supreme Court has construed them. unrelated. . . . Similarly, a directly adverse conflict may arise when a lawyer is required to cross-examine a client who appears as a witness in a lawsuit involving another client, as when the testimony will be damaging to the client who is represented in the lawsuit.”). We note that Rule 1.7’s comments have changed slightly, but not meaningfully, since the period applicable to the cases before us. See also Colo. RPC 1.9 cmt. 1 (“After termination of a client-lawyer relationship, a lawyer has certain continuing duties with respect to confidentiality and conflicts of interest and thus may not represent another client except in conformity with this Rule.”). We note that Rule 1.9’s comments have also changed slightly, but not meaningfully, since the period applicable to the cases before us. 5 9 ¶19 In Holloway, the trial court appointed a public defender to represent three codefendants, despite objections that the codefendants’ conflicting interests would necessarily compromise the public defender’s ability to effectively represent each of them. 435 U.S. at 477. The trial court repeatedly denied motions for appointment of separate counsel. The Supreme Court noted that the danger with this concurrent representation lies “in what the advocate finds himself compelled to refrain from doing, not only at trial but also as to possible pretrial plea negotiations and in the sentencing process.” Id. at 490 (emphasis in original). For instance, conflicting representations can prevent an attorney from pursuing strategies favorable to one client that might inculpate another, like seeking a plea deal in exchange for testimony against a codefendant. Id. Holloway established an automatic reversal rule when counsel is forced to represent codefendants over a timely objection, unless the trial court first determines that no conflict exists. See Mickens, 535 U.S. at 168 (citing Holloway, 435 U.S. at 488). ¶20 In Sullivan, two private attorneys represented the defendant and his alleged co-conspirators, who were tried separately. 446 U.S. at 337–38. After the jury convicted him of murder, Sullivan argued that his attorneys’ conflicting interests denied him effective assistance of counsel. Id. at 338. Significantly, no one had objected to the concurrent representation at trial. Id. at 337–38. ¶21 The Court explained that although concurrent representation of codefendants almost always involves conflicts, a reviewing court may presume that counsel was 10 ineffective only when the trial court denied a defendant the opportunity to show that the conflict imperiled his right to a fair trial. Id. at 348. The Court reasoned: Holloway requires state trial courts to investigate timely objections to multiple representation. But nothing in our precedents suggests that the Sixth Amendment requires state courts themselves to initiate inquiries into the propriety of multiple representation in every case. Defense counsel have an ethical obligation to avoid conflicting representations and to advise the court promptly when a conflict of interest arises during the course of trial. Absent special circumstances, therefore, trial courts may assume either that multiple representation entails no conflict or that the lawyer and his clients knowingly accept such risk of conflict as may exist. Indeed, as the Court noted in Holloway, trial courts necessarily rely in large measure upon the good faith and good judgment of defense counsel. Id. at 346–47 (citations omitted). Thus, without a defendant’s objection, the mere fact that an attorney concurrently represented codefendants was insufficient to prompt a court to initiate an inquiry into possible conflicts of interest. Id. ¶22 The Sullivan Court held that to obtain reversal absent objection, a defendant must show that a conflict actually affected the adequacy of the representation. Id. at 348–49. “This requires showing both that counsel was placed in a situation where conflicting loyalties pointed in opposite directions (an ‘actual conflict’) and that counsel proceeded to act against the defendant’s interests (‘adversely affect[ing] his performance’).” 3 Wayne R. LaFave et al., Criminal Procedure § 11.9(d) (3d ed. 2013). ¶23 Holloway and Sullivan apply to conflicts arising from multiple concurrent representation. The Supreme Court later announced a different standard for other ineffectiveness claims. In Strickland, the defendant sought post-conviction relief, alleging that counsel was ineffective during his sentencing proceeding. 466 U.S. at 675. The Court held that the purpose of the right to counsel is to ensure a fair trial and to 11 justify reliance on the verdict, and therefore “any deficiencies in counsel’s performance must be prejudicial to the defense in order to constitute ineffective assistance under the Constitution.” Id. at 692. “The defendant must show that there is a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Id. at 694. ¶24 Sullivan, then, occupies a middle ground between Holloway’s per se reversal rule and Strickland’s requirement that a defendant demonstrate prejudice to the outcome of the trial. See Dunlap, 173 P.3d at 1073 n.24 (“The [Sullivan] standard operates as an exception to the normal requirements of Strickland.”). The Sullivan exception applies “needed prophylaxis in situations where Strickland itself is evidently inadequate to assure vindication of the defendant’s Sixth Amendment right to counsel.” Mickens, 535 U.S. at 176. ¶25 In the years immediately following Sullivan, it was unclear whether defendants needed to show a separate adverse effect in addition to a conflict of interest when addressing ineffective assistance claims based on multiple concurrent representation. The Supreme Court created some confusion in Wood when it vacated the defendants’ convictions and remanded with the instruction that “[i]f the court finds that an actual conflict of interest existed at that time . . . it must hold a new revocation hearing that is untainted by a legal representative serving conflicting interests.” 450 U.S. at 273–74. Conspicuously absent from this instruction was any mention of an adverse effect resulting from the conflict, leading some courts to conclude that a separate showing was not required. See Mickens, 535 U.S. at 171–72 (rejecting the argument that Wood’s 12 remand instruction eliminated the need to show adverse effect). ¶26 Before Mickens, this court concluded that a defendant need not separately show that counsel’s performance was adversely affected by a conflict. Castro, 657 P.2d at 944–45. In Castro, the defendant claimed ineffective assistance of counsel because his lawyer simultaneously represented the district attorney for the same judicial district in unrelated proceedings. Id. at 936. This dual representation created “a real and substantial conflict that placed the defense attorney in a situation inherently conducive to and productive of divided loyalties.” Id. at 945. We concluded that “[a] lawyer representing such conflicting interests cannot avoid being adversely affected at various stages of the criminal prosecution of an accused,” and we reversed and remanded without considering what adverse effect, if any, the concurrent representation may have had. Id. at 944–46.6 ¶27 Two years later, in Armstrong v. People, 701 P.2d 17, 26 (Colo. 1985), Justice Quinn, who penned the majority opinion in Castro, specially concurred to argue that “the Supreme Court did not intend to require proof of adverse effect as a necessary component of a claim of ineffective assistance of counsel based on an actual conflict of interest.” But the majority in Armstrong did not resolve the tension between Sullivan’s requirement that a defendant show an adverse effect and Wood’s suggestion that an “actual conflict of interest” alone suffices. Id. at 20–21. In doing so, we did not purport to create a different rule under our state constitution than exists under the federal constitution. 6 13 ¶28 Mickens resolved this tension. In Mickens, the Supreme Court clarified that the phrase “actual conflict of interest” in Wood’s remand instruction was shorthand for Sullivan’s holding: As used in the remand instruction, however, we think “an actual conflict of interest” meant precisely a conflict that affected counsel’s performance—as opposed to a mere theoretical division of loyalties. It was shorthand for the statement in Sullivan that “a defendant who shows that a conflict of interest actually affected the adequacy of his representation need not demonstrate prejudice in order to obtain relief.” 535 U.S. at 171 (emphasis in original). “[T]he Sullivan standard is not properly read as requiring inquiry into actual conflict as something separate and apart from adverse effect. An ‘actual conflict,’ for Sixth Amendment purposes, is a conflict of interest that adversely affects counsel’s performance.” Id. at 211 n.5. Thus, a defendant who alleges that a conflict of interest deprived him of effective assistance of counsel must show (1) that counsel had a conflict of interest (2) that adversely affected the representation. ¶29 Because Castro eliminated the need for defendants to demonstrate an adverse effect, Mickens implicitly overruled it; therefore, we do so explicitly now. C. The Post-Mickens Landscape ¶30 Mickens left unresolved two issues relevant to the cases before us: first, whether Sullivan applies in cases involving conflicts from simultaneous representation of not only codefendants, but also defendants and trial witnesses; and, second, whether Sullivan applies in cases of successive representation. We explore each in turn. 1. Multiple Concurrent Representation of Witnesses ¶31 The Supreme Court has never expressly limited Sullivan to cases involving only 14 joint representation of codefendants, not even in Mickens when it plainly could have. ¶32 In Mickens, a Virginia jury convicted the defendant of murder and sentenced him to death. 535 U.S. at 164. Court-appointed trial counsel for Mickens had briefly been counsel for the victim in an unrelated juvenile proceeding at the time of the murder. During federal habeas proceedings, Mickens alleged that he had been denied his Sixth Amendment right to the effective assistance of trial counsel as a result of the conflict arising from this multiple representation. The federal district court denied Mickens’s petition. The Court of Appeals for the Fourth Circuit, sitting en banc, affirmed the denial of habeas relief. Relying on Sullivan, the Fourth Circuit held that a defendant must show “both an actual conflict of interest and an adverse effect even if the trial court failed to inquire into a potential conflict about which it reasonably should have known.” Id. at 165 (citing Mickens v. Taylor, 240 F.3d 348, 355–56 (4th Cir. 2001) (en banc)). Because Mickens failed to demonstrate adverse effect, he lost. Id. ¶33 The Supreme Court granted certiorari to address “what a defendant must show in order to demonstrate a Sixth Amendment violation where the trial court fails to inquire into a potential conflict of interest about which it knew or reasonably should have known.” Id. at 164. The “nub” of this question, as Justice Scalia explained in writing for the majority, was whether Holloway and its progeny provided an automatic reversal exception to the general rule of Strickland “under the circumstances in the present case.” See id. at 167. Mickens argued that Wood’s remand instruction established an “unambiguous rule” that where the trial judge neglects a duty to inquire into a potential conflict, the defendant, to obtain reversal of judgment, need only show a 15 conflict of interest and not any adverse effect. Id. at 172. The Mickens Court flatly rejected this assertion. Id. In doing so, the Court observed that “the rule applied when the trial judge is not aware of the conflict (and thus not obligated to inquire) is that prejudice will be presumed only if the conflict has significantly affected counsel’s performance—thereby rendering the verdict unreliable, even though Strickland prejudice cannot be shown.” Id. at 172–73. The Court then held: Since this was not a case in which (as in Holloway) counsel protested his inability simultaneously to represent multiple defendants; and since the trial court’s failure to make the Sullivan-mandated inquiry does not reduce the petitioner’s burden of proof; it was at least necessary, to void the conviction, for petitioner to establish that the conflict of interest adversely affected his counsel’s performance. The Court of Appeals having found no such effect, the denial of habeas must be affirmed. Id. at 173–74 (emphasis added). ¶34 Thus, in a case involving an alleged conflict of interest stemming from trial counsel’s former representation of a victim—not concurrent or successive representation of a codefendant—the Court embraced the concept of individualized inquiry into adverse effect, without requiring a showing of Strickland prejudice. Had the Court wanted to dispose of the case under Strickland, and thereby eliminate any middle ground for multiple representation cases not involving codefendants, it could have done so. It did not.7 Wood also suggests that Sullivan extends to representation of multiple clients with potentially conflicting interests, even if they are not codefendants. 450 U.S. 261. In Wood, the attorney’s potential conflict arose where an employer hired and paid the attorney to represent three employees. Id. at 266. The employer and employees were not codefendants. In remanding the case, the Court held that an actual conflict’s existence would depend upon “whether counsel was influenced in his basic strategic 7 16 ¶35 Courts have thus continued to apply Sullivan to conflicts arising from multiple concurrent representation of defendants and witnesses. See, e.g., United States v. Blount, 291 F.3d 201, 210 (2d Cir. 2002) (applying Sullivan “in light of the Supreme Court’s recent decision in Mickens” where another member of defense counsel’s law firm represented a government witness in an unrelated proceeding); see also Wilkins v. Stephens, 560 F. App’x 299, 309 (5th Cir. 2014) (applying Sullivan where defense counsel formerly represented victim). ¶36 Sullivan stressed “the high probability of prejudice arising from multiple concurrent representation, and the difficulty of proving that prejudice.” Mickens, 535 U.S. at 174 (citing Sullivan, 446 U.S. at 348–49). In discussing this in Mickens, the Court reiterated, “[U]ntil a defendant shows that his counsel actively represented conflicting interests, he has not established the constitutional predicate for his claim of ineffective assistance.” Id. (second emphasis added) (citing Sullivan, 446 U.S. at 350). Because Mickens involved a claim of ineffective assistance of counsel based on a conflict resulting from representation of someone other than a codefendant, we conclude that the Supreme Court remanded for evaluation of “adverse effect” with the thought that other forms of multiple representation could constitute the “actively conflicting decisions by the interests of the employer who hired him”—whether the attorney was also representing the employer was of no consequence. Id. at 272. The Wood Court emphasized “the inherent dangers that arise when a criminal defendant is represented by a lawyer hired and paid by a third party,” id. at 268–69, and that additional dangers may arise where pursuing the payor’s interests requires sacrificing the defendant’s interests, id. at 269–70. Thus, the Court’s analysis was premised on the divided loyalty resulting from multiple representation, seemingly irrespective of whether the clients were codefendants. 17 interests” necessary to demonstrate a Sixth Amendment violation under Sullivan. The question left unanswered in Mickens is what test should be used to assess adverse effect. We answer that question after first examining successive representation.8 2. Successive Representation ¶37 Whether Sullivan should be extended to cases of successive representation “remains, as far as the jurisprudence of the [United States Supreme] Court is concerned, an open question.” Id. at 176. Federal circuit courts have reached divergent conclusions, but the majority applies Sullivan in cases of successive representation.9 ¶38 Our post-Mickens precedent has also left open the possibility that Sullivan extends to successive representation cases. In Dunlap, this court applied the Sullivan standard to a conflict arising from successive representation because the parties In dicta, the Mickens Court questioned the assumption that Strickland should not govern claims of ineffectiveness based on alleged conflicts resulting from other forms of divided loyalty (for example, counsel’s personal or financial interests, including employment concerns, romantic entanglements, and fear of antagonizing the trial judge). Id. at 174–75. What should be made of that dicta is a matter for another day. Our analysis today concerns only Sullivan’s application to alleged conflicts arising from multiple representation. 8 Compare Alberni v. McDaniel, 458 F.3d 860, 874 (9th Cir. 2006) (state supreme court did not err by applying Sullivan to successive representation); United States v. Infante, 404 F.3d 376, 391, 395 n.12 (5th Cir. 2005) (applying Sullivan to successive representation closely related in subject matter and timing); Hall v. United States, 371 F.3d 969, 974 (7th Cir. 2004) (same); Moss v. United States, 323 F.3d 445, 462 (6th Cir. 2003) (applying Sullivan to successive representation that was “nearly identical” to the first representation), with Jalowiec v. Bradshaw, 657 F.3d 293, 316–17 (6th Cir. 2011) (questioning whether Moss’s application of Sullivan is appropriate for successive representation but concluding that defendant failed to demonstrate adverse effect and was not entitled to relief); Gray v. Farris, 560 F. App’x. 740, 743 n.1 (10th Cir. 2014) (noting that the Supreme Court has never applied Sullivan outside the context of multiple concurrent representation); Schwab v. Crosby, 451 F.3d 1308, 1325–28 (11th Cir. 2006) (same). 9 18 assumed it applied. 173 P.3d at 1073 n.24. We do the same today. Because the parties have not briefed the issue, we assume, without deciding, that the Sullivan standard applies to alleged conflicts arising from successive representation. With these principles in mind, we now turn to the facts of the cases before us. D. Application to West and Cano 1. Conflicts ¶39 The cases before us involve public defenders’ simultaneous representation of the defendants and the prosecution witnesses against them. In neither case was there any acknowledgement or record of the conflict at trial. The post-conviction courts found that these situations did not raise conflicts of interest for counsel, but two divisions of the court of appeals reversed and remanded for evaluation under Sullivan. ¶40 Communications between the Mesa and El Paso offices of the State Public Defender potentially tainted the Mesa public defender’s representation of West. The record reveals that Mesa’s investigator requested D.S.’s files in an unrelated open matter from the El Paso County Office during West’s case. Even assuming that the Mesa public defender did not actually acquire confidential information about D.S., as the trial court found, he undeniably had access to it. It is evident that representation of both West and D.S. created “a situation inherently conducive to and productive of divided loyalties.” Castro, 657 P.2d at 945. ¶41 Although the public defender representing the potential alternate suspect, Aguilar, was not involved in Cano’s case, the Adams County Office consisted of approximately 15 attorneys who routinely consulted each other on cases and worked in 19 close proximity to one another. We may presume that Cano’s public defenders in the same office had access to confidential material about Aguilar, and vice versa. ¶42 The Mesa County Office’s potential imputed10 conflict involving D.S. and the Adams County Office’s potential conflict with Aguilar both involve concurrent representation and therefore fall under Sullivan. ¶43 D.E.S., however, presents a slightly different issue: West’s trial counsel once entered an appearance for D.E.S., and although he had no open cases with public defenders during West’s proceedings, the Mesa County Office had represented him many times. The potential conflict with D.E.S. was therefore successive. ¶44 Whether “actual” conflicts existed in West’s and Cano’s cases will turn on whether the alleged conflicts adversely affected their counsels’ performances. West and Cano can thus prevail on their Crim. P. 35(c) motions on remand if they demonstrate Confidential information obtained by one public defender may be imputed to other public defenders. Rodriguez, 719 P.2d at 704. Lawyers generally are “vicariously bound by the obligation of loyalty owed by each lawyer with whom the lawyer is associated.” Colo. RPC 1.10 cmt. 2 (setting forth general principles of imputed disqualification). In affirming the court of appeals, however, we express no opinion regarding which ethics rule applies to these public defenders—Colo. RPC 1.10 (the general imputation rule) or Colo. RPC 1.11 (the special conflicts of interest rule for government employees). Nor do we hold that conflicts of interest are always imputed to other public defenders, either between or within regional offices. Our role here is “not to enforce the Canons of Legal Ethics, but to . . . assure vindication of the defendant’s Sixth Amendment right to counsel.” Mickens, 535 U.S. at 176; see also Nix v. Whiteside, 475 U.S. 157, 165 (1986) (“[B]reach of an ethical standard does not necessarily make out a denial of the Sixth Amendment guarantee of assistance of counsel.”). An attorney’s ethical obligations do not dictate the scope of the Sixth Amendment right to conflict-free counsel. Were it otherwise, this court could redraw the boundaries of the Sixth Amendment each time the Rules of Professional Conduct are revised. 10 20 that the public defenders’ concurrent or successive representation of prosecution witnesses had an adverse effect on their counsels’ representation of them. 2. Adverse Effect ¶45 While the United States Supreme Court in Mickens has defined “actual conflict” under the Sixth Amendment, it has not defined “adverse effect.” This lack of guidance has led courts, both in our court of appeals and the federal circuit courts, to apply inconsistent standards for determining an adverse effect. Therefore, we now explore the different adverse effect tests and describe how a defendant may demonstrate “that an actual conflict of interest adversely affected his lawyer’s performance.” Sullivan, 446 U.S. at 348. ¶46 In West, the court of appeals remanded with instructions to the trial court to apply the Sullivan standard because the trial court had made no findings “as to whether [West’s trial counsel]’s conflict adversely impacted [his] performance.” Slip op. at 12– 13. The court did not, however, instruct the trial court on what standard to use in determining whether an “adverse effect” exists. ¶47 In Cano, the court of appeals remanded with instructions to the trial court to apply the standard set forth in People v. Kenny, 30 P.3d 734, 745 (Colo. App. 2000), which instructs the defendant to “identify something that counsel chose to do or not do, as to which he had conflicting duties, and . . . show that the course taken was influenced by that conflict.” Slip op. at 14 (emphasis added) (citing Vance v. Lehman, 64 F.3d 119, 125 (3d Cir. 1995)). The Kenny standard recognizes that a separate adverse effect test is necessary, but it is insufficient given Sullivan’s prophylactic purpose. 21 ¶48 The majority of federal circuit courts requires defendants to satisfy a two-part standard by showing “that some plausible alternative defense strategy or tactic might have been pursued, and that the alternative defense was inherently in conflict with or not undertaken due to the attorney’s other loyalties or interests.” United States v. Levy, 25 F.3d 146, 157 (2d Cir. 1994) (emphasis added) (citations and internal quotation marks omitted); see also United States v. DeCologero, 530 F.3d 36, 77 (1st Cir. 2008) (same); Perillo v. Johnson, 205 F.3d 775, 807 (5th Cir. 2000) (same); United States v. Wells, 394 F.3d 725, 733 (9th Cir. 2005) (same); United States v. Bowie, 892 F.2d 1494, 1500 (10th Cir. 1990) (same); see also McFarland v. Yukins, 356 F.3d 688, 705–06 (6th Cir. 2004) (holding that an adverse effect “requires a choice by counsel, caused by the conflict of interest” (emphasis added)). ¶49 Other circuit courts require the defendant to satisfy a more rigorous three-part standard. In addition to the two parts used in the aforementioned circuits, these courts instruct that the defendant must also show that the alternative strategy was “objectively reasonable” under the facts known to the attorney at the time of the strategic decision. See United States v. Nicholson, 611 F.3d 191, 197 (4th Cir. 2010); see also Morelos v. United States, 709 F.3d 1246, 1252 (8th Cir. 2013) (“To prove a conflict produced an adverse effect, a defendant must identify a plausible alternative defense strategy or tactic that defense counsel might have pursued, show that the alternative strategy was objectively reasonable under the facts of the case, and establish that the defense counsel’s failure to pursue that strategy or tactic was linked to the actual conflict.” 22 (citation and internal quotation marks omitted)); Brownlee v. Haley, 306 F.3d 1043, 1080 n.17 (11th Cir. 2002) (same). ¶50 Lastly, the Seventh Circuit requires a defendant to prove an adverse effect “by showing that there is a reasonable likelihood that his counsel’s performance would have been different had there been no conflict of interest.” Hall, 371 F.3d at 974 (emphasis added) (citing Stoia v. United States, 22 F.3d 766, 771 (7th Cir. 1994)). This language is similar to Strickland’s standard for showing prejudice, which requires a defendant to show “a reasonable probability that, but for counsel’s unprofessional errors, the result of the proceeding would have been different.” Strickland, 466 U.S. at 694 (emphasis added). ¶51 We conclude that the two-part standard is too deferential to counsel’s subjective assessment of his representation. See, e.g., Nicholson, 611 F.3d at 207 (citing Beets v. Scott, 65 F.3d 1258, 1269 (5th Cir. 1995) (en banc) (recognizing that Strickland imposes a “more deferential standard of attorney competence” than Sullivan)). Research indicates that attorneys “systematically understate both the existence of conflicts and their deleterious effects.” Tigran W. Eldred, The Psychology of Conflicts of Interest in Criminal Cases, 58 U. Kan. L. Rev. 43, 48 (2009). An adverse effect therefore becomes too difficult for a defendant to prove because the inquiry relies largely on the attorney’s interpretations of his decisions amid the conflict. See Holloway, 435 U.S. at 491 (“[T]o assess the impact of a conflict of interests on the attorney’s options, tactics, and decisions in plea negotiations would be virtually impossible.”). 23 ¶52 Likewise, we find the Seventh Circuit’s reasonable likelihood standard inconsistent with Supreme Court precedent. While the Court has not defined “adverse effect,” it has consistently held that the Sullivan standard requires an actual conflict. We cannot fathom how the Court would then require only a reasonable likelihood that counsel’s performance would have been different. See Mickens, 535 U.S. at 163 (“‘[A]n actual conflict of interest’ mean[s] precisely a conflict that affected counsel’s performance—as opposed to a mere theoretical division of loyalties.”); see also Sullivan, 446 U.S. at 350 (“We hold that the possibility of conflict is insufficient to impugn a criminal conviction.”). ¶53 In selecting a standard, we are mindful that Sullivan’s prophylactic approach exists to eliminate the burden Strickland imposes on a defendant to demonstrate outcome-determinative prejudice. Dunlap, 173 P.3d at 1073. We aim to adopt a standard that balances the defendant’s interest in conflict-free counsel with the public’s interest in the finality of verdicts. ¶54 After carefully evaluating the alternatives, we strike that balance by adopting the Fourth Circuit’s three-part standard in Nicholson. The standard’s second prong— focusing on whether the alternative strategy or tactic was objectively reasonable under the facts known to the attorney at the time of the strategic decision—promotes conflict-free counsel by eliminating a defendant’s forced reliance on the attorney’s subjective assessment of his representation. By allowing a defendant to rely in part on a reasonableness inquiry, we ensure that the standard is not too deferential to attorneys. See Nicholson, 611 F.3d at 207. 24 ¶55 The “objectively reasonable” prong simultaneously ensures the finality of verdicts because it shields attorneys against more tenuous claims of ineffective assistance of counsel where the alternative strategy or tactic would have proved unwise, illogical, or otherwise undesirable under the factual circumstances. See, e.g., Dunlap, 173 P.3d at 1076 (holding that counsel’s decision to abandon the alternate suspect theory was “strategic” given the case’s unexpected direction). ¶56 This three-part standard is also attractive because the Fourth Circuit relied on it in Mickens v. Taylor, 240 F.3d 348 (4th Cir. 2001) (en banc), aff’d without consideration of this point, 535 U.S. 162. While the standard’s propriety was not before the Supreme Court in Mickens, the Court could have disavowed the standard’s application when it discussed the impropriety of applying Sullivan to all forms of attorney conflicts. See Mickens, 535 U.S. at 174–75. Because that did not happen, the Nicholson test we embrace today is alive and well in the Fourth Circuit. ¶57 Accordingly, we hold that to show an adverse effect, a defendant must (1) identify a plausible alternative defense strategy or tactic that counsel could have pursued, (2) show that the alternative strategy or tactic was objectively reasonable under the facts known to counsel at the time of the strategic decision, and (3) establish that counsel’s failure to pursue the strategy or tactic was linked to the actual conflict. ¶58 The first prong—the existence of a plausible alternative defense strategy or tactic that counsel could have pursued—requires record evidence that clearly indicates counsel possessed sufficient information to merit considering an alternative strategy or 25 tactic. Ultimately, this inquiry requires the defendant to identify unpursued strategies and tactics that were obviously in the defendant’s interest under the circumstances. ¶59 Once the defendant identifies a plausible alternative that counsel might have pursued, he must show that the alternative was objectively reasonable under the facts known to counsel at the time of the strategic decision. Like the Nicholson court, we emphasize that “the ultimate question involves a conclusion of law reached under an objective standard, whether, considering the facts known to the lawyer, the alternative defense strategy was ‘objectively reasonable.’” Nicholson, 611 F.3d at 206–07. This inquiry “does not induce the court to contemplate whether the alternative strategy was subjectively reasonable to the lawyer, nor does it require or permit the court to view the lawyer’s performance under the ‘highly deferential’ standard spelled out in Strickland.” Id. at 207 (citing Strickland, 466 U.S. at 689). ¶60 Instead, the inquiry properly focuses on factors that will vary from case to case, including, but not limited to, the charge(s) against the defendant, the evidence, the information that the defendant communicated to the attorney and upon which the attorney based his decisions, the attorney’s ethical obligations, the likelihood that pursuing the alternative strategy would damage the defendant’s credibility and jeopardize his chances for future Crim. P. 35(c) relief, and the alternative strategy’s viability given all of the above. See, e.g., Mickens, 240 F.3d at 361–62; see also Nicholson, 611 F.3d at 208–11. ¶61 The final prong—that counsel’s failure to pursue the strategy or tactic was linked to the actual conflict—can be proved in two ways: first, by showing that the 26 alternative strategy or tactic was “inherently in conflict with . . . the attorney’s other loyalties or interests”; or, second, by showing that the alternative strategy or tactic was “not undertaken due to those other loyalties or interests.” Nicholson, 611 F.3d at 212 (emphasis added) (citation and internal quotation marks omitted). ¶62 An alternative strategy or tactic is “inherently in conflict with counsel’s other loyalties or interests” if the two are “inconsistent with each other.” Id. at 213 (citation and internal quotation marks omitted). This inquiry does not consider counsel’s subjective belief that he forewent the alternative strategy for reasons unrelated to the conflict. ¶63 The second subpart recognizes that not all conflicts of interest cases involve an inherent conflict. Where there is no inherent conflict, but problematic circumstances exist, a defendant must prove that the alternative strategy or tactic was not pursued due to the attorney’s other loyalties or interests. While the defendant does not have to present unequivocal proof of a link, record evidence should strongly indicate that counsel’s failure to pursue the alternative strategy resulted from a “struggle to serve two masters.” Sullivan, 446 U.S. at 349 (quoting Glasser v. United States, 315 U.S. 60, 75 (1942)).11 In quantifying the requisite level of evidence, we do not suggest that this inquiry deviates in any way from the defendant’s burden to establish a link by a preponderance of the evidence. Rather, we intend only to assist courts in determining whether a defendant has satisfied this inquiry. 11 27 ¶64 In holding that a defendant must satisfy the three-part standard adopted today in addition to proving a conflict of interest exists, we overrule Castro to the extent that it holds otherwise. IV. Conclusion ¶65 In order to prevail on an ineffective assistance of counsel claim predicated on trial counsel’s alleged conflict of interest arising from concurrent or successive representation of trial witnesses against a defendant, we hold that the defendant must show by a preponderance of the evidence both a conflict of interest and an adverse effect resulting from that conflict. To show an adverse effect, a defendant must (1) identify a plausible alternative defense strategy or tactic that trial counsel could have pursued, (2) show that the alternative strategy or tactic was objectively reasonable under the facts known to counsel at the time of the strategic decision, and (3) establish that counsel’s failure to pursue the strategy or tactic was linked to the actual conflict. A defendant may prove the link under the third prong by showing that the alternative strategy or tactic was inherently in conflict with counsel’s other loyalties or interests or by showing that the alternative strategy or tactic was not undertaken due to those other loyalties or interests. ¶66 The record supports petitioners’ allegations that their attorneys represented the potentially conflicting interests of trial witnesses. Therefore, we now remand the cases to the court of appeals to return them to the trial courts to determine whether the alleged conflicts adversely affected their counsels’ performance and thereby created actual conflicts. 28 ¶67 We affirm in part the court of appeals’ judgments in both West and Cano and instruct the trial courts to consider whether, under Sullivan and consistent with this opinion, West and Cano received ineffective assistance of counsel by virtue of their attorneys’ alleged conflicts of interest and are therefore entitled to new trials. JUSTICE COATS dissents, and JUSTICE EID and JUSTICE MÁRQUEZ join in the dissent. 29 JUSTICE COATS, dissenting. ¶68 While I applaud the majority for acknowledging our longtime misinterpretation of the limited exception to the Strickland ineffective assistance standard for actual conflicts of interest, I nevertheless believe the majority continues to understand that exception far too broadly. Unlike the majority, I consider it clear that the Sullivan exception, or prophylaxis, was never intended to provide a different standard for conflicts of interest generally, but instead was created for, and has been applied by the Supreme Court solely to, the inherently prejudicial situation posed by the simultaneous representation of multiple defendants accused of jointly committing the same offense. Because I believe the majority, like those inferior federal tribunals upon which it chooses to rely, misreads (or is at least unmoved by) the applicable jurisprudence of the Supreme Court, and because there is no assertion here that the Strickland standard has itself been met, I would simply reverse the judgment of the court of appeals. I therefore respectfully dissent. ¶69 In 1984, the Supreme Court articulated for the first time a comprehensive understanding of ineffective assistance, making clear that this aspect of the Sixth Amendment right to counsel merely protects a criminal defendant from being prejudiced by a deficient performance from his counsel. United States v. Cronic, 466 U.S. 648, 654 (1984); Strickland v. Washington, 466 U.S. 668, 692 (1984). Excepting only a few narrow circumstances in which the Court had previously presumed prejudice— notably where counsel was not made available, was prohibited by the trial court from participating in a critical aspect of the proceeding, or acted under a conflict of interest— 1 this constitutional right is now held to be violated only upon a demonstration of likely prejudicial impact on the outcome of a particular adjudication. See Mickens v. Taylor, 535 U.S. 162, 171 (2002); Strickland, 466 U.S. at 687 (1984); Cronic, 466 U.S. at 665; see generally 3 Wayne R. LaFave et al., Criminal Procedure, § 11.7(d) (3d ed. 2013). With regard to the exception for conflicts of interest, several years prior to finally arriving at its comprehensive standard in Strickland, the Court found it appropriate to presume prejudice where a trial court declines to inquire further into defense counsel’s assertion that he will be unable to adequately represent the interests of multiple codefendants at the same trial. Holloway v. Arkansas, 435 U.S. 475, 485 (1978). Shortly thereafter it held that relief from an alleged conflict arising from such joint representation, which is raised initially only after conviction in a case in which the trial court neither knew nor had reason to know of any particular conflict requiring further inquiry, would be contingent upon a demonstration that the conflict actually affected the adequacy of counsel’s representation. Cuyler v. Sullivan, 446 U.S. 335, 349–50 (1980). ¶70 Although it has continued to recognize this standard for assessing the impact of “actual conflicts,” as an exception to the prejudice prong, or requirement, of Strickland, in Mickens, the Court clarified the scope of the exception in two important respects. First, it emphasized that “the Sullivan standard is not properly read as requiring inquiry into actual conflict as something separate and apart from adverse effect.” Mickens, 535 U.S. at 172 n.5. Rather, “[a]n ‘actual conflict,’ for Sixth Amendment purposes, is a conflict of interest that adversely affects counsel’s performance.” Id. 2 Acknowledging this unambiguous proposition from Mickens, the majority now overturns our holding to the contrary in People v. Castro, 657 P.2d 932, 945 (Colo. 1983). ¶71 In addition, however, the Supreme Court openly disparaged lower federal appellate court holdings applying Sullivan “‘unblinkingly’ to ‘all kinds of alleged attorney ethical conflicts.’” Mickens, 535 U.S. at 174–75. Instead, the Court emphasized the limited applicability of Holloway and Sullivan, both of which “stressed the high probability of prejudice arising from ‘multiple concurrent representation,’ and the difficulty of proving that prejudice.” Id. It is the interpretation of this second limitation on Mickens, with regard to the kinds of conflicts susceptible of categorization as actual (rather than merely potential) for purposes of the Sullivan prophylaxis, as to which I part company with the majority. ¶72 The “multiple concurrent representation,” id. at 175, for which the “Sullivan prophylaxis” has been considered necessary, was in no way intended as a reference to concurrent representations of every conceivable kind, simply because they were concurrent. Rather, it was a clear and unmistakable reference to the multiple concurrent representation of defendants for the same offense, the representations with which the Court was faced in both Holloway and Sullivan. Concurrent representation of different clients generally, of course, does not present an ethical concern at all, unless the representation of one will be directly adverse to another or there is a significant risk that the representation of one will be materially limited by counsel’s responsibilities to the other. See generally Colo. RPC 1.7 (Conflict of Interest: Current Clients). Unlike concurrent, or for that matter even successive, see Colo. RPC 1.9 (Duties to Former 3 Clients), representation of different clients, which may under certain circumstances, but generally does not, involve conflicting interests, the potential for conflict in the representation of codefendants has long been recognized as so grave that defense counsel are ordinarily admonished from even accepting such representation. See ABA Standards for Criminal Justice, 4-3.5 Conflicts of Interest (3d ed. 1993); see generally 3 Wayne R. LaFave et al., Criminal Procedure, § 11.9(a) (3d ed. 2013). While the Court did not unequivocally restrict the applicability of the Sullivan exception to joint representation of defendants being tried for the same crime, it did make clear that this problem of joint representation was the impetus for developing the standard in the first place and that its applicability beyond this narrow class of cases could extend to other kinds of conflicts, if at all, only to the extent that with regard to those other kinds of conflicts the Strickland standard would, for similar reasons, be unable to assure vindication of the defendant’s Sixth Amendment right to counsel. Mickens, 535 U.S. at 174–75.1 ¶73 The Mickens Court did not draw a conceptual line of demarcation between concurrent and successive representation, leaving the applicability of the Sullivan The Court’s pre-Mickens case of Wood v. Georgia, 450 U.S. 261 (1981), is offered by the majority as an example of a broader application of the exception, but Wood clearly involved joint representation concerning the same offense and differed from the joint representations in Holloway and Sullivan, at most, in that the conflict of concern arose between the defendants and their employer, who paid for the representation, rather than among the codefendants themselves. As the Fifth Circuit noted in Beets v. Scott, 65 F.3d 1258, 1267 (5th Cir. 1995), an opinion cited favorably in Mickens for criticizing the broader application of Sullivan to conflicts generally, Mickens, 535 U.S. at 174, while Wood did not indicate whether counsel formally represented the employer as well, he “was at least in the functional equivalent of a joint representation.” 1 4 prophylaxis to the latter an open question while implying its applicability to the former, as the majority holds. Rather, it distinguished successive from concurrent representation only with regard to the representation of defendants either jointly charged or joined for trial as permitted by the federal rules and only in demonstrating that “[n]ot all attorney conflicts present comparable difficulties.” Id. at 175. Similarly, it singled out successive representation as a kind of representation as to which the applicability of the Sullivan prophylaxis remained an open question only by way of emphasizing that the question of successive representation, the precise factual situation presented in Mickens itself, was not argued by the parties and was not before the Court on certiorari review. Nothing in the Supreme Court’s opinion in Mickens remotely suggests an intent to extend the Sullivan prophylaxis beyond joint representation to either successive representation or concurrent representation of every kind.2 To the contrary, Mickens makes abundantly clear that applicability of the Sullivan exception may be extended beyond the specific “multiple concurrent representation” situations at issue in Holloway and Sullivan only to other situations, should they exist, with a comparably high probability of, and comparably high difficulty of proving, prejudice. I do not believe, and the majority does not assert or make any attempt to demonstrate, At one point in its analysis, the majority asserts that by assuming that the Sullivan exception applies to successive representations, it does no more than this court had already done in Dunlap v. People, 173 P.3d 1054, 1073 n.24 (Colo. 2007). See maj. op. ¶ 38. There, of course, we could assume, without deciding, the applicability of the Sullivan exception because we found there to be no adverse effect, and therefore no ineffective assistance, in any event. Here the majority bases its remand order on the proposition that the existence of adverse effect may be determinative of the question of effective assistance. 2 5 that the concurrent representations of witnesses or alternate suspects alleged in these two cases fall within a class of conflicts sharing a comparably high probability of prejudice with the joint representations dealt with by the Court in Holloway and Sullivan.3 ¶74 Apparently unconvinced by its own broad reading of Mickens, as extending the applicability of the Sullivan prophylaxis beyond the joint representation of codefendants, the majority finds succor in the fact that the conflict at issue in Mickens itself actually involved prior representation of the victim rather than joint representation of a codefendant. At least three different times, the majority attempts to bolster its interpretation with the observation that the Court could therefore have resolved the case simply as one not involving joint representation, had it indeed considered that fact dispositive. But the inference drawn by the majority from the Court’s failure to rule on such alternate ground is precisely the inference the final third of Mickens is devoted to warning against. In excruciating detail, the Supreme Court explains that “[l]est today’s holding be misconstrued,” the only question presented, argued, and accepted for review was whether the Sullivan exception would still require a demonstration of deficient performance by defense counsel. Although it is openly critical of application of the Sullivan exception by various lower federal courts to By contrast, notwithstanding the majority’s assertion that most federal circuits apply Sullivan to cases of successive representation, see maj. op. ¶ 37, I understand those circuits as wrestling with the individual circumstances of each case to determine whether they are more or less like the conflicts as to which the Supreme Court has found the Sullivan prophylaxis necessary, rather than as accepting the applicability of Sullivan to successive representations as a class. 3 6 “conflicts” writ large, the Court emphasizes that, for the reasons stated, no inference may be drawn from its failure to exclude from the Sullivan prophylaxis the kind of conflict at issue in its own case. ¶75 The majority actually refers to the Supreme Court’s concluding caution concerning the non-fungibility of conflicts as “dicta.” “However, there is dicta and then there is dicta, and then there is Supreme Court dicta.” Schwab v. Crosby, 451 F.3d 1308, 1325 (11th Cir. 2006) (criticizing the Second Circuit for similarly making light of the Supreme Court’s so-called “postscript” in Mickens). While the final third of Mickens itself implicitly acknowledges that it is not necessary to the resolution of the question before the Court, a majority of the Court (not merely a plurality) clearly considers it important to make express precisely what it has not decided and why it has not done so. When the Supreme Court presents us with a “thoroughly reasoned, and carefully articulated analysis . . . describing the scope of one of its own decisions . . . , as an entire, separately enumerated section of [its] opinion—three long, citation-laden paragraphs, consisting of more than five hundred words,” id., I, for one, consider it meaningful. ¶76 Even if I were to agree that the Sullivan prophylaxis could apply to the conflicts alleged in these cases, I would nevertheless reject the Fourth Circuit’s tri-partite test as confusing, unnecessary, and a questionable proliferation of special multi-factor tests neither mandated nor even supported by Supreme Court jurisprudence. In Mickens, the Supreme Court could not have more clearly held that the Sullivan prophylaxis excuses individual inquiry into whether counsel’s inadequate performance undermined the reliability of the verdict—that is, whether it was prejudicial—not whether counsel’s 7 performance was deficient in the first place. Mickens, 535 U.S. at 166–67. Where counsel has not sooner openly conceded his inability to represent codefendants because of their conflicting interests, as in Holloway, a requirement that the multiple representation at issue have an adverse effect clearly communicates that counsel’s performance must have actually been deficient and that the deficiency in his performance must have resulted from a particular conflict arising from that representation. Reliance on the Sullivan prophylaxis to excuse the second prong of Strickland requires no more than that the defendant establish the first prong of Strickland and, in place of the second, establish a causal connection between counsel’s deficient representation of the defendant and a conflict arising from the qualifying multiple representation, however broad that may be. ¶77 Rather than suggesting a parallel universe for attorney-conflict challenges, the Supreme Court merely excuses proof that the outcome of the defendant’s trial would likely have been different, upon a showing that his counsel’s performance was deficient and that the deficiency was caused by the qualifying multiple representation. There is no shortage of case law, in the Strickland context, assessing whether counsel’s choices concerning the pursuit or disregard of particular defenses, examination of witnesses, requests for or objections to jury instructions, and virtually every other kind of choice made (or neglected) in conducting a defense were plausible tactical choices or fell below the standard of reasonable competence. In contrast to the tri-partite scheme adopted by the majority, the only element of the Sullivan prophylaxis not already included in the 8 Strickland test is a causal connection between counsel’s deficient performance and a particular kind of conflict with a high probability of prejudicing the defendant’s case. While there will always be room for debate about the nature and proximateness ¶78 required of the causal connection at issue, there can be no question that Sullivan simply excuses the second prong of the Strickland test upon a showing that the first prong resulted from a particular kind of conflict. Because the lower federal courts do not interpret federal constitutional law or Supreme Court jurisprudence for this court, and because I find the court’s reasoning in United States v. Nicholson, 475 F.3d 241, 249 (4th Cir. 2007), particularly unpersuasive, I would not saddle the nisi prius courts of this jurisdiction with the same findings required of those federal district courts subject to Fourth Circuit review. ¶79 Because I do not believe the prejudice exception of Sullivan can apply to the kinds of representation involved in these cases and because, in any event, neither defendant has asserted, much less proved, that the defense provided to him fell below the standard of reasonable competence, I would simply reverse the judgment of the court of appeals. ¶80 I therefore respectfully dissent. I am authorized to state that JUSTICE EID and JUSTICE MÁRQUEZ join in the dissent. 9