Argued September 16, 1975
Decided December 22, 1975
Appeal from the Appellate Division of the Supreme Court in the Second Judicial Department, GEORGE J. BALBACH, J.
Herbert A. Lyon, William M. Erlbaum and Charles Wender for appellant.
Nicholas Ferraro, District Attorney (Thomas A. Duffy, Jr., and Barry Alan Schwartz of counsel), for respondent.
Chief Judge BREITEL Defendant has twice been convicted, after jury trials, of manslaughter in the first degree for killing her daughter. She was sentenced from 5 to 20 years' imprisonment.
On this appeal, defendant contends that she was wrongfully denied a hearing by the Supreme Court on her motion to vacate judgment. The Appellate Division affirmed.
The critical issue is whether the denial, without a hearing, of defendant's motion to vacate judgment upon the ground of newly discovered evidence may be reviewed by this court. The dissenter's challenge that there has been a denial of justice makes imperative consideration and elaboration of the issue and the questions of law and fact which resolve them conclusively.
There should be an affirmance. The power to review a discretionary order denying a motion to vacate judgment upon the ground of newly discovered evidence ceases at the Appellate Division. This court, limited in a noncapital criminal action to reviewing questions of law only, has no power to review the discretionary denial of a motion to vacate judgment upon the ground of newly discovered evidence. Not to be confused is this court's power to set aside a conviction on the ground that it is not supported by proof sufficient to establish guilt beyond a reasonable doubt as a matter of law (see People v Santos, 38 N.Y.2d 173, and cases cited).
Even if the denial of such a motion were reviewable, the same result would obtain. The "newly discovered evidence" consisted only of an affidavit of a "witness" submitted some seven years after the events described in it had occurred. The affidavit was complete on its face, that is, it presented fully, and in the best possible light, all the evidence the "witness" could possibly offer on defendant's behalf. Assuming affiant's sincerity, his averments were in part conclusory and were highly unreliable as to accuracy of recollection. The averments did not disclose a probability, as opposed to a mere possibility, that the jury would have rejected the overwhelming evidence of defendant's guilt and returned a verdict more favorable to defendant. Thus, the court below properly denied the motion without a hearing.
In 1968, defendant was convicted, after jury trial, of manslaughter in the first degree for killing her daughter. That conviction was reversed on appeal and a new trial ordered ( 33 A.D.2d 793, affd 26 N.Y.2d 319).
Thereafter, in 1971, defendant was convicted, after jury trial, of murder in the first degree for killing her son, five years of age, and, for the second time, of manslaughter in the first degree for killing her daughter, four years of age. The Appellate Division reversed the murder conviction, on the law and the facts, and dismissed the applicable murder count in the indictment. That court also reversed the manslaughter conviction, on the law only, and ordered a new trial ( 41 A.D.2d 933). This court affirmed with respect to the dismissal of the murder charge, but reversed the order granting a new trial and remitted the appeal to the Appellate Division for a determination of the facts ( 36 N.Y.2d 230, COOKE and FUCHSBERG, JJ., dissenting in an opn by COOKE, J.).
The Appellate Division, on remittal, affirmed the judgment of conviction. It also affirmed an order denying defendant's double-branched motion to vacate judgment upon (1) the ground of newly discovered evidence and (2) asserted improper conduct by the prosecution in withholding from her information allegedly helpful to her defense ( 48 A.D.2d 663). Leave to appeal to this court was granted by a Justice of the Appellate Division. On the present appeal, defendant has limited her argument to whether she was entitled to a hearing in connection with her motion to vacate judgment.
The evidence upon the second trial compels the conclusion that defendant killed her daughter. At the second trial, Joseph Rorech, one of defendant's paramours, testified that she had confessed to him that she had killed her daughter. Rorech also recounted other conversations with defendant during which she indicated her guilt.
On July 14, 1965, defendant's daughter's body had been found in a vacant lot near defendant's residence. An autopsy revealed that the little girl had been asphyxiated by human agency. From a scientific analysis of the contents of her stomach, it was established that the girl had died from less than one half hour to no more than two hours after eating. Defendant admitted to police officers investigating the disappearance of her children that, on the evening of July 13, she had fed them a meal which began at 7:30 P.M. and ended at 8:15 P.M. Thus, it would appear that the daughter was killed between approximately 8:30 and 9:00 or 9:30 P.M. Defendant's own statements establish that she, separated from her husband, had exclusive custody of the children during those hours.
As Judge JONES recounted, on behalf of a majority of the court, in People v Crimmins ( 36 N.Y.2d 230, 242, supra), "In addition to other compelling circumstantial evidence, there was eyewitness testimony * * * that on the night before the daughter's body was found, defendant, carrying what was described as a `bundle' and accompanied by an unidentified man, was seen leading her son from the Crimmins home; that as the man threw the `bundle' into a parked car defendant cried out, `Please don't do this to her', to which the man responded, `Does she know the difference now? * * * Now you're sorry." This eyewitness was Mrs. Sophie Earomirski, one of defendant's neighbors.
Notably, Judge JONES in referring to "compelling" evidence, addressed himself to the circumstantial evidence in the case rather than to the testimonial evidence of Mrs. Earomirski, a witness whose credibility was severely attacked by the defense.
Moreover, still other evidence pointed to defendant's guilt. For example, defendant's trial theory was that her children had been abducted from the house by outside kidnappers. Since the children presumably were asleep in their locked bedroom at the time of the "abduction", if the defense theory were valid, the kidnappers must have entered the Crimmins residence and removed the children through the bedroom window. Yet, police investigation the following day established that layers of dust on a chest of drawers in the bedroom abutting the window and on the window sill were undisturbed. As the court stated, "the prosecution conclusively exploded defendant's theory of an outside kidnapper" (36 N.Y.2d, at p 243).
Thus, the majority concluded (36 N.Y.2d, at pp 242, 243, supra): "there was overwhelming proof that this defendant was guilty of manslaughter in the death of her daughter * * * We read this record as leading only to a single, inexorable conclusion, as two juries have indeed found: defendant was criminally responsible for the death of her daughter."
It was not on this appeal but on the first appeal, after the first trial, to this court, that Mrs. Earomirski's testimony was described, in the opinion from which the dissenter quotes, as "essential" to the prosecution (26 N.Y.2d, at pp 322, 324).
Indeed, the proof of defendant's guilt was so overwhelming that the case has withstood an attack based upon trial errors committed by the prosecution. Those errors were held to be harmless: in the case of constitutional error, beyond a reasonable doubt; and in the case of nonconstitutional error, because "there is no significant probability in the light of the overwhelming proof that, had it not been for the errors which occurred, this jury would have acquitted the defendant or that a third jury might do so" (36 N.Y.2d, at p 243). Of course, the dissenters on that appeal did not and still do not agree with that evaluation (see dissenting opn by COOKE, J., at pp 243-250).
This statement of the harmless error rule with respect to nonconstitutional errors did not constitute a new rule. As the court stated, the "doctrine [of harmless error] has received expression in our court over the last 20 years" (36 N.Y.2d, at p 239). The rule restated on the second Crimmins appeal was an accommodation to what had developed in the area of harmless error over the years, from the time of the Code of Criminal Procedure to the CPL, and has survived the latter day imposition by the Supreme Court of a harmless "beyond a reasonable doubt" rule for Federally-declared constitutional error. Nonconstitutional error, as has been the statutory rule since 1881 and the decisional law in this State for over one hundred years, is harmless when, given the overwhelming proof of a defendant's guilt, there is no significant probability that the jury would have acquitted the defendant if the error had not been committed (see, e.g., Code Crim Pro, § 542 [L 1881, ch 504, § 542]; CPL 470.05, subd 1; Shorter v People, 2 N.Y. [Comstock] 193, 202; People v Corey, 148 N.Y. 476, 493-494; People v Marendi, 213 N.Y. 600, 619; People v Crimmins, 36 N.Y.2d 230, 239-240, and cases cited at p 242; see, also, Stokes v People, 53 N.Y. 164, 180).
It has been useful in the context of this case to examine the harmless error rule applicable to nonconstitutional error, for it is similar to the statutory standard for determining when a motion to vacate judgment upon the ground of newly discovered evidence should be granted (see CPL 440.10, subd 1, par [g]).
The "newly discovered evidence" upon which defendant's motion to vacate judgment is in part based is an affidavit by one who came forward with his story some seven years after the events he narrates. The would-be witness averred that, in the early morning of July 14, 1965, shortly after 1 A.M., he had picked up two small children, a boy and a girl, "hitchhiking" in the vicinity of 150th Street and "probably" 75th Avenue in Queens County. He learned from the boy "that he [the boy] knew exactly where `home' and his `daddy' was." He drove, under the boy's direction, for several blocks, and eventually let the children out at the southwest corner of 162nd Street and 71st Avenue. Supposedly a responsible individual, affiant was "quite sure" these small children "were safe and sound" at this spot where he left them in the wee hours of the morning. If they had been the Crimmins children they would have been four and five years of age, respectively.
Affiant did not learn the names of the children. Despite this, he stated unequivocally in the affidavit that these two children were the Crimmins', a conclusion admittedly reached only after reading about the killing in the newspapers a few days later. He did state in the affidavit that the boy "could very well have" identified his companion as "Missy" (defendant's daughter's nickname) instead of "my sister", as he had first thought. He does admit, however, that he arrived at this conclusion only after reading newspaper accounts of the first trial, three years after the incident. The affidavit concludes with an uncertain hypothesis of how the Crimmins children met their fate.
The trial court, as noted, denied, without a hearing, defendant's motion to vacate judgment upon the ground of newly discovered evidence based on the affidavit.
CPL 440.10 (subd 1, par [g]) provides that:
"At any time after the entry of a judgment, the court in which it was entered may, upon motion of the defendant, vacate such judgment upon the ground that: * * *
"New evidence has been discovered since the entry of a judgment based upon a verdict of guilty after trial, which could not have been produced by the defendant at the trial even with due diligence on his part and which is of such a character as to create a probability that had such evidence been received at the trial the verdict would have been more favorable to the defendant; provided that a motion based upon such ground must be made with due diligence after the discovery of such alleged new evidence".
CPL 440.10 (subd 1, par [g]) is the successor to subdivision 7 of section 465 of the former Code of Criminal Procedure, which provided for application for a new trial on similar, but somewhat more stringent, grounds (see 6 Zett, New York Criminal Practice, par 50.3, subd , par [g]). Unlike a post-conviction application for a new trial under the code, which had to be made within one year after judgment, no time limitation is prescribed for a motion to vacate judgment under CPL 440.10 (subd 1, par [g]) (compare former Code Crim Pro, § 466).
CPL 440.10, captioned "Motion to vacate judgment", was intended to encompass contentions which formerly could have been raised on an application for postconviction relief in the nature of coram nobis, a postconviction application for a new trial upon the ground of newly discovered evidence, and those which may be raised in State and Federal habeas corpus proceedings, as well as other contentions (see Denzer, Practice Commentary, McKinney's Cons Laws of NY, Book 11A, CPL 440.10, at p 183; Pitler, New York Criminal Practice Under the CPL, pp 830-831).
Historically, neither an order denying coram nobis relief nor an order denying a postconviction application for a new trial upon the ground of newly discovered evidence was appealable (see People v Gersewitz, 294 N.Y. 163, 166, app dsmd 326 U.S. 687; People v Trezza, 128 N.Y. 529, 533). This was so because the right to appeal in a noncapital criminal case is purely statutory, and no statute authorized such appeals (id.; compare a prejudgment motion for a new trial upon the ground of newly discovered evidence, which, as a part of the judgment roll, could be appealed under section 517 of the former Code of Criminal Procedure [see People v Priori, 163 N.Y. 99, 107-108; Cardozo, Jurisdiction of the Court of Appeals (2d ed), § 122]).
In 1947, however, section 517 of the former Code of Criminal Procedure was amended to authorize an appeal from an order denying a "motion to vacate a judgment" (L 1947, ch 706; see Matter of Bojinoff v People, 299 N.Y. 145, 149). The amendment was initially interpreted to apply to a denial of a postconviction application for a new trial upon the ground of newly discovered evidence (see People v Greenfield, 275 App. Div. 862, 863). In 1953, however, section 517 was again amended to make clear that an appeal was authorized only from an order denying an "application for a writ of error coram nobis" (L 1953, ch 602; see People v Palumbo, 282 App. Div. 1059). Thus, before the enactment of the CPL, a denial of a postjudgment application for a new trial upon the ground of newly discovered evidence remained nonappealable (see, e.g., People v Baxter, 40 A.D.2d 551; People v La Rocca, 37 A.D.2d 974; People v Hendry, 13 A.D.2d 793; Pitler, op. cit., at p 834).
With the passage of CPL 450.15 (subd 1) and CPL 450.90 (subd 1), an order denying a motion under CPL 440.10 (subd 1, par [g]) to vacate judgment upon the ground of newly discovered evidence was made appealable (see Pitler, op. cit., at p 834). While this would appear to resolve the issue of appealability, it does not settle the question of reviewability by this court. The Legislature is free to expand or contract by statute the classes of noncapital criminal cases which may be appealed to the Court of Appeals. The court's power to review may not, however, be expanded by statute alone, and the CPL must be read in context of constitutional limitations, whether the CPL is explicit or not (compare N Y Const, art VI, § 3, subd b with subd a).
Section 3 of article VI of the Constitution provides that "The jurisdiction of the court of appeals shall be limited to the review of questions of law except where the judgment is of death". Thus, CPL 470.35 (subd 1) provides that, upon an appeal to the Court of Appeals from an order of an intermediate appellate court affirming a judgment, sentence or order of a criminal court, this court may consider only questions of law (see, e.g., People v Robinson, 36 N.Y.2d 224, 228).
The power to vacate judgment upon the ground of newly discovered evidence and grant a new trial rests within the discretion of the court (see, e.g., People v Salemi, 309 N.Y. 208, 215; People v Patrick, 182 N.Y. 131, 178; People v Buchanan, 145 N.Y. 1, 30, app dsmd 158 U.S. 31 [capital cases]). It has been repeatedly held that this type of discretion is unlimited in the lower courts and thus this court has no power in a noncapital case to review their exercise of discretion (see People v Fein, 18 N.Y.2d 162, 169, app dsmd and cert den 385 U.S. 649; People v Mistretta, 7 N.Y.2d 843, 844; People v Girardi, 303 N.Y. 887, 888; People v Luciano, 275 N.Y. 547, 548; People v Bonifacio, 190 N.Y. 150, 152; Cohen and Karger, Powers of the New York Court of Appeals, p 586 and n 13; p 723 and n 97; p 745 and n 40; see, also, Cardozo, op. cit., at pp 49, 292; but see People v Welcome, 37 N.Y.2d 811 [a brief memorandum decision apparently assuming reviewability because the court did not dismiss the appeal, but relying only upon a capital case, People v Shilitano [ 218 N.Y. 161]).
Before completing discussion of the distinction between appealability and reviewability of issues in this court, reference should be made to People v Slaughter ( 37 N.Y.2d 596), recently decided by this court. There, under CPL 330.30 rather than CPL 440.10, a motion was made before judgment for a new trial upon the ground of newly discovered evidence. Hence, its denial was appealable to the Court of Appeals even under the statutes and practice antedating the CPL, as discussed above. Without more the exercise of discretion in the Slaughter case would not have been reviewable in this court, although the denial itself would be appealable. However, there was more. The allegedly newly discovered evidence related to possibly false testimony by police officers, a now classic ground for relief in coram nobis because of the "fraud on the court", even as the alleged suppression by the prosecutor of exculpatory evidence in this Crimmins case is both appealable and reviewable in this court, as discussed below. The court directly disposed of the issue therefore in the Slaughter case, citing notably to the treatise by Cohen and Karger (supra) at page 742, which refers to this court's very limited powers of review in noncapital criminal cases.
Moreover, even in a capital case, where its jurisdiction is not limited to issues of law, this court has been most reluctant to overturn the lower courts' exercise of discretion in denying a motion for a new trial upon the ground of newly discovered evidence (see, e.g., People v Salemi, 309 N.Y. 208, 215, supra; People v Patrick, 182 N.Y. 131, 178, supra; People v Buchanan, 145 N.Y. 1, 30, app dsmd 158 U.S. 31, supra).
Because this case has been so much litigated and because of the dissenter's plaint that justice has been denied, it is excusable, perhaps even necessary so that the specter of justice denied be laid to rest, to examine the question as if it were reviewable (cf. People v Fein, 18 N.Y.2d 162, 169, supra). Nevertheless, the denial, without a hearing, of the motion to vacate judgment upon the ground of newly discovered evidence was proper.
Whether or not to hold a hearing on a motion to vacate judgment upon the ground of newly discovered evidence is discretionary with the court to which the motion is addressed (see, e.g., People v Welcome, 37 N.Y.2d 811, supra; People v Eng Hing, 212 N.Y. 373, 386; People v Patrick, 182 N.Y. 131, 178, supra; People v White, 44 A.D.2d 749, 750; People v Dinan, 15 A.D.2d 786; People v Maynard, 80 Misc.2d 279, 283). As this court has made clear in capital cases where its power to review is not limited to the law, a hearing should be held to promote justice if the issues raised by the motion are sufficiently unusual and suggest searching investigation (People v Shilitano, 218 N.Y. 161, 169, 170, supra; see People v Arata, 254 N.Y. 565, 565-566; cf. People v White, 309 N.Y. 636, 640-641; see, also, People v Seidenshner, 152 N.Y.S 595). This standard is, of course, equally applicable to noncapital cases (see People v Luciano, 164 Misc. 167, 171, affd 251 App. Div. 887, app dsmd 275 N.Y. 547; People v Gordon, 142 Misc. 25, 25-26). To grant such a hearing where the court is able to reach its conclusion on the papers alone would serve no end of justice but would only protract futile litigation (see People v Luciano, 164 Misc, at p 171, supra).
Although defendant has limited her argument to whether she was entitled to a hearing on her motion, the dissent instead has addressed the merits of the denial of the motion. Indeed, the dissent offers no rationale for its unstated assumption that a hearing on the motion would somehow result in a decision more favorable to defendant from the hearing Judge. More likely is the People's characterization of what a hearing would accomplish: "Indeed, to have afforded appellant a hearing would merely have provided the prosecution with an opportunity to impeach [affiant's] credibility."
This is so because the affidavit is detailed and complete on its face. It obviously represents a determined effort by affiant, or whoever prepared his affidavit, to present all of the evidence he could possibly offer on defendant's behalf. Indeed, the affidavit goes beyond what affiant thought he had observed in the early morning of July 14; affiant gives his conclusion, based upon his later reading of a newspaper account of the killing, that the children he picked up were the Crimmins'. He offers an imaginative alternative hypothetical explanation, worthy of concoction by an A. Conan Doyle, consonant with affiant's sworn version of what he thought he saw and heard, of how the Crimmins children met their fate. Thus, a hearing on this motion would have been unnecessary and would have served only to postpone resolution of the controversy.
Moreover, the motion was denied by the very trial court which, in the dissenter's words, must have been sensitive to the "pulsebeat of the courtroom." Indeed, it is for this reason that the motion to vacate judgment must be made to the court in which it was entered (see CPL 440.10, subd 1).
However "understandable" affiant's reasons for the seven-year delay in disclosure may be regarded, the fact remains that affiant's recollection of the events of the early morning of July 14, 1965 must be recognized as inevitably influenced and inevitably colored by the passage of time and his acknowledged familiarity and undoubtedly obsessed concern with newspaper accounts of the case. Indeed, affiant admitted that his intriguing conclusion that the two children he picked up were the Crimmins' was the product of his independently-acquired newspaper "knowledge" of the case. Jurors and courts are confined, but also benefited, by a testimonial record developed in the adversarial process.
Moreover, even fully crediting affiant's sincerity, the denial, without a hearing, was proper because the court correctly assessed its probable impact upon the jury. Defendant has twice been convicted, after jury trials, of killing her daughter. As discussed above, her guilt was shown by overwhelming proof. There is no probability, as opposed to speculative possibility, that the jury would have returned a verdict more favorable to defendant had they had affiant's testimony. In saying even this, one must assume that affiant's testimony would be admissible, that is, that his belatedly disclosed, newspaper-stimulated description of the two children he "picked up" was sufficient to identify them as the Crimmins children, an uncertain assumption at best.
In sum, a hearing would only have delayed resolution of the motion supported by an affidavit complete on its face and, therefore, was not necessary. Moreover, there was no significant probability that the evidence in the affidavit if introduced at trial would have resulted in a verdict more favorable to defendant. Thus, the motion was properly denied without a hearing.
Defendant also moved to vacate judgment upon an additional ground, namely, that of asserted improper conduct by the prosecution in withholding from her information, provided by another deponent, allegedly helpful to her defense. This branch of her motion, brought under CPL 440.10 (subd 1, pars [f], [h]), is in the nature of an application for a writ of error coram nobis (see Denzer, Practice Commentary, McKinney's Cons Laws of NY, Book 11A, CPL 440.10, at p 183). The denial of this branch of her motion may, contrary to the rule which obtains with respect to newly discovered evidence, be reviewed by this court to determine whether there has been an abuse of discretion as a matter of law (see Matter of Bojinoff v People, 299 N.Y. 145, 151, supra; Cohen and Karger, op. cit., at p 745 and n 43). The distinction is based upon the character of the ancient remedy of the writ of error coram nobis, a remedy fashioned to inform the court of facts, such as, for example, fraud not reflected in the record, which as a matter of law would undermine the basis of its judgment (see, e.g., People v Sullivan, 3 N.Y.2d 196, 198; see, generally, Cohen, Post-Conviction Relief in the New York Court of Appeals: New Wine in Broken Bottles, 35 Brooklyn L Rev 1, 6). Even in coram nobis the findings of fact in the courts below are not reviewable in the Court of Appeals. The facts must be taken as found. The rules of law, however, as applied to the facts are, of course, reviewable in this court (see People v White, 309 N.Y. 636, 641, supra). When the issue in the Court of Appeals is whether in coram nobis a hearing was properly denied, that determination may be reviewed only for an abuse of discretion as a matter of law (see People v Richetti, 302 N.Y. 290, 294-295; accord, e.g., People v Beasley, 25 N.Y.2d 483, 487; People v Bagley, 23 N.Y.2d 814, 815; People v Silverman, 3 N.Y.2d 200, 203; People v Guariglia, 303 N.Y. 338, 342-343; cf. People v Session, 34 N.Y.2d 254, 256). The bifurcating principle is that in coram nobis the judgment is attacked as infirm, while on motion to vacate judgment upon the ground of newly discovered evidence the validity of the judgment is not attacked, only the likelihood of a similar verdict being rendered if there were an enlargement of the evidence on the principle issue.
The affidavit of the other deponent was insufficient to ground an inference of improper suppression of evidence by the prosecution. Perhaps most important it does not contain relevant or competent evidence on the issue of guilt or innocence, but at best supplies clues arguably meriting investigation. The defendant has now had for some time opportunity, through unusually adept lawyers, to make such an investigation. There is still nothing beyond a remote speculative inference to offset what may be a strange coincidence involving two other little children, crediting the affidavit without qualification. The issue was not the truth and accuracy of the affidavit but whether one could say that there had been an improper suppression by the police rather than a reasonable exercise of judgment. This issue was determinable from the papers. There is no disagreement in the court concerning this branch of defendant's motion.
It is true that the road to justice in this 10-year-old case has been long. Much too long, indeed, on any view. Its length should not be further extended, however, by proferred "newly-discovered" evidence, baseless and weightless, nor should it be extended by erroneous, however dextrous, misapplication of this court's limited power to review and violation of time-tested and accepted grounds for ordering new trials in criminal actions. The road to justice, short or long, should be the road to justice, and not a primrose path.
Accordingly, the order of the Appellate Division should be affirmed.
While I concur in the majority opinion affirming the Appellate Division's order, which affirmed an order of the Supreme Court, Queens County, denying defendant's motion to vacate a judgment of conviction and for a new trial upon the ground of newly discovered evidence, I do not retreat in any respect, except for the due regard which must be accorded a prevailing opinion, from the position taken and the dissent which I authored and in which I urged affirmance of an earlier order of the Appellate Division which reversed the judgment of conviction as to the manslaughter count, on the law, and granted a new trial on that count (see People v Crimmins, 36 N.Y.2d 230, 243 et seq.). The issues on this appeal and that one are not the same. Here, even if it were reviewable, it cannot be said as a matter of law that the court, in denying the application to vacate the judgment based on newly discovered evidence, abused its discretion.
The road to justice has been long in this 10-year-old case in which a mother was charged with the killing of her young son and daughter. Yet, I do not think we may say with any conclusiveness of law or fact that the end of that road was properly reached when, without even conducting a hearing, the trial court rejected defendant's motion for a new trial made on the basis of the affidavit of F. Sutherland Macklem, whose trustworthiness and whose averments have hitherto never been questioned. In the determination of that issue it should be irrelevant that she was twice found guilty by a jury. Her first conviction had to be reversed for jury misconduct; the second was flawed by now conceded trial error. The triumph of justice rests on its ability to overcome such impediments to a just decision.
Rather, as suggested by Justices RABIN and MARTUSCELLO at the Appellate Division ( 48 A.D.2d 663), the special circumstances here required that there be a plenary hearing as a precondition to the determination of whether the newly discovered evidence would have resulted in a verdict "more favorable to the defendant" (CPL 440.10, subd 1, par [g]). Accordingly, and for the further reasons which follow, I would modify the order and remand for a hearing.
The reversal of defendant's original conviction of manslaughter in the first degree for the death of her daughter, Alice Marie Crimmins, on the night of July 13, 1965 ( 33 A.D.2d 793), was affirmed by us unanimously in an opinion in which we said, with respect to the testimony of Sophie Earomirski concerning certain events which took place at 2 A.M. on July 14, 1965, that "the credibility of her testimony is essential to the prosecution's case." ( 26 N.Y.2d 319, 324.) In contrast, the same opinion assesses the first trial testimony of Joseph Rorech, to whom defendant is unconvincingly alleged to have made an oral confession, as a witness who was "seriously challenged" and "subjected to searching cross-examination" (p 324).
Before retrial, the District Attorney secured a superseding indictment, which, in addition to the manslaughter accusation with respect to the death of her daughter, also charged the defendant with murder in the first degree of her son Edmund Crimmins, Jr. There followed the second conviction, but the Appellate Division reversed and dismissed the Edmund murder count and ordered a new trial on the Alice manslaughter count. ( 41 A.D.2d 933. ) We affirmed the dismissal, thus terminating the Edmund charge, but, as to so much of the order as related to young Alice's death, the majority of our court voted to remand the case to the Appellate Division for determination of the facts. ( 36 N.Y.2d 230.)
The Appellate Division had reversed the second verdict because the trial had been riddled with multiple errors, which it found had deprived the defendant of a fair trial. The majority of our court, though accepting the existence of these errors, denoted them as "harmless" by articulating and applying a rule on the basis of which it determined that "there is no significant probability in the light of the overwhelming proof that, had it not been for the errors which occurred, [the] jury would have acquitted the defendant" (36 N.Y.2d, at p 243) and which it acknowledged was much less demanding than the Federal test under which it suffices that there be (p 241) "a reasonable possibility that the * * * [error] might have contributed to the conviction * * * (Fahy v. Connecticut, 375 U.S. 85, 86)."
It is noteworthy that none of the acknowledged errors involved the testimony of Sophie Earomirski. Instead, by emphasizing its "compelling" nature and that it was the only alleged "eyewitness testimony" (36 N.Y.2d, at p 242), the court made her testimony an error-free cornerstone for its characterization of the prosecution's case as "overwhelming". Indeed, the appraisal of the Earomirski testimony as pivotal was unanimous, for Judge COOKE in his dissent, in which I joined, accurately referred to her as a "major witness". (36 N.Y.2d, at p 248.) The Appellate Division's subsequent affirmance on the same facts, since it followed on our constraining decision, of course adds little to our discussion.
Viewed in this frame of reference, the detailed Macklem affidavit clearly goes right to the heart of the case. Significantly, there was proof that the children, one of whose bodies was later found alongside a highway and the other in a vacant lot, shared a room, the door to which was kept locked, after they retired, by means of a hook and eye catch their father had installed to keep them from raiding the refrigerator. The window provided a possible means of street level exit from the room. Mr. Macklem, an electronics scientist whose reliability has gone unchallenged, tells us that after the time when Mrs. Earomirski's testimony at least tended to establish that the defendant was seen removing the dead body of the child Alice from her home, he picked up two hitchhiking children on a roadway in the vicinity of the Crimmins residence. He adds that their description tallied with that of the Crimmins children in, among other things, their ages, their self-identification as brother and sister, the fact that the boy was the "bigger" of the two and the fact that the girl was referred to by the boy by the similar-sound appellation "my sister" or "Missy", the latter being the one by which little Alice was in fact called. Mr. Macklem relates, normally enough it seems to me, that he made the association between these two children and the Crimmins children when he read newspaper reports of the case only a few days later.
When scientific rules of probability are applied to the many points of correspondence, including the coincidence of the hour, place, date, age, sex and number of children, the mathematical odds that result so overwhelmingly weigh against the possibility that these were not the Crimmins children that, if the affidavit is taken at face value, it is impermissible to say that such testimony might not have made the difference between a verdict of innocence and one of guilt. If Macklem's testimony stood up in accordance with his affidavit, it would knock out such a vital part of the underpinning for the denomination of the case presented against the defendant at her second trial as "overwhelming" so that, even allowing for the newly enunciated harmless error doctrine, the so-called "harmless" errors almost automatically would have to be elevated to the status of reversible ones.
This is all the more so since I believe a painstaking analysis of the trial proof left plenty of room to question whether this was an appropriate case to apply the broadened definition of harmless error in the first place. Ready illustration is found in the three items of proof on which the majority here relies so heavily to paint a picture of overwhelming guilt, e.g. the so-called confession to Rorech, the testimony of dust given by Detective Piering and the medical examiner's testimony concerning what light the autopsy shed on the time of death. Each one but compounds the doubt.
The relationship between Rorech and the defendant appears to have been a most intimate one. But it comes through from the record with crystal clarity that neither the long-standing marital discord which separated her husband and herself or the interest she developed in other men detracted from her devotion as a mother to her two small children and her resolute determination that she continue to have custody of them. It is against that background that we must view Rorech's testimony that on one overwrought occasion the defendant uttered the words, "Joseph, forgive me, I killed her". That was the alleged confession. Though Rorech and the defendant were with each other constantly, almost unbelievably he maintained that she, neither on that occasion or any other, ever offered or was questioned by him as to any details. There was no explanation as to how the crime was perpetrated, no excuses or rationalizations for its commission, no justifications to mitigate moral guilt for such an horrendous and unnatural a double infanticide, no remorse, not even any troubled talk. Just the bold statement.
The words of this sparse confession must raise more psychological and factual questions than they answer. Is there anyone with some experience in life who has not witnessed the emotional and verbal self-flagellation of a parent who, no matter how blamelessly, has suffered the death of a child, whether through illness or violence or even suicide, expressed in almost identical language? Further, for about two years after the words were supposed to have been said, Rorech, though himself a suspect, neither in many meetings with the defendant's then counsel, to whom he had recommended her, nor on his two different Queens County Grand Jury appearances, or in questioning by the Nassau County police, even once mentioned any such statement. It appeared on the horizon for whatever it was worth only 48 hours before the trial and after he had been supplanted in defendant's affections by one Anthony Grace and had worked out an immunity arrangement with the prosecution. The recognized need to bolster his testimony indeed appears to have led to one of the trial errors, an improper revelation to the jury of the administration of a sodium pentothal test to the witness, interestingly enough on an occasion before there was any recollection of a confession.
The proof about dust was at least equally vulnerable to attack. In order to controvert any argument that the children had exited through the window, the prosecuting attorney put on Detective Piering to testify that a chest below the window had an undisturbed film of dust, visible to him when a clean circle was formed upon his removal of a round lamp base. It turned out he could not have so visualized it because the lamp base had three separate, discrete legs. Having testified that it was only a "faint film" at the first trial two years after his observations, by the second one, when the issue had escalated in importance, he described the chest as "covered" with dust and stated that his recollection on this point was refreshed only two weeks earlier by a review of notes on this matter contained in police records. An examination of the police records and of Piering's own notes showed they contained no entry of dust whatsoever. Nor was there any such finding of dust in the special report of forensic experts from the police laboratory who promptly made a fine tooth comb examination of the children's bedroom. No wonder the prosecutor on summation found it necessary to explain, "Now, Piering, he's no genius. He's no Ellery Queen".
So far as the autopsy is concerned, it is claimed that the contents of the daughter's stomach proved she died sometime between 8:30 and 9:30 P.M., so that she could not have been alive at the time Mrs. Earomirski claims she saw the "bundle". That operates on the assumption the child's last meal was taken at about 7:30. The fact is the autopsy proves no such thing. Dr. Grimes, the medical examiner, who first examined the body the next afternoon, did not profess to know when the last meal was taken; he merely testified that from the undigested state of the food she died shortly after ingesting a meal, whenever that was. In fact, since the best, uncontradicted and most disinterested testimony was that the child last ate at home at 7:30 P.M. and was last known to be alive at 9 P.M., the absence of any evidence of digestion was incompatible with the passage of the hour and a half, since the evidence proved that digestion begins at once and is a continual and progressive process. Furthermore, the contents of the meal she had at home, described to the police by the mother well before there was any autopsy report, varied considerably from the different foods found on the examination and, therefore, suggested that the child might have had another meal at some unknown time and unknown place considerably after the one taken at home. Thus, the autopsy proof, far from concluding the matter, left very much open the question of whether she was alive at the time covered by the Earomirski testimony and, incidentally, would have made the testimony of Mr. Macklem all the more vital.
All this is not to say that Alice Crimmins was not in fact guilty. She may very well have been. But it points up the essentially judgmental and factual nature of the preliminary determinations which have to be passed upon when we apply a broad harmless error rule. And, in accordance with the analogy which the majority, correctly I believe, draws between that rule and the standard for determining when a motion to vacate a judgment upon the ground of newly discovered evidence should be granted, the latter calls for a similarly permissible exercise by us of power to determine whether in a particular judgmental and factual setting there has been an abuse of discretion as a matter of law. In either case that does not mean we are passing on facts as such, but rather considering them to the extent that they are a foundation for the application of law.
I dissented from the "harmless error" decision because, once meaningful error is established, I do not believe that, from the remote fastness of an appellate court, the pulsebeat of the courtroom can be perceived with sufficient reliability to predict whether, when there has been such error, the result would have been affected. Put another way, an appellate court is too different from a lay jury to permit one to substitute for the other. However, accepting that decision as a fait accompli and now acting on constraint of it, I respectfully suggest that, in light of the Macklem affidavit, any argument about harmless error begs the question.
This is not an instance of a quest for a new trial on evidence that was or, by the exercise of due diligence, could have been available to the defendant at an earlier time. Concededly, immediately upon its discovery, defendant moved with great expedition. Nor is this a situation where an investigation by the District Attorney has resulted in the disclosure of even a single collateral fact making the affiant's assertions suspect. In fact, there is not the slightest suggestion that Mr. Macklem was unwilling to expose his story to interrogation by the People, in either a formal or an informal setting, whether at the initiative of the People or of the defendant. Defendant's counsel did volunteer to produce him. And the substance of the affidavit, obviously, is not merely cumulative or impeaching in nature.
The fact that Mr. Macklem did not come forward until nearly seven years after the event is not to be blithely ignored. But, besides being a circumstance beyond the control of the defendant, standing alone, especially in a case of the seriousness and complicated litigation history of the one before us, that fact was not enough to allow for the trial court's rejection of the motion out of hand. The delay merely emphasized the urgent need for the kind of scrutiny afforded only by an adversary hearing, with its opportunity for direct personal confrontation. This is all the more so here since, curiously, Mrs. Earomirski, as to whom both trial records reveal evidence of instability, had herself not come forward as a witness until about two years after the crime. For that matter, as already indicated, that was true of Rorech as well.
Mr. Macklem himself does not leave his delay unexplained. His affidavit tells us the following:
"The first time I realized that these two children were the Crimmins children was when I saw the report in the Thursday edition of the New York Times.
"Because there was no mention in the press of a possible suspect, I refrained from coming forward, because I was terrified that I might be wrongly accused of a crime.
"I was certain that the children were the Crimmins' children because I left them at 162nd Street; because their ages seemed to coincide with the ages reported in the press, because the children were brother and sister and the brother was bigger.
"I followed the first trial in the newspapers and when I read what was alleged to have happened I then knew that it was impossible. The little girl could not have been killed two to four hours after a 7:30 or 8:00 o'clock meal.
"The girl and boy were very much alive when I left them and Mrs. Earomirski could not have seen what she claimed to have seen at 2:00 A.M. July 14.
"When Mrs. Crimmins was convicted the first time, I talked to my lawyer about the case, pretending that I was discussing it purely as a matter in the news. He said that the conviction would not stand and that she would surely be released on bail shortly.
"The more I thought of the case, the more I convinced myself that she had to be acquitted during the second trial.
"Having kept silent that long, I allowed myself to maintain my silence, particularly because this was not an ordinary public trial, but a brutally publicized trial, and I feared the publicity that would result from any revealed connection with the case.
"When the second trial resulted in a conviction for murder, I could not longer remain silent, but confided in my attorney, and asked him to contact the attorneys for Mrs. Crimmins."
Though his conduct hardly comports with the civic ideal, an elementary knowledge of human nature makes Mr. Macklem's actions quite understandable. Hesistation about becoming "involved" often breeds delay, which in turn leads to embarassment at having already delayed so long.
In these circumstances, it appears to me that it was an abuse of discretion as a matter of law not to grant a hearing on the basis of Mr. Macklem's affidavit. The difference between a determination on papers and one that permits confrontation with a human being is not a slender one and is not to be discounted casually. The appraisal of Mr. Macklem's affidavit involved an evaluation of his powers of observation, recollection, motivation, intelligence and veracity. Were they all to be assumed the motion should have been granted. If they were not, since no basis for impeachment of any of these subjective qualities was offered by the prosecution, a determination should only have been made after testing the affiant with the aid of tried tools such as cross-examination.
Until today, I thought that we had a reservoir of power to review an abuse of discretion in a criminal case, even though we have exercised it sparingly (People v Slaughter, 37 N.Y.2d 596; People v Welcome, 37 N.Y.2d 811; People v Englese, 7 N.Y.2d 83; Matter of Bojinoff v People, 299 N.Y. 145, 151; People v La Brake, 28 N.Y.2d 625, revg on dissent with additional comments 35 A.D.2d 631; Cohen and Karger, Powers of the New York Court of Appeals, § 198, p 745 [1952 rev ed]). I agree with the majority that questions of fact may not be reviewed, but whether there has been an abuse of discretion is a question of law, not of fact. It matters not that the discretion which we review necessarily was exercised in a factual setting. And that is so even though we must look at the facts in order to determine whether the discretion was indeed abused. As has been recently said in a somewhat similar context, "more than a merely mechanical categorization between law and fact is merited if this Court is to perform its great function to review." (People v Santos, 38 N.Y.2d 173, 176.)
Since, in the absence of any enabling statute, the right of appeal to us from a postjudgment application for a new trial based upon freshly discovered evidence did not exist prior to the enactment of the CPL in 1971, with the fullest respect to my colleagues, the prior history which the majority opinion relates is, therefore, hardly relevant. Because there was no right of appeal there was then, a fortiori, also no right of review. Given this fact, a close reading of the cases decided during that period indicates that, when they speak in terms of reviewability, they must be interpreted either as engaging in dicta or as loosely referring to nonreviewability when nonappealability was intended. Indeed, one suspects that such verbal usage had its genesis in the fact that, at that time, postjudgment orders relating to motions on newly discovered evidence were not part of the judgment roll and, accordingly, were not even physically available for "review" (see People v Priori, 163 N.Y. 99, 107-108).
In any event, none of the cases suggest any constitutional limitations on the right of review. And now that the Legislature has granted a right of appeal, the concomitant right to review on such appeal should follow. It hardly seems likely that those who gave birth to the statute intended it to be stillborn. For a statute usually has some purpose or object to accomplish and we are required to so assume (Montgomery v Daniels, 38 N.Y.2d 41).
Significantly, not a single case in our court since the CPL came into effect in 1971 has found an order denying such a motion for new trial nonreviewable here. Indeed, at about the very time the present case was argued, we decided People v Welcome ( 37 N.Y.2d 811, 812-813, supra), where in reviewing an identical order denying a new trial based upon freshly discovered evidence, we specifically ruled on whether "the Trial Judge abused his discretion in denying the motion without a hearing". Similarly, in People v Slaughter ( 37 N.Y.2d 596, supra), as a fair reading of its opinion makes clear, we again exercised the power to review. The importation pro hac vice of a possible rationale not there expressed does not detract from our point.
Furthermore, CPL 440.10 makes no distinction between posttrial, postjudgment motions, irrespective of whether they are based on newly discovered evidence or not. It also makes no special distinction for cases that, before 1971, fell within the traditional scope of the common-law writ of coram nobis, which is now embraced within the same statutory scheme that applies to all discretionary postjudgment motions. There is no reason now, on the basis of a history whose premise is gone, to carve out an exception for motions on the basis of newly discovered evidence.
In reviewing whether discretion has here been abused it is well also to consider that legislative changes have relaxed the statutory standard for a hearing. So the new evidence must now be only "of such character as to create a probability that had such evidence been received at the trial the verdict would have been more favorable to the defendant." (CPL 440.10, subd 1, par [g]; People v Maynard, 80 Misc.2d 279, 283-284; see, also, 6 Zett, New York Criminal Practice, par 50.3, subd , par [g]; cf. ABA Project on Minimum Standards for Criminal Justice, Standards Relating to Post Conviction Remedies, § 2.1, subd [a], par [v]; Uniform Post Conviction Procedure Act, § 1, subd a, par ; Kyle v United States, 297 F.2d 507, 514.) The support for the motion here met all the cited standards.
In People v Rensing ( 14 N.Y.2d 210, 214), after sentencing it was disclosed that Rensing's codefendant, who gave damaging testimony against him at trial, was insane at the time and had a long history of mental illness. Despite overwhelming evidence of guilt, including a confession, our court reversed Rensing's conviction and ordered a new trial without a hearing. Judge (later Chief Judge) FULD went on to distinguish People v Salemi ( 309 N.Y. 208), on which the majority here relies, noting "that the matter relied upon by Salemi for a new trial `was not new matter at all, but had been before the court and jury on the main trial' (309 N.Y., at p. 212)." ( 14 N.Y.2d, p 214.) True, the reversal in that case took place in a capital case, where we have the power to review the facts as such, but it appears to me that this should have no bearing on the threshold question, raised by a motion for a new trial based on newly-discovered evidence, as to whether a hearing to develop facts should be held in the first place. (Cf. People v Ledwon, 153 N.Y. 10; People v Silver, 33 N.Y.2d 475.)
Therefore, while I am in accord with the majority on the separate matter raised by the Feldman affidavits, I conclude that the failure to order at least a hearing on the Macklem affidavit, which stands uncontradicted and unimpeached, was an abuse of discretion as a matter of law and, accordingly, I would so modify the order of the Appellate Division.
Judges JASEN, GABRIELLI, JONES and WACHTLER concur with Chief Judge BREITEL; Judge COOKE concurs in a separate opinion; Judge FUCHSBERG dissents and votes to modify in another separate opinion.