HALLIGAN, J.
Matthew C. Hug, for appellant.
Emily Schultz, for respondent.
HALLIGAN, J.
The defendant raises various challenges to his convictions for two counts of murder in the second degree and two counts of tampering with physical evidence. We conclude that the trial court did not err in denying the defendant's motion to sever the charges relating to each murder from each other. With respect to the improper, undisclosed business relationship between the defendant's trial counsel and an Assistant District Attorney (ADA) involved in the defendant's prosecution, the record supports the trial court's determination that although this arrangement created a potential conflict of interest, it did not operate on the defense and thus did not require vacatur of the conviction. As to the defendant's remaining claims, we conclude that they are either without merit or unreviewable, and affirm.
This appeal arises from Edward Mero's convictions for two separate murders, committed almost two years apart and tried together. The first victim was the defendant's roommate, who was found dead in their shared apartment in 2013 after a fire that occurred under circumstances that later raised questions about her cause of death. The second victim was a woman whom defendant had hired to go on a date with him in December 2014; her body was found in a shallow grave in May 2015. The defendant was arrested in 2017 and charged with two counts of murder in the second degree and two related counts of tampering with physical evidence.
The charges were joined in a single indictment, and the defendant moved to sever them pursuant to CPL 200.20 (3). He argued that the cases "could not be any more different" because the victims were unrelated and their deaths occurred nearly two years apart under dissimilar circumstances, that his defenses for each would be different, and that the evidence against him for each murder was "underwhelming." The trial court denied the motion, explaining that the distinct evidence relating to each murder would enable the jury to consider separately the proof for each charge, and joinder thus would not unduly prejudice the defendant. The jury convicted the defendant of both murders and the associated tampering charges.
During trial, defense counsel reported an incident to the trial court that ultimately resulted in dismissal of a juror. Counsel explained that a juror had been present during a "boisterous" and "inflammatory" conversation in the courthouse lobby. She admitted to the court that she had made "derogatory comments" about her client . The parties agreed to strike the juror.
During post-trial proceedings, counsel denied participating in the exchange and asserted she was merely present for the comments.
In December 2019, the defendant moved to vacate his convictions under CPL 440.10, arguing that his trial counsel and the ADA who prosecuted him had an improper business relationship that constituted a conflict of interest. A hearing revealed that the defendant's trial counsel had paid the ADA to write several briefs for her clients over a four-year period. The trial court denied the motion to vacate, concluding that although there was a potential conflict of interest, it did not operate on the defense.
The defendant appealed the judgment of conviction and sentence and the order denying his CPL 440 motion, claiming, among other things, that the trial court abused its discretion in denying the motion to sever and erred in denying the motion to vacate due to a conflict of interest.
The Appellate Division affirmed the judgment and the order. On severance, the Court reasoned that the defendant's characterization of the evidence for each murder as "underwhelming" belied any suggestion that proof of one murder was more significant than proof of the other, and that the defendant's argument that the crimes "could not be any more different" refuted his contention that the jury would struggle to consider them separately (see 221 A.D.3d 1242, 1250 [3d Dept 2023]). The Appellate Division agreed with the trial court that the potential conflict of interest did not operate on the defense and rejected the defendant's other challenges to the conviction.
Two Justices dissented with respect to severance, taking the view that the proof of the second murder was "significantly more abundant in quantity and significant in scope" than the proof relating to the first murder (id. at 1253 [Reynolds Fitzgerald, J., dissenting in part]). The dissenting Justices concluded that, even with limiting instructions, the jury would likely "focus on the abhorrent common nature of the crimes," rather than "the fundamental differences of proof" (id.). A dissenting Justice granted leave to appeal (40 N.Y.3d 1095 [2024]), and we now affirm.
We begin with the defendant's motion to sever the charges related to the first murder from those related to the second murder. CPL 200.20 (2) (c) allows joinder of offenses "based upon different criminal transactions" that "are defined by the same or similar statutory provisions and consequently are the same or similar in law" (CPL 200.20 [2] [c]). There is no dispute that joinder of the offenses under this provision was proper in the first instance. The defendant was charged with two counts of murder in the second degree and two corresponding counts of tampering with physical evidence, and acknowledged that joinder was "permissible" in moving to sever. Additionally, there were several overlapping witnesses who testified regarding both murders, namely the medical examiner and a jailhouse informant to whom the defendant had confessed committing both murders.
Where counts have been joined under CPL 200.20 (2) (c), the CPL provides that the trial court has discretion to sever them if doing so would be "in the interest of justice and for good cause shown" (id. 200.20 [3]). This provision has its origins in a 1936 amendment to the Code of Criminal Procedure that allowed joinder of "two or more acts or transactions constituting crimes of the same or a similar character which are neither connected together nor parts of a common scheme or plan" (Code Crim Proc § 279 [eff 1936]).
The statute specifies two situations that establish good cause: first, where there is "[s]ubstantially more proof on one or more such joinable offenses than on others and there is a substantial likelihood that the jury would be unable to consider separately the proof as it relates to each offense"; and second, where there is "[a] convincing showing that a defendant has both important testimony to give concerning one count and a genuine need to refrain from testifying on the other, which satisfies the court that the risk of prejudice is substantial" (CPL 200.20 [3] [a], [b]). The statute expressly notes that good cause is not limited to these grounds (see id. 200.20 [3]), and we have explained that "a defendant's fundamental right to a fair trial free of undue prejudice" must be protected when offenses are joined for trial (People v Lane, 56 N.Y.2d 1, 8 [1982]). Denial of a defendant's motion to sever is reviewed for abuse of discretion (see People v Ford, 11 N.Y.3d 875, 879 [2008]).
The defendant's arguments before us on severance are cursory, and primarily summarize the points made by the Appellate Division dissent. The defendant asserts that "the jury was incapable of considering the proof separately with respect to each offense" because the proof of each crime was "markedly different" and "dissimilar." But that is generally true whenever offenses are joined pursuant to CPL 200.20 (2) (c), and the key question is whether it would be difficult for the jury to consider separately the proof for each offense, a point we address below.
The defendant does not show there was "[s]ubstantially more proof" of the first offense than the second (CPL 200.20 [3] [a]). Rather, he characterized the amount of proof for each as similarly "underwhelming," and asserted that there was a "clear danger" the jurors would treat the evidence cumulatively because of the "abhorrent nature of the allegations." Even if we were to accept that characterization of the evidence, the question, again, is whether the jury would find it difficult to segregate the proof for each set of offenses (see Ford, 11 N.Y.3d, at 879; United States v Werner, 620 F.2d 922, 929 [2d Cir 1980] [despite the risk that juries may use evidence cumulatively where charges are joined for trial," 'the objection disappears'" provided that" 'the accused's conduct on several separate occasions can properly be examined in detail'" and" 'the only consideration is whether the trial as a whole may not become too confused for the jury'" (quoting United States v Lotsch, 102 F.2d 35, 36 [2d Cir 1939] [L. Hand, J.])]).
As both the trial court and the Appellate Division concluded, the defendant did not make that showing here. The trial court gave thorough instructions to the jury, once before trial began, once before jury deliberation began, and once in response to a note requesting a readback of the instructions . Absent evidence to the contrary, we assume the jurors followed these instructions (see People v Stone, 29 N.Y.3d 166, 171 [2017]; see also People v Adames, 42 A.D.3d 328, 329 [1st Dept 2007] ["(T)he court's jury instructions were sufficient to prevent the evidence of (one incident) from having any prejudicial effect with respect to the (joined incident)"]; cf. People v Davis, 225 A.D.3d 62, 75 [1st Dept 2024] ["(I)t cannot be fairly said that the evidence was easily segregable in the minds of the jurors where the trial court failed to give a limiting instruction to that effect"] [internal quotation marks and citation omitted]) . Given the facts presented here, the defendant failed to demonstrate "a substantial likelihood that the jury would be unable to consider separately the proof as it relates to each offense" (CPL 200.20 [3] [a]).
The court instructed as follows: "Now, you have observed that the People have joined in this single indictment two separate and distinct cases: On or about January 27, 2013; and on or about December 3, 2014. Now, these cases are joined in a single indictment for reasons that do not concern you. The fact that a defendant is charged with one crime constitutes no proof that he committed another crime also charged in a single indictment. You are instructed to segregate and to keep separate in your minds the evidence applicable to each of the cases presented to you. You may not infer guilt on one accusation simply because there is another accusation as well, nor may you make any negative inference because the defendant has been twice accused. You are to vote on each case separately and solely on the basis of the evidence applicable to that case. Proof on one case must not be considered as proof in the other case. It is your obligation to evaluate the evidence as it applies or fails to apply to each case separately. Each instruction on the law must be considered by you as referring to each case separately. You must return a separate decision for each case, and those decisions may be but need not be the same. It is your sworn duty to give separate consideration to each individual case."
People v Stanard (32 N.Y.2d 143 [1973]), on which Judge Rivera in dissent relies (see Rivera, J., dissenting op at 15), does not suggest otherwise. That case had nothing to do with severance, and our fact-specific holding that an instruction would not cure a combination of extensive testimony on unrelated criminal activity and a prosecutor's improper suggestion that the defendant could be held accountable for that conduct (see Stanard, 32 N.Y.3d at 148) does not bear on whether the jurors here could segregate the proof for the two murders.
The jury notes, which requested a readback of the medical examiner's testimony only as it related to the second victim, and days later, the testimony regarding only the first victim, confirm that the jury did consider separately the proof for each offense.
The defendant and Judge Rivera in dissent suggest that joinder inevitably prejudiced the defendant given the "heinous" nature of the charged crimes (Rivera, J., dissenting op at 13). However, the defendant points to nothing akin to the "unique" and "peculiar" context in People v Shapiro that mandated severance-circumstances that included what this Court termed "aberrant sexual practices," the aggregation of 64 counts of one offense with 2 counts other, more serious offenses, and a request by the defendant to testify regarding one set of offenses, but not the other (50 N.Y.2d 747, 755-756 [1980]). Moreover, we note that the Appellate Division Departments have permitted joinder of homicides on numerous occasions (see e.g. People v Bonner, 94 A.D.3d 1500, 1501 [4th Dept 2012]; People v Perez, 47 A.D.3d 409, 411 [1st Dept 2008]; Adames, 42 A.D.3d at 329; People v Oliveira, 2 A.D.3d 122, 123 [1st Dept 2003]), and nothing in our case law suggests joinder is impermissible solely because of the nature of the crime.
None of the other factors to which Judge Rivera points indicates that the trial court abused its discretion in declining to sever (see Rivera, J., dissenting op at 12-14). The dissent notes the People's statement at oral argument that proof of the second murder was stronger than proof of the first (see id. at 12), but proof of joined offenses will rarely be identical in strength. Additionally, CPL 200.20 (3) (a) provides for severance if the jury cannot segregate proof-a showing which, as noted, was not made here. The dissent describes the trial as "complex" (Rivera, J., dissenting op at 14), but that does not mean the jury could not segregate the evidence of each charge.
Finally, both dissents make sweeping contentions regarding the CPL's joinder and severance provisions. Judge Rivera takes the view that "wholly unrelated offenses" cannot be joined "because of the inherent prejudice to a defendant from propensity evidence" (Rivera, J., dissenting op. at 8-9; see also id. at 9 n 1). But joinder of factually unrelated offenses does not necessarily present a greater risk of prejudice than joinder of factually related offenses (see e.g. People v Forest, 50 A.D.2d 260, 262 [1st Dept 1975] [risk of "conviction by reason of () cumulative effect" required severance where "all three crimes (were) of a similar nature"]; cf. Herring v Meachum, 11 F.3d 374, 378 [2d Cir 1993] ["(B)ecause the evidence with respect to each murder was distinct and easily compartmentalized, the risk of jury confusion at petitioner's trial was significantly limited"]). The Chief Judge would go even further, saying that offenses may be joined under CPL 200.20 (2) (c) only if all the evidence of one offense would be admissible in the case in chief of the other (see Wilson, Ch. J., dissenting op at 2). But the Legislature addressed joinder under this circumstance when it enacted CPL 200.20 (2) (b), which authorizes joinder of cases in which "proof of the first offense would be material and admissible as evidence in chief upon a trial of the second" (or vice versa) (CPL 200.20 [2] [b]). In any event, all that is before us are the arguments raised by the defendant, who did not contest that CPL 200.20 (2) (c) authorized joinder in the first instance and does not press the dissents' various points about legislative history or critiques of our joinder scheme.
Accordingly, we hold that the trial court did not abuse its discretion in denying the defendant's motion to sever the charges against him.
The defendant argues that the improper business relationship between his defense counsel and an ADA who was prosecuting him created an impermissible conflict of interest that required vacatur of his conviction. He first contends that the arrangement created an actual conflict of interest, which requires reversal if not waived (see People v Solomon, 20 N.Y.3d 91, 95 [2012]). Such a conflict arises when an attorney "has divided and incompatible loyalties within the same matter necessarily preclusive of single-minded advocacy" (People v Brown, 33 N.Y.3d 983, 987 [2019] [cleaned up and internal quotation marks omitted]). That is not what occurred here. The ADA was paid for drafting four appellate briefs and a motion for a change of venue in other cases litigated outside of the county where this prosecution took place. The record indicates that the work was performed on nights and weekends, without access to defense counsel's office or files.
The defendant also argues, in the alternative, that the arrangement created a potential conflict of interest necessitating reversal. We have held that reversal of a conviction is required if a defendant shows that a potential conflict actually operated on the conduct of his defense (see People v Sanchez, 21 N.Y.3d 216, 223 [2013]). Evidence that unconflicted counsel would have acted in a "more vigorous, less inhibited manner" may establish this point (People v Harris, 99 N.Y.2d 202, 211 [2002]). The inquiry presents a mixed question of law and fact, and thus we will reverse only if the Appellate Division's determination lacks record support (see People v Ennis, 11 N.Y.3d 403, 411 [2008]).
We certainly do not condone the conduct of trial counsel. But the trial court characterized defense counsel's advocacy as "meaningful" and "zealous[] and effective[]" notwithstanding the incident that led to the removal of the juror within earshot of the disparaging comments about the defendant. The juror was dismissed with the consent of all parties, and no other jurors heard the remarks. The defendant has produced no evidence suggesting the business conflict had anything to do with the removal of the juror, or that it otherwise operated on his defense.
Nor is the conflict here akin to that in People v Shinkle (51 N.Y.2d 417 [1980]). There, the defendant's attorney left his position with the Legal Aid Society in the months preceding the defendant's trial and joined the District Attorney's office in a senior leadership position (id. at 420). We held that the resulting "unmistakable appearance of impropriety" and "continuing opportunity for abuse of confidences" was significant enough to disqualify the entire District Attorney's office from handling the prosecution (id.). Indeed, this Court noted that the attorney had extensively interviewed the defendant, "was intimately familiar with the contents of his file[,] and assisted in the formulation of defense strategy" (id. at 419). The circumstances here are different. The prosecutor here never represented the defendant, and the record indicates that defense counsel did not share information or confidences with the prosecutor regarding the defendant. The defendant's reliance on People v Wandell is similarly misplaced (75 N.Y.2d 951 [1990] [Mem]). We limited our holding to the unique circumstances presented there: the concurrent representation of a defendant and the People's chief witness against him, following a reversal in a prior case for a similar omission by the same attorney (see id. at 952).
The defendant's remaining claims are meritless. First, the defendant argues that he was deprived effective assistance of counsel, noting defense counsel's conflict of interest, the incident that led to the removal of a juror, and various other points about trial strategy. As the trial court explained, the disparaging comments made in the courtroom lobby "clearly operated on the defense of the case" to the extent it resulted in the dismissal of a juror. But the court also determined that this incident was unrelated to the potential conflict of interest between defense counsel and the ADA, that the ADA "was not present during the lobby fiasco," and that there was "no causal relationship between [defense counsel's] disparaging remarks and the business relationship with [the ADA]." The trial court further held that defense counsel provided meaningful representation and was detailed in her work. The defendant makes several other claims about the performance of his trial counsel, including the decision to forgo calling several witnesses and not to investigate a claim about a trooper's offhand remark. But the defendant failed to demonstrate that these decisions were not strategic.
Second, the defendant argues that his convictions are unsupported by legally sufficient evidence, but the record refutes this claim. As to the first murder, the evidence at trial showed that the defendant and the victim had a loud argument the night she died, the victim's body showed signs she was dead before the fire started, the defendant bragged to coworkers about his ability to get rid of a body by arson and offered contradictory alibi statements to the police, and a jailhouse informant testified that the defendant confessed the murder and arson to him. With respect to the second murder, the evidence showed that defendant was the last person to see the victim, was one of few people with access to the property on which her body was found, got rid of his car shortly after the victim disappeared, implicated himself in an interview with police, and confessed to two jailhouse informants who testified to specific details of the murder.
Third, the defendant contends that two men he met in jail, each of whom testified at trial that the defendant had confessed to one or both of the murders, were acting as agents of the state and their statements should therefore have been suppressed. As the Appellate Division correctly noted, a defendant must be able to show that the State was "actively involved in, or even encouraged" the decision to volunteer the statements (221 A.D.3d at 1248 [internal quotation marks omitted]). The record lacks evidence to that effect.
Fourth, the defendant argues that admission of a text message and a short colloquy with investigators constituted an impermissible comment on defendant's invocation of his right to silence. Without considering the merits of this claim, we conclude that any error was harmless given the court's curative instruction (cf. People v McLean, 243 A.D.2d 756, 756-757 [3d Dept 1997]). Finally, the defendant's weight of the evidence claim is unreviewable in this Court (see People v Bleakley, 69 N.Y.2d 490, 495 [1987]; People v Danielson, 9 N.Y.3d 342, 349 [2007]).
Accordingly, the order of the Appellate Division should be affirmed.
WILSON, Chief Judge (dissenting)
I agree with my dissenting colleague that our fundamental prohibition on convictions based on propensity requires us to reverse and remit for two separate trials. I write separately because, where two unrelated crimes are charged, our Molineux doctrine must take precedence over administrative convenience. I therefore interpret CPL 200.20 (3) as incorporating the Molineux rule by requiring severance unless the evidence from each of the joined charges would fit under a Molineux exception.
It is helpful to think of this case in terms of Molineux. Mr. Mero was tried jointly for the murders of two young women. One woman was buried in a shallow grave; her badly decomposed body showed that she had suffered significant blunt force trauma to her head and face. The other young woman was Mr. Mero's roommate. Her badly charred body was recovered after a fire in their shared apartment, where Mr. Mero had been hours earlier; the body showed no signs of smoke inhalation, suggesting she was dead before the fire began.
Suppose, for a moment, that Mr. Mero had first been tried separately for the murder of the woman found in a shallow grave. Whether he had been convicted or acquitted of that murder, our Molineux rule would have prevented the introduction of the evidence against him, the fact of conviction (or acquittal), or the fact of arrest, from introduction in a separate trial for the murder of his roommate. The same would be true if he had been tried first, separately, for the murder of his roommate. There is no argument that any Molineux exception applies here. The efficiency arguments, such as they are, are that the same medical examiner and jailhouse informant were involved in both cases. But our Molineux doctrine does not allow the convenience of witnesses to overcome Mr. Mero's right to a fair trial.
It is also worth thinking about the following question: why did the People want to try these two unrelated crimes together? If the evidence for each crime was very strong, the People should have wanted two chances-with two separate juries-to put Mr. Mero away for 25 to life. It is only if the evidence for one or both was not very strong that the People would sacrifice two bites at the apple for one. That is precisely what our Molineux rule prevents.
I
I begin from the same well-established premise as Judge Rivera: a defendant may not be convicted on evidence of criminal propensity (People v Molineux, 168 NY 264, 291 [1901]). We do not permit proof of guilt based on prior bad acts or prior convictions, except in very limited circumstances, such as to prove identity where the similarities between two offenses are "unusual enough to compel the inference that the defendant committed both" and identity is not otherwise conclusively established (People v Agina, 18 N.Y.3d 600, 603 [2012], quoting People v. Beam, 57 N.Y.2d 241, 251 [1982]), or to prove intent "when proof of the act falls short of demonstrating that the defendant acted with a particular state of mind and where proof of a prior act is relevant to that issue" (People v Alvino, 71 N.Y.2d 233, 242 [1987]). Unless a Molineux exception applies, the defendant's guilt must be proved by evidence related to the charged crime only. Based on that fundamental rule, we have repeatedly held that evidence of past crimes is inadmissible unless it serves a non-propensity purpose, and its probative value outweighs the risk of prejudice (see e.g., People v Weinstein, ___ N.Y.3d ___, 2024 NY Slip Op 02222 at *9, 2024 WL 1773181 [Ct. App. Apr 25, 2024]). Like Judge Rivera, I believe that CPL 200.20 (2) (c), which permits joinder of offenses defined by the same or similar statutory provisions, creates the clear risk of convictions based on propensity.
I depart from my dissenting colleague in my analysis of how the Molineux rule should inform our reading of CPL 200.20 (3), which gives courts the authority to sever offenses "in the interest of justice and for good cause shown." Judge Rivera would hold that trial courts abuse their discretion in denying severance where joinder "carries[s] significant risks that the jury would determine guilt based on the appearance of defendant's criminal disposition and... the evidence of guilt on one... count[] would spill over and bolster the evidence of the other," and those risks cannot be overcome (Rivera, J. dissenting op at 2). In applying that rule, she suggests that "considerations of judicial economy" may sometimes justify joinder of offenses (id. at 15-16). In my view, there is no place for such considerations in our analysis. The provisions at issue here were intended to give the state the tools it needed to fight organized crime by allowing prosecutors to join a defendant's interrelated offenses for trial. Nothing in the legislative history indicates a desire to erode the Molineux rule, which was deeply entrenched by the time these provisions were enacted. I would therefore apply our Molineux jurisprudence to hold that good cause to sever exists as a matter of law where, as here, there is no non-propensity purpose for which evidence of one offense could be admitted at the trial of the other. Good cause would also exist where the risk of prejudice outweighs the probative value of joinder.
II
"In matters of statutory... interpretation, 'legislative intent is the great and controlling principle, and the proper judicial function is to discern and apply the will of the enactors'" (Sedacca v Mangano, 18 N.Y.3d 609, 615 [2012]). In 1936, when the legislature enacted the provisions at issue here, the Molineux rule was already recognized as fundamental to New York law. We do not construe statutes to derogate the common law unless they do so clearly . Here, neither the text nor the legislative history of CPL 200.20 (2) (c) and (3) indicates that the legislature intended to displace Molineux. Instead, the "good cause" provision is best read to incorporate the Molineux rule to require severance where evidence of one offense would not be admissible to prove the defendant's guilt of the other.
Molineux has both a common law and a constitutional dimension; our more recent Molineux jurisprudence makes clear that Molineux is grounded in the constitutional right to a fair trial (e.g. People v Telfair, 41 N.Y.3d 107, 117 [2023] ["[C]onsistent with our duty to safeguard the rights of criminal defendants to a fair trial, we have reversed in numerous cases when Molineux evidence was improperly admitted"]).
A
It is a fundamental principle of our justice system that a person is presumed innocent until proven guilty. We have long recognized that the presumption of innocence means that the state may not introduce evidence of a past crime as proof of a person's criminal propensity (People v Molineux, 168 NY 264, 291 [1901]; see also People v Arafet, 13 N.Y.3d 460, 465 [2009] ["The point of Molineux is to prevent a jury from convicting a defendant because of his criminal propensity"]). The Molineux rule "serves as a judicial bulwark against a guilty verdict based on supposition rather than proof, on 'collateral matters or [] because of [a defendant's] past' or on the defendant's 'bad character' alone" (Weinstein, 2024 NY Slip Op at *7, quoting People v Alvino, 71 N.Y.2d 233, 241 [1987]).
The "time-tested rule against propensity evidence" (id.) substantially predates the legislature's enactment, in 1936, of the joinder and severance provisions at issue here. In 1901, we described the Molineux rule as "universally recognized and [] firmly established in all English-speaking lands,... rooted in that jealous regard for the liberty of the individual which has distinguished our jurisprudence from all others, at least from the birth of the Magna Charta" (People v Molineux, 168 NY 264, 291 [1901]). Molineux quoted our 1873 decision in Coleman v People, where we explained:
"It would be easier to believe a person guilty of one crime if it was known that he had committed another of a similar character, or, indeed, of any character; but the injustice of such a rule in courts of justice is apparent. It would lead to convictions, upon the particular charge made, by proof of other acts in no way connected with it, and to uniting evidence of several offenses to produce conviction for a single one" ( id. at 292, quoting Coleman v People, 55 NY 81, 90 [1873]).
As we recognized in 1901, there are exceptions to the broad rule that evidence of one crime may not be introduced in a prosecution for another. Evidence may be relevant, for example, to establish motive, intent, the absence of mistake or accident, a common scheme, or identity (id. at 293). Where evidence of other crimes is relevant to "some issue, other than mere criminal propensity," and its "probative value exceeds the potential for prejudice resulting to the defendant," it may be admitted (Weinstein, 2024 NY Slip Op at *9, quoting People v Hudy, 73 N.Y.2d 40, 55 [1988] and People v Alvino, 71 N.Y.2d 233, 242 [1987]).
"The legislature... is presumed to be aware of the common law" (Gletzer v Harris, 12 N.Y.3d 468, 476-477 [2009], citing McKinney's Cons Laws of NY, Book 1, Statutes § 301 [a]). We therefore strictly construe statutes in derogation of the common law, and common law rules "are to be no further abrogated than the clear import of the language used in the statute" (Matter of Midland Ins. Co., 16 N.Y.3d 536, 547 [2011], quoting Transit Comm. v Long Is. R.R. Co., 253 NY 345, 355 [1930]; see also McKinney's Cons Laws of NY, Book 1, Statutes § 301 [a]; Oden v Chemung County Indus. Dev. Agency, 87 N.Y.2d 81, 86 [1995]; Matter of Excelsior Pictures Corp. v Regents of Univ of State of N.Y., 3 N.Y.2d 237, 245 [1957]). This presumption extends to statutes that derogate from the common law rules of evidence (McKinney's Cons Laws of NY, Book 1, Statutes § 305).
The Molineux rule was firmly established in our jurisprudence long before the legislature enabled the joinder of offenses for trial in 1936. Our pre-1936 decisions lead inexorably to the conclusion that joinder of offenses-especially offenses of the same nature-may lead to convictions based on criminal propensity. Recall that our 1873 decision in Coleman explained that "[i]t would be easier to believe a person guilty of one crime if it was known that he had committed another of a similar character, or, indeed, of any character" (55 NY at 90).
CPL 200.20 (2) (c)-the provision under which the charges against Mr. Mero were joined-makes offenses joinable when they are "defined by the same or similar statutory provisions and consequently... the same or similar in law." To address the potential for prejudice arising from joinder of unrelated offenses, CPL 200.20 (3) provides that when offenses are joined under section (2) (c), the court may "in the interest of justice and for good cause shown... in its discretion, order that any such offenses be tried separately."
We must presume that in enacting section 279 of the Code of Criminal Procedure-the precursor to CPL 200.20-the legislature was aware of the Molineux rule. I therefore would not construe CPL 200.20 to derogate from the common law unless the "clear import" of the language requires it. Nothing in the language or the legislative history of CPL 200.20 suggests that the legislature wanted us to depart from the Molineux rule in determining when good cause exists to sever offenses of the same kind joined for trial. In fact, the legislative history of CPL 200.20 (2) (c) and (3) supports the conclusion that the legislature intended to enable joinder of interrelated offenses for the specific purpose of enabling prosecutors to combat organized crime.
B
CPL 200.20 (2) (c) and (3) were enacted to enable the state to successfully prosecute organized crime. In 1936, Thomas E. Dewey, then the Special Prosecutor for racketeering and organized crime in New York County, proposed legislation that would permit courts to join multiple offenses in one trial . That legislation created section 279 of the Code of Criminal Procedure-the predecessor to CPL 200.20 (2) (c) and (3). In his memorandum to Governor Lehman, Dewey explained the problem section 279 was intended to solve:
The relevant provisions were originally enacted as section 279 of the Code of Criminal Procedure. In early versions of the Criminal Procedure Law, section 279 of the Code became section 100.20. When the new CPL was enacted in 1970, section 279 became CPL 200.20.
"More than a century ago, when the State criminal procedure was established, the criminal problem was simple, crime was isolated and organized crime unknown.... Today, as you well know, crime is syndicated and organized.... Seldom, if ever, does any major criminal commit any crime under circumstances in which apprehension is possible. Only the tool commits the extortion, the assault, the stench bomb throwing or the acts of coercion; and if he be convicted there are a hundred others to take his place while the major criminal enterprise goes on without interruption. The only way in which a major criminal can be punished is by connecting to him, through various layers of subordinates, the related by separate crimes committed on his behalf." (Mem. from Thomas E. Dewey to Governor Herbert H. Lehman, Bill Jacket, L. 1936, ch. 328, at 6).
He wrote that New York law-which prohibited joinder "except in a conspiracy, which is a mere misdemeanor"-created "a procedural strait-jacket... Though the organization is conceived and functions to prey upon hundreds of men in the same status, each of its offenses must at present be tried separately before a separate court and a separate jury" (id. at 6-7).
In fact, according to former Judge George Medalie, Mr. Dewey had a very specific "major criminal" in mind when he proposed section 279: Lucky Luciano. In a 1941 meeting of the federal Advisory Committee on Rules of Criminal Procedure, Judge Medalie told his fellow committee members that the New York statute "was really prepared with a view to Lucciano [sic]" (Minutes of the Advisory Committee on Rules of Civil Procedure [1941], Part 3, at 529-530, https://www.uscourts.gov/sites/default/files/fr_import/CR09-1941-min-Part3.pdf.pdf).
Governor Lehman recommended the bill's passage in a special message to the Legislature in which he read Mr. Dewey's memorandum in its entirety. In addition to repeating Mr. Dewey's warnings about the rise of organized crime, Governor Lehman added his own. He explained that the existing rules "frequently hampered" efforts to "combat the highly organized criminal groups of today," and that the changes would "permit police and prosecutors to convict the 'higher-ups' in racketeering activities," who are "the really dangerous and despicable enemies of society" (Governor's Mem, Bill Jacket, L. 1936, ch. 328, at 9).
Neither the Governor nor Special Prosecutor Dewey cited judicial efficiency as a free-standing goal. To the extent Mr. Dewey referenced efficiency at all, he did so in the specific context of organized crime prosecutions, writing:
"Aside from the great advantage which this procedure gives to professional criminals, the expense to the People in separately prosecuting the separate but interrelated offenses of the single criminal organization is very great. The People should not be compelled to suffer such expense and delay where a single criminal racket is on trial" (emphasis added).
Not only does Mr. Dewey's reference to efficiency underscore the centrality of organized crime prosecutions to the enactment of section 279, it also explicitly highlights that the efficiency arose from the joint prosecution of "interrelated" offenses-not from trying unrelated offenses together.
The other reference to efficiency in the legislative history is a Bar Association letter recommending passage of the bill. Like Mr. Dewey, the Bar Association argued that prosecuting "similar or connected offenses" together would increase judicial efficiency-there was no attempt to justify joining completely unrelated crimes (Letter from N.Y. St. Bar Ass'n, Bill Jacket, L. 1936, ch. 328, at 14). The Bar Association's letter explained that the existing rules were "defective" because "in any complicated situation involving the commission of more than one crime, although there be a common intent or motive, the propriety of joining the various crimes thus committed is a matter of some doubt" (id. at 13[emphasis added]). Both Mr. Dewey's memorandum and the Bar Association's letter describe situations that fit within a Molineux exception. There is no reason to think the legislature meant to do more than that.
Nor does the legislative history's reference to the federal rules support justifying joinder of offenses on the basis of judicial economy. Section 279 of the Code of Criminal Procedure was, in part, modelled on the federal rules for joinder. But in 1936, the federal rules included no "good cause" provision. At that time, federal law permitted joinder of offenses "[w]hen there are several charges against any person for the same act or transaction, or for two or more acts or transactions connected together, or for two or more acts or transactions of the same class of crimes or offenses, which may be properly joined" (18 USC 557 [1934] [emphasis added]) . The federal statute did not provide for severance upon a showing of good cause.
Today, Rule 8 (a) of the Federal Rules of Criminal Procedure allows joinder "if the offenses charged-whether felonies or misdemeanors or both-are of the same or similar character, or are based on the same act or transaction, or are connected with or constitute parts of a common scheme or plan" (Fed. Rules Crim. Pro., rule 8 [a]).
Contemporaneous federal caselaw supports the argument that federal law did not permit joinder when it would be inconsistent with fundamental rules protecting the rights of defendants. U.S. Supreme Court jurisprudence required trial courts to "determine whether, in a given case, a joinder... is consistent with the settled principles of criminal law," (Pointer v United States, 151 U.S. 396, 400 [1894]), including the "fundamental... principle" that "the court must not permit the defendant to be embarrassed in his defense by a multiplicity of charges embraced in one indictment and to be tried by one jury" (id. at 403). In Pointer, "[t]here was such close connection between the two killings in respect of time, place, and occasion that it was difficult, if not impossible, to separate the proof of one charge from the proof of the other" (id. at 404). The Court held that in light of the connection between the two jointly tried murders it was "clear that the accused was not confounded in his defense... and that his substantial rights were not prejudiced" by consolidation (id.). When the Federal Rules of Criminal Procedure came into effect in 1944, they included a provision granting the trial court authority to sever "[i]f the joinder of offenses... appears to prejudice a defendant or the government" (Fed. Rules Crim. Pro., rule 14 [a]). The 1944 Advisory Committee on Rules of Criminal Procedure cited Pointer to explain rule 14 (Notes of Advisory Committee on Rules [1944], (Fed. Rules Crim. Pro., rule 14).
Subsequent legislative history does not change that analysis. In 1970, when New York adopted the new Criminal Procedure Law recommended by the Bartlett Commission, section 279 became CPL 200.20. The Bartlett Commission explained that section 200.20 "substantially restate[d]" existing law (Bartlett Commission, Staff Notes on Proposed Criminal Procedure Law § 100.20, at 173 [1967]). With respect to the requirements for severance, a staff comment to the 1967 proposed draft explained:
"The proposed draft, like the existing Criminal Code, is not inattentive to the possibility of prejudice when offenses are joined or consolidated solely because they are 'of the same or a similar character' (§ 100.20 [2(d)]). It expressly provides that in such instance (for example, two separate robberies), the trial court, in the interest of justice and for good cause shown, may in its discretion order [severance]" (id.).
Although section 200.20 underwent some revisions between 1967 and 1970, the good cause provision did not change.
The 1967 version of CPL 200.20 (2) (c) permitted joinder of multiple "offenses of the same or a similar character," copying section 279. In 1968, that became "defined by the same or similar statutory provisions and consequently... of the same or a similar legal character" (Bartlett Commission, Proposed New York Criminal Procedure Law § 100.20, at 88 [1968]). In 1969, "of the same or a similar legal character" became "the same or similar in law" (Bartlett Commission, Proposed New York Criminal Procedure Law § 200.20, at 115 [1969])
The legislative history fully supports reading CPL 200.20 (3) to incorporate our longstanding prohibition on the admission of propensity evidence. It does not suggest that judicial economy can outweigh the risks of prejudice that come with the joinder of unrelated offenses for trial .
Judicial economy is not unimportant, but it does not allow us to shortchange the right to a fair trial. We have, for example, highlighted the value of efficiency in the context of joinder of defendants for trial. In People v Mahboubian, we were asked to determine whether a joint trial of two defendants accused of jointly orchestrating a "staged theft of Persian antiquities" was proper (74 N.Y.2d 174, 180 [1989]). In that context, we held that "a strong public policy favors joinder, because it expedites the judicial process, reduces court congestion, and avoids the necessity of recalling witnesses" (id. at 183). Our holding in Mahboubian is not relevant here. First, the statutory provisions providing for joinder of defendants were enacted at a different time, through different legislation. Unlike the provisions governing joinder of offenses, which were created in 1936, the provisions governing joinder of defendants were created by the Bartlett Commission in the late 1960s. A staff comment to the Bartlett Commission's 1967 proposed Criminal Procedure Law explained that joinder of defendants was "not covered by the Criminal Code" (Bartlett Commission, Staff Notes on Proposed Criminal Procedure Law § 100.40, at 174 [1967]). The legislative history of the offense joinder provisions, as discussed supra, does not support an emphasis on judicial economy. Second, the joinder of multiple defendants presents different risks from the joinder of multiple offenses against a single defendant. While joining multiple defendants together for trial may be highly prejudicial in some circumstances, that is not due to the concerns underlying Molineux, but rather because the defenses of the defendants may be antithetical to each other, such as where one defendant's attorney attempts to impeach the testimony of another defendant (People v Cardwell, 78 N.Y.2d 996, 998 [1991]).
III
Reading the Molineux rule into CPL 200.20 (3) is consistent with the statutory context, and in particular with CPL 200.20 (2) (b). Section (2) (b) permits joinder where "proof of the first offense would be material and admissible as evidence in chief upon a trial of the second" or vice versa. On first glance, incorporating Molineux into CPL 200.20 (3) might seem inconsistent with the existence of section (2) (b). But CPL 200.20 (2) (b) reaches offenses that would not be joinable under (2) (c). Imagine, for example, charges for tax fraud and murder, where the tax fraud occurred years before the murder, but the prosecution argued that the murder victim was killed for threatening to reveal the fraud. Those offenses would not be joinable as arising from the same transaction, or as involving the same legal provisions. But they would be joinable under section (2) (b) if evidence of the fraud were admissible to prove motive for the murder. Interpreting good cause under 200.20 (3) to incorporate the Molineux rule would not, therefore, render 200.20 (2) (b) redundant.
Also, CPL 200.20 (2) (b) was enacted after sections 200.20 (2) (c) and (3). When section 279 was created in 1936, it permitted joinder where offenses (1) arose from "the same act or transaction," (2) arose from "two or more acts or transactions connected together or constituting parts of a common scheme or plan," or (3) constituted "crimes of the same or a similar character" (L 1936, ch. 328). Section 200.20 (2) (b) was added to the statute during the Bartlett Commission's drafting process, sometime between 1967 (when the draft bill did not include a [2] [b] equivalent) and 1970, when the new Criminal Procedure Law was enacted . The addition of section (2) (b) in the late 1960s does not tell us anything about the 1936 legislature's intent in creating a good cause exception to joinder for offenses of the same kind.
The Bartlett Commission also modified CPL 200.20 (2) (c) by adding the phrase "even though not joinable pursuant to paragraph (b)." This change does not alter the meaning of the good cause provision, which the Commission imported without modification.
IV
Here, the proper remedy is remittal for two separate trials. In denying Mr. Mero's request for severance, Supreme Court held that no good cause existed for severance because the "jury will readily be able to consider the proof separately as it relates to each charge" given that "the evidence in each case is markedly distinct." That is the wrong standard.
It was an abuse of discretion for Supreme Court to refuse to sever because the only reason to introduce evidence of one murder at a trial for the other would be to suggest Mr. Mero had a propensity to kill. At Supreme Court, the People conceded that they did not have "a Molineux purpose" for introducing evidence of one murder to show Mr. Mero's guilt in the other. As the court held, the record supports that concession. That means that in determining Mr. Mero's guilt in each killing, the only thing a juror could infer from evidence of the other was propensity. The only possible reason to permit joinder under these circumstances is efficiency. But efficiency cannot justify depriving Mr. Mero of his right to a fair trial. It was therefore an abuse of discretion for Supreme Court to deny severance.
Requiring that a Molineux exception be met for the joinder of two distinct crimes is also the best way to make practical sense of the statutory scheme. It would be easier for a jury to keep distinct proof of two unrelated crimes that did not involve the same penal law violation-for example, an automobile theft committed a year after an unrelated assault charge-than to keep separate the proof of two unrelated murder charges, and the risk of conviction by propensity is greater in the latter than the former. The legislature's concern for the jury's ability to keep the proof in each offense separate-evidenced in CPL 200.20 (3) (a)-thus weighs against reading CPL 200.20 (2) (c) to permit the joinder of completely unconnected crimes.
We must face the reality that when faced with a defendant charged with two crimes-especially two crimes that are the same in law-"the natural and inevitable tendency of the tribunal... is to give excessive weight to the vicious record of crime thus exhibited, and either to allow it to bear too strongly on the present charge, or to take the proof of it as justifying a condemnation irrespective of guilt of the present charge" (People v Telfair, 41 N.Y.3d 107, 114 [2023], quoting People v Zackowitz, 254 NY 192, 198 [1930] [Cardozo, J.]). We should not read the joinder statutes quietly to circumvent our time-honored protections against such prejudice. The legislature did not intend to do so. I would read the joinder and severance provisions enacted in 1936 as the legislature intended: to allow joinder where offenses are interrelated, and to require severance when joinder would serve no purpose but prejudice, or where the risk of prejudice outweighs any probative purpose.
RIVERA, J. (dissenting)
Defendant was convicted of two unrelated murders, which occurred years apart, by different methods. There were no eyewitnesses to either death and no forensic evidence linked defendant to the crimes. The bodies were found in different locations. The common denominator was that defendant knew both victims: one was his roommate and the other, a female acquaintance. Defendant maintained his innocence and contended that one of the deaths was caused by an accidental fire. Nevertheless, the murders and the related evidence-tampering charges were joined in a single indictment. Eventually, they were tried before the same jury, over defendant's request for separate trials to avoid inevitable prejudice to his defense. Joinder of these counts rested solely on the ground that they "are defined by the same or similar statutory provisions" (CPL 200.20 [2] [c]).
Failing to sever the trials carried significant risks that the jury would determine guilt based on the appearance of defendant's criminal disposition and that the evidence of guilt on one of the murder counts would spill over and bolster the evidence of the other. The court's instructions and counsel's advocacy could not overcome this prejudice to the defense. Therefore, the trial court abused its discretion by denying severance. On that ground, I would reverse defendant's convictions and order separate trials on each of the respective murder and related counts.
I.
Section 200.20 (2) (c) of the Criminal Procedure Law permits joinder of wholly unrelated offenses so long as "such offenses are defined by the same or similar statutory provisions and consequently are the same or similar in law." But CPL 200.20 (3) authorizes a court, upon request by the defendant or the prosecution, to order that such dissimilar counts be tried separately "in the interest of justice and for good cause shown." This subparagraph also sets forth nonexhaustive examples of "good cause," the common thread being the risk of prejudice to the defense. For example, good cause exists in "situations where there is... (a) Substantially more proof on one or more such joinable offenses than on others and there is a substantial likelihood that the jury would be unable to consider separately the proof as it relates to each offense" or "a defendant has both important testimony to give concerning one count and a genuine need to refrain from testifying the other, which satisfies the court that the risk of prejudice is substantial" (CPL 200.20 [3] [a], [b]).
II.
Our prohibition on the admission of evidence of crimes and bad acts unrelated to the offense charged guards against conviction based on a defendant's perceived criminal disposition. As relevant to the issue on this appeal, it serves as a judicial safeguard against prejudicial joinder of offenses similar at law but factually unconnected. The concerns underlying that rule, along with the history and purpose of CPL 200.20, confirms that the murder-related counts here should have been tried separately.
A.
One of our bedrock criminal law principles is that a defendant may not be convicted based on propensity evidence (see People v Molineux, 168 NY 264 [1901]). A defendant's conviction for an offense must rest solely on the evidence admitted at trial and the factual findings and inferences drawn therefrom. Evidence of other crimes or bad acts may not be admitted to bolster evidence of another crime or to establish a defendant's likelihood of having committed the crime charged because they have a bad character or appear predisposed to criminality (see id. at 293 [internal quotation marks omitted]). Absent a relevant connection between the offenses, reliance on evidence of a criminal disposition is inherently prejudicial and if admitted, requires reversal of the conviction." 'It would be easier to believe a person guilty of one crime if it was known that [they] had committed another of a similar character, or, indeed of any character; but the injustice of such a rule in courts of justice is apparent. It would lead to convictions, upon the particular charge made, by proof of other acts in no way connected with it, and to uniting evidence of several offences to produce conviction for a single one' "(id. at 292, quoting Coleman v People, 55 NY 81, 90 [1873]). "Thus, '[w]hen we limit Molineux or other propensity evidence, we do so for policy reasons, due to fear of the jury's 'human tendency' to more readily 'believe in the guilty of an accused person when it is known or suspected that [they have] previously committed a similar crime'" (People v Weinstein, 2024 NY Slip Op 02222, at *7 [Ct App Apr. 25, 2024], quoting People v Brewer, 28 N.Y.3d 271, 276 [2016], quoting People v Ventimiglia, 52 N.Y.2d 350, 359 [1981], and citing Molineux, 168 NY at 313).
The exceptions to the bar on propensity evidence share the common characteristic of relevance to the offense prosecuted. Thus, we have identified evidence of motive, intent, impossibility or improbability of mistake or accident, a common plan or scheme, and identity as relevant bases for admission of evidence of other criminal conduct (see Molineux, 168 NY at 293). Even when evidence of a defendant's criminal propensity is relevant to some other, proper issue, the evidence may only be admitted if "its probative value exceeds the potential for prejudice resulting to the defendant" (see People v Alvino, 71 N.Y.2d 233, 242 [1987]).
The Legislature was well aware of this well-established principle and the limited exceptions to the bar against propensity evidence when it introduced the provisions at issue on this appeal (see Matter of Estate of Youngjohn v Berry Plastics Corp., 36 N.Y.3d 595, 606 [2021] ["(T)he legislative history of a particular enactment must be reviewed in light of the existing decisional law which the (l)egislature is presumed to be familiar with and to the extent it left it unchanged, that it accepted"]).
B.
Until 1936, the Code of Criminal Procedure required that an "indictment must charge but one crime and in one form except that the crime may be charged in separate counts to have been committed by different means; and when the acts complained of may constitute different crimes, such crimes may be charged in separate counts" (People v Dimick, 107 NY 13, 31 [1887], citing Code Cr Proc §§ 278, 279). Then, the Legislature enacted a predecessor of today's CPL 200.20. That enactment liberalized the law of joinder, including by providing that "the court may order" consolidated trials of "two or more acts or transactions constituting crimes of the same or a similar character" (Code Cr Proc § 279 [eff 1936]). In such cases, however, "the court, in the interest of justice and for good cause shown, may, in its discretion, order that the different charges... be tried separately" (id.).
Crucially, this liberalization did not displace established principles of criminal law. The Governor and a key proponent supported the 1936 amendment as an "anti-racketeering bill" (Letter from Thomas E. Dewey to Charles Poletti, Feb 27, 1936, Bill Jacket, L 1936, ch 328), one that would permit prosecution of "the separate but inter-related offenses of the single criminal organization" (Governor's Mem at 2, Bill Jacket, L 1936, ch 328 [quoting Dewey]). Joinder would also provide for the efficient resolution of criminal matters (see id.). Consistent with these purposes, the bill's supporters emphasized that it was "of course, not a new or radical departure in the administration of criminal justice" (id.). Indeed, the bill tracked with minor changes an analogous federal statute, the predecessor to today's Federal Rules of Criminal Procedure 8 and 14. Thus, as the Governor noted, New York's new Section 279 would "bring[ ] to the state courts a large body of considered judicial decisions from all of the Federal Courts, where the statute ha[d] been in just and successful administration" for nearly a century (id. at 2-3).
That body of federal case law balanced the administrative benefits of joinder with the concern for undue prejudice against defendants. The leading case at the time was Pointer v United States (151 U.S. 396 [1894]; see People ex rel. Pincus v Adams, 274 NY 447, 454 [1937] [discussing Pointer ]; De Luca v United States, 299 F 741, 743-745 [2d Cir 1924] [same]). There, the United States Supreme Court indicated that if a joint trial of multiple charges risked prejudice to "the substantial rights of the accused," "the court, according to the established principles of criminal law, can compel" severance (Pointer, 151 U.S. at 403). Whether a court abused its discretion by denying severance depended "on the special circumstances" of each case (id. at 402). Pointer itself concerned two charges of murder. "The indictment showed that the two murders were committed on the same day, in the same county and district, and with the same kind of instrument. These facts alone justified" the trial court's denial of severance (id. at 403). And at trial, "the wisdom of the course pursued by the court became manifest, for it appeared that the two murders were committed at the same place, on the same occasion, and under such circumstances, that the proof in respect to one necessarily threw light upon the other" (id. at 404). In short, under the principles adopted by the 1936 amendment to the Code of Criminal Procedure, denial of severance was most appropriate where charges were factually related. Where charges were utterly distinct except in law, a trial court risked abusing its discretion by permitting a joint trial.
Subsequent developments reinforce that the Legislature was attentive to this risk. In 1967, the Temporary Commission on Revision of the Penal Law and Criminal Code, known as the Bartlett Commission, proposed to replace the Code of Criminal Procedure. Section 279, however, was largely untouched. The Commission sought "solely" to "substantially restate[ ]" and "give added clarity" to the 1936 provision (State of New York Temporary Commission on Revision of the Penal Law and Criminal Code, Proposed New York Criminal Procedure Law at 173 [1967]). The Commission explained that "the proposed draft, like the existing Criminal Code, is not inattentive to the possibility of prejudice when offenses are joined or consolidated because they are 'of the same or a similar character,'" as in an indictment of "two separate robberies" (id.). In such cases, the Commission noted that severance might well be appropriate or, perhaps, required. The Commission's draft became what is now CPL 200.20.
In 1984, the Legislature amended § 200.20 to encourage severance in appropriate cases. The 1984 amendments, which remain effective today, identified situations in which there exists "good cause" to sever counts of an indictment that are similar only in law. One of these is "when there is... [s]ubstantially more proof on one or more such joinable offenses than on others and there is a substantial likelihood that the jury would be unable to consider separately the proof as it relates to each offense" (CPL 200.20 [3] & [3] [a]). According to both the State Executive Chamber and the New York City Mayor, this clause codified the rule of People v Forest (50 A.D.2d 260 [1st Dept 1975]; see Mem of Executive Chamber at 1-2, Bill Jacket L 1984, ch 672; Mem of Mayor of City of New York at 2, Bill Jacket L 1984, ch 672). In Forest, the defendant had been convicted following a joint trial of three counts of robbery (see Forest, 50 A.D.2d at 261). For two of those counts, eyewitnesses identified the defendant as the perpetrator; for the third, the victim identified the defendant only from a photo array, and only after first selecting a different person (see id.). Even though the robberies all occurred within one month, and even though all involved purse snatchings in the Bronx, the Court held that the trial court abused its discretion by denying severance (see id. at 261-262). Factual similarities among the counts, although decisive in Pointer, were not enough to justify joinder in Forest.
Thus, in the near-century since it first liberalized the joinder rules, the Legislature has never retreated from the established legal principles protecting defendants against undue prejudice. Indeed, it has only reaffirmed them. While offenses similar in nature may be joined, the legislature fully understood that wholly unrelated offenses - those lacking any connection whatsoever other than the prosecution's allegation that the defendant committed the crimes so charged-would be inadmissible under general criminal law principles because of the inherent prejudice to a defendant from propensity evidence.
Evidence concerning one criminal transaction is often inadmissible in a trial of another criminal transaction, even where there is some alleged factual connection between them (see e.g. People v Weinstein, 2024 NY Slip Op 02222, at *8 [Ct App Apr. 25, 2024]). In such cases, joinder would be impermissible under CPL 200.20 (2) (b), but permissible under CPL 200.20 (2) (c) (see e.g. People v Adames, 42 A.D.3d 328, 329 [1st Dept 2007]). In concluding as much, I depart from the Chief Judge (see Wilson, Ch J, concurring op at 2). Requiring separate trials for charges with no factual connection vindicates the Legislature's choice to expand joinder beyond Molineux 's strict terms, while also respecting its decision to preserve the longstanding principles embodied in that case. If, as the majority suggests, "joinder of factually unrelated offenses does not necessarily present a greater risk of prejudice than joinder of factually related offenses" (majority op at 9), that is reason to grant severance in some cases involving the latter; it is no reason to abandon this Court's role in supervising trial courts' exercises of discretion.
C.
Scholars have long warned of the grave risks attending joinder in cases like the one presented on this appeal. In 1980, Judge Friendly noted that CPL 200.20 (2) (c)'s federal analog-the "same or similar character" provision in Federal Rule of Criminal Procedure 8(a)-had "not been met with the favor of commentators" (United States v Werner, 620 F.2d 922, 927 [2d Cir 1980]). The criticism continues more than forty years later. For example, the American Bar Association's Standard for Criminal Justice states: "Whenever two or more unrelated offenses"-that is, offenses not based upon the same conduct, criminal episode, or plan-"have been joined for trial, the prosecuting attorney or the defendant shall have a right to a severance of the offenses" (ABA Criminal Justice Standards: Joinder & Severance standard 13-3.1 [emphasis added], available at https://www.americanbar.org/groups/criminal_justice/resources/standards/joinder-severance/). The leading treatise on federal practice and procedure likewise questions "whether joinder of this type should ever be permitted" (1A Charles Alan Wright, Arthur R. Miller, & Andrew D. Leipold, Federal Practice & Procedure § 223 [5th ed June 2024 update]). And that text's counterpart on criminal procedure suggests that "[a]bsent reform" along the ABA's proposed lines, "some change in the law (or at least in judicial attitudes) is badly needed" (5 Wayne R. LaFave, Jerold H. Israel, Nancy J. King, & Orin S. Kerr, Criminal Procedure § 17.1 [f] [4th ed Nov 2024 update]).
These concerns are not speculative. In 1989, there already existed empirical literature concerning joinder of offenses, based mostly upon experimental studies (James Farrin, Note, Rethinking Criminal Joinder: An Analysis of the Empirical Research and Its Implications for Justice, 52 L & Contemp Probs, 325, 326 [1989]). That "empirical data unequivocally show[ed] that the probability of a defendant being convicted significantly increases if offenses are joined rather than tried separately" (id. at 332). The effect was stronger still "when the cases joined were similar as opposed to dissimilar" (id.). Moreover, limiting instructions to juries were "insufficient to counter the demonstrated prejudicial effects of joinder" (id. at 336). More recently, an examination of all federal criminal defendants who stood trial over a five-year period found that "a defendant's chances of conviction increase by more than 10% if [they] stand[ ] trial on more than one count," regardless of whether the defendant stands trial alone or with other defendants (Andrew D. Leipold & Hossein A. Abbasi, The Impact of Joinder and Severance on Federal Criminal Cases: An Empirical Study, 59 Vand L Rev 347, 383-384 [2006]). This enhanced risk of conviction from joinder has profound significance for defendants standing trial-and even for those bargaining with prosecutors before trial (see Andrew Manuel Crespo, The Hidden Law of Plea Bargaining, 118 Colum L Rev 1303, 1316-1323 [2018]).
Our role as New York's high court demands that we confront these concerns. For Judge Friendly, the answer was clear: "The Rule is what it is; if commentators do not like it, the road for them to seek amendment is open" (Werner, 620 F.2d at 928). Presumably, the majority views matters similarly. But there is one problem: our rule is not the federal rule. "In order to prevail," the defendant in Werner had to "show not simply some prejudice but substantial prejudice" from the failure to sever (id.). Federal courts still adhere to that requirement (see e.g. United States v Lingala, 91 F4th 685, 693 [3d Cir 2024]; United States v Page, 657 F.3d 126, 129 [2d Cir 2011]). This Court, however, has never held that a defendant must establish actual prejudice. To the contrary, as the majority acknowledges, we review denial of a defendant's severance motion only for abuse of discretion, considering the risk of prejudice as it existed at the time of the motion (see majority op at 5; People v Ford, 11 N.Y.3d 875, 879 [2008]; People v Shapiro, 50 N.Y.2d 747, 756 [1980]). Consequently, we may-indeed, we must-ensure that trial courts exercise their discretion with due regard for the possibility that a joint trial might not be a fair trial.
III.
Turning to the appeal before us, defendant requested severance on the ground that the evidence in the cases was underwhelming, and, thus, "the simple 'What are the odds?' factor that defendant couldn't possibly be uninvolved in both cases, create[d] immediate prejudice." He also contended that the prosecutor's evidence would be sharply different in each case, given defendant's claims that the roommate died in an accidental fire and that the murder of the female acquaintance was a case of misidentification.
Trial courts have "reasonable latitude" to grant or deny severance of trials for multiple offenses (People v Lane, 56 N.Y.2d 1, 8 [1982]). But that latitude is not unlimited. In this case, the trial court erred by denying defendant's request. Presenting evidence of both murders to the same jury presented a substantial risk that the jury would convict based on what it might conclude was his criminal disposition and an equally high risk that the jury would aggregate the evidence, so that proof of one murder would bolster evidence of the other. The result was that the jury might well convict on insufficient evidence of defendant's guilt on either or both murder counts. Such a prospect is anathema to "our system of justice" (Weinstein, 2024 NY Slip Op 02222, at *1).
Several factors, taken together, rendered denial of severance untenable. First, the prosecution conceded that the case against defendant for the murder of the female acquaintance was stronger than that for the murder of the roommate. Evidence of the former murder would be confirmatory of guilt of the latter. The danger of "conviction by reason of [the charges'] cumulative effect rather than on the strength of the specific evidence regarding each crime" was precisely the sort of risk CPL 200.20 (3) was enacted to avoid (Forest, 50 A.D.2d at 262; see Governor's Mem at 1-2, Bill Jacket, L 1984, ch 672).
The majority's discussion of this factor is puzzling. If the majority means that neither conviction was against the weight of the evidence, that is a judgment that this Court lacks authority to render in this case (see People v Danielson, 9 N.Y.3d 342, 349 [2007]). More fundamentally, the majority appears to impose a prejudice requirement where none exists. A defendant's request for severance "must be judged in the context of the... circumstances against which it is made" (Shapiro, 50 N.Y.2d at 757). This Court's retrospective analysis of the trial evidence has no place in assessing whether the trial court abused its discretion by denying a pretrial severance motion. The majority's additional observation that CPL 200.20(3)(a) "provides for severance if the jury cannot segregate proof" and the court made no such finding (see majority op at 9), further illuminates the error below. The court simply ignored the fundamental precept of the Molineux rule: propensity evidence cannot be segregated from evidence of the charged offense because of the jurors' natural and human tendency to bolster or disregard the weakness of the prosecution's evidence and find guilt based upon the defendant's criminal past.
Second, the cases were completely different factually, except for one common feature: defendant knew both victims. As defendant argued in his initial severance motion, the jury could well conclude that it was not coincidence that defendant was the only connection between the two murdered victims. In this respect, this case differs sharply from those where this Court has found no abuse of discretion in denying severance. For example, in People v Ford -relied upon at oral argument by the prosecution-there were "two successive robberies in May 2004 in elevators of residential buildings on Park Avenue in Manhattan" (11 N.Y.3d at 876).
Third, the crimes are heinous, committed by someone without regard for human life. Indeed, if the jury credited the prosecution's evidence, it could have concluded that the roommate's murder was premeditated. Joinder of offenses is always prejudicial (see Werner, 620 F.2d at 929), but the prejudice is greater when the crimes charged provoke the revulsion of the jury (see Shapiro, 50 N.Y.2d at 756). Given the understandable human tendency to feel animosity towards the perpetrator of two brutal murders, there was a high risk that the jurors would convict on all counts not because they would find guilt beyond a reasonable doubt based on the evidence specific to each case, but because they would view the cases cumulatively. In other words, viewing the evidence in totality, the jury would conclude that, defendant murdered the victims.
Fourth, and finally, the issues were complex. Unlike many cases in which there was no abuse of discretion in denying severance, this was not a trial in which "the evidence as to each [offense] was short and simple" (United States v Lotsch, 102 F.2d 35, 36 [2d Cir 1939] [L. Hand, J.]). To the contrary, dozens of witnesses testified over a lengthy trial about matters ranging from municipal apparel to cell phone records to pulmonary edema. If ever jurors were at risk of impermissibly aggregating evidence, it was in this trial.
In sum, much as in Shapiro, "the superficial closeness of the indictments here, resulting largely from a common focus on the same kind of" violent conduct, "was likely to eclipse the very fundamental difference between them" (Shapiro, 50 N.Y.2d at 756). Thus, "there is a substantial likelihood that the jury would be unable to consider separately the proof as it relates to each offense" (CPL 200.20 [3] [a]).
For the same reason, and contrary to the majority's view, the court's instructions could not eliminate the risk that the jury would convict defendant of both crimes based on the combined strength of the evidence (cf. majority op at 6-7). We assume that the jury will follow the court's instructions, and here the court told the jury that "[p]roof on one case must not be considered as proof in the other case. It is your obligation to evaluate the evidence as it applies or fails to apply to each case separately." However, there are cases where instructions are not enough. Even where instructions are "precise and impeccable," they cannot cure an "overabundance" of improperly prejudicial evidence (People v Stanard, 32 N.Y.2d 143, 148 [1973]). This is just such a case.
Nor, as the majority suggests, are the jury notes relevant (cf. majority op at 8 n 4). The propriety of the court's decision on the motion to sever is measured as of the time of the motion, and not based on hindsight. The question is one of potential risk, not of whether an erroneous ruling was vindicated at trial. In any case, the jury notes here did not establish that the jurors cabined the evidence to prevent tainting their deliberations on the counts related to one murder with consideration of evidence related to the other. The notes merely established that the jury referenced the evidence by victim's name. The notes provide no insight into how they considered that evidence and whether they improperly synthesized it to reach their verdict.
The majority fails to identify any factual similarity between the Appellate Division cases it cites and the underlying facts of defendant's prosecution (see majority op at 8). The majority's silence emphasizes the outlier status of the joinder of the counts here and why the court abused its discretion by denying severance. In People v Bonner, the evidence established defendant's common "motive," "modus operandi," and "common scheme of using physical abuse to instill fear and obedience in the prostitutes who worked for him" (94 A.D.3d 1500, 1501 [4th Dept 2012]). In People v Perez, the counts were joinable under CPL 200.20 (2) (b), "on the basis of overlapping evidence..., especially with regard to defendant's confessions and their surrounding circumstances" (47 A.D.3d 409, 411 [1st Dept 2008]; see also Perez v Conway, 09 CIV. 5173, 2011 WL 1044607, at *4 [SDNY Mar. 18, 2011] ["While (defendant) initially told police that two drug-dealers killed Drakeford and that he had only helped disposed of her body, he admitted that his tale was a lie and confessed that he had killed both Pollard and Drakeford in later videotaped interviews"]). In People v Adames (42 A.D.3d 328 [1st Dept 2007]) the prosecution sought to link both murder counts to defendant's involvement in the narcotics trade (see generally Brief for Appellant in Adames, 42 A.D.3d, *46-64; Brief for Respondent in Adames, 42 A.D.3d, *40-53). And in People v Oliveira, "the evidence of guilt as to both incidents was overwhelming" and defendant's "proposed testimony would have opened the door to evidence of the crimes charged in the second incident" (2 A.D.3d 122, 123 [1st Dept 2003]; see also Oliveira v Phillips, 05 CIV. 564 (SAS), 2007 WL 2890211 [SDNY Sept. 28, 2007]). These cases illustrate judicial recognition of the need for some relevant connection between the charges.
Finally, considerations of judicial economy cannot support affirmance. While trial courts should weigh considerations of efficiency against the defendant's interest, "compromise of a defendant's fundamental right to a fair trial free of undue prejudice as the quid pro quo for the mere expeditious disposition of criminal cases" is intolerable (Lane, 56 N.Y.2d at 8). Thus, where the risk of undue prejudice becomes sufficiently great, no appeal to judicial economy can justify denying severance. Here, the risk was significant and thus intolerable.
IV.
The Legislature never intended for joinder in cases like this one. To have done so would indeed have been the "radical departure" that liberalized joinder's proponents explicitly disclaimed (Governor's Mem at 2, Bill Jacket, L 1936, ch 328). However, if the judiciary exercises its discretion as broadly as the majority now permits - in contravention of our criminal law principles and notwithstanding the prejudicial effects on defendants - then it is time for the Legislature to step in and take corrective action.
Order affirmed. Opinion by Judge Halligan. Judges Garcia, Singas, Cannataro and Troutman concur. Chief Judge Wilson dissents in an opinion. Judge Rivera dissents in a separate opinion.
Comments