Opinion
Indictment No. 39-1996
01-05-2023
The PEOPLE of the State of New York v. Andrew KRIVAK, Defendant.
Robert V. Tendy, District Attorney (Larry Glasser of counsel), for the People. Oscar Michelen and Karen A. Newirth for defendant. Tania Brief and Lauren Gottesman for Innocence Project, Inc., amicus curiae.
Robert V. Tendy, District Attorney (Larry Glasser of counsel), for the People.
Oscar Michelen and Karen A. Newirth for defendant.
Tania Brief and Lauren Gottesman for Innocence Project, Inc., amicus curiae.
Robert J. Prisco, J.
On May 19, 2022, the People filed a motion for an Order seeking, inter alia , to preclude the testimony of Kyle C. Scherr, Ph.D. (hereinafter "Dr. Scherr"), on the subject matter of false confessions or, in the alternative, to hold a hearing as to the scope of such testimony. The People also sought to preclude the testimony of Robert A. Leonard, Ph.D. (hereinafter "Dr. Leonard"), or alternatively, to hold a hearing pursuant to Frye v. United States , 293 F. 1013 [D.C. Cir. 1923], regarding his proposed testimony on the subject matter of linguistics, i.e., the science of language analysis.
On June 14, 2022, Defendant filed an Affirmation in Support and a Memorandum of Law opposing the People's motion to preclude the testimony of Defendant's proposed experts.
On June 21, 2022, the Innocence Project, Inc., served and filed an Amicus Brief seeking the Court's denial of the People's motion to preclude the testimony of Dr. Scherr.
On June 22, 2022, the People filed an Affirmation and a Memorandum of Law in response to Defendant's opposition papers.
On July 13, 2022, the Innocence Project, Inc., served and filed a response to the People's opposition papers.
On October 5, 2022, the Court issued a Decision and Order directing that a hearing take place to determine the scope of the testimony to be offered by Dr. Scherr and the degree, if any, to which Dr. Scherr could rely upon a report that was prepared in connection with a polygraph examination to which the defendant submitted. The Court also ordered a hearing to determine the admissibility of the proposed testimony of Dr. Leonard.
On November 7 and 9, 2022, testimony was taken from Dr. Scherr, a Professor of Psychology at Central Michigan University, and a proposed expert in police interrogations, confession decision-making, and wrongful convictions and exonerations.
On November 9, 10, 14, and 15, 2022, testimony was elicited from Dr. Leonard, a proposed expert in linguistics who is, among other positions, a Professor of Linguistics at Hofstra University, and the Director of Hofstra's Forensic Linguistics Graduate Program, the University's Institute for Forensic Linguistics, Threat Assessment, and Strategic Analysis, and of the Forensic Linguistics Capital Case Innocence Project at the University.
On November 15, 16, 21, and 22, 2022, Natalie Schilling, Ph.D. (hereinafter "Dr. Schilling"), a Professor of Linguistics at Georgetown University, and a proposed expert in forensic linguistic matters such as authorship attribution, speaker identification and profiling, testified in rebuttal to Dr. Leonard.
On November 29, December 1, and 8, 2022, counsel argued their respective positions based upon the testimony of Drs. Scherr, Leonard, and Schilling, the other evidence that was submitted over the course of the hearing, and the applicable case law.
After consideration of the above, the Court makes the following determination:
Frye Analysis
Pursuant to Frye , "the court [is] required to determine whether the proposed expert opinion testimony was based on principles and methodologies generally accepted within the relevant scientific community" ( People v. Powell , 37 N.Y.3d 476, 489, 162 N.Y.S.3d 235, 182 N.E.3d 1028 [2021] ; see Matter of State of New York v. Ronald S. , 186 A.D.3d 1227, 1228, 129 N.Y.S.3d 486 [2d Dept. 2020] ; People v. Wilson , 107 A.D.3d 919, 919-920, 967 N.Y.S.2d 756 [2d Dept. 2013], lv. denied 21 N.Y.3d 1047, 972 N.Y.S.2d 544, 995 N.E.2d 860 [2013] ). After reviewing their qualifications and listening to their testimony, the Court has determined that Drs. Scherr, Leonard, and Schilling are qualified to render expert testimony as they "possess adequate skill, training, education, knowledge or experience to proffer it" ( People v. Gilbert , 199 A.D.3d 1048, 1057, 156 N.Y.S.3d 533 [3d Dept. 2021], quoting People v. May , 188 A.D.3d 1309, 1310-1311, 133 N.Y.S.3d 685 [3rd Dept. 2020], lv. denied 36 N.Y.3d 974, 138 N.Y.S.3d 457, 162 N.E.3d 686 [2020] ; see People v. Lee , 96 N.Y.2d 157, 162, 726 N.Y.S.2d 361, 750 N.E.2d 63 [2001] ; People v. Godallah , 132 A.D.3d 1146, 1150, 19 N.Y.S.3d 119 [3rd Dept. 2015] ).
The curriculum vitae for the proposed experts were admitted into evidence.
While recognizing that "the inquiry under Fry e ‘is separate and distinct from the admissibility question applied to all evidence - whether there is a proper foundation - to determine whether the accepted methods were appropriately employed in a particular case’ " ( People v. Powell , 37 N.Y.3d at 489, 162 N.Y.S.3d 235, 182 N.E.3d 1028, quoting People v. Brooks , 31 N.Y.3d 939, 941, 73 N.Y.S.3d 110, 96 N.E.3d 206 [2018] ; see People v. LeGrand , 8 N.Y.3d 449, 457, 835 N.Y.S.2d 523, 867 N.E.2d 374 [2007] ; Parker v. Mobil Oil Corp. , 7 N.Y.3d 434, 447, 824 N.Y.S.2d 584, 857 N.E.2d 1114 [2006] ), in the interest of judicial economy, evidence pertaining to the admissibility and scope of Defendant's proposed experts’ testimony at trial was also submitted and addressed.
This evidence was presented over the course of eight (8) days.
DR. KYLE C. SCHERR
Testimony of Dr. Scherr
Dr. Scherr testified that he was retained by the defense to evaluate the facts and evidence provided to him for the purpose of identifying known risk factors that can lead to a false confession. Dr. Scherr generated a report dated October 27, 2021, in which he identified nine (9) situational factors associated with false confessions that were allegedly present during Defendant's interrogation, to wit: high pressure presumption of guilt, the polygraph environment, minimization and implied promises of leniency, contamination, duration, the error correction tactic, failure to corroborate, the failure to record, and misinformation. According to Dr. Scherr, as the number of factors increase, so does the risk that an innocent person will falsely confess. Dr. Scherr further testified that research shows that people have a lack of appreciation and awareness of situational factors that can put an innocent person at risk for falsely confessing.
Dr. Scherr also prepared a supplement to his report dated June 8, 2022.
No dispositional factors relating to Defendant were addressed by Dr. Scherr in his hearing testimony. However, this Court does not interpret People v. Bedessie , 19 N.Y.3d 147, 947 N.Y.S.2d 357, 970 N.E.2d 380 [2012], as requiring that a proposed expert's testimony must address both dispositional and situational factors to be admissible (emphasis added) (see People v. Boone , 146 A.D.3d 458, 459, 46 N.Y.S.3d 520 [1st Dept. 2017], lv. denied 29 N.Y.3d 1029, 62 N.Y.S.3d 298, 84 N.E.3d 970 [2017] ).
Dr. Scherr cited to a study conducted by Alceste [2020], which showed that ninety-four (94) percent of respondents indicated that in cases of disputed confessions, expert testimony would aid in their understanding of the factors that can lead to false confessions.
The Law
"[T]he ‘phenomenon of false confessions’ is not only genuine, but has ‘moved from the realm of startling hypothesis into that of common knowledge, if not conventional wisdom’ " ( People v. Days , 131 A.D.3d 972, 977, 15 N.Y.S.3d 823 [2d Dept. 2015], lv. denied 26 N.Y.3d 1108, 26 N.Y.S.3d 767, 47 N.E.3d 97 [2016] ), quoting People v. Bedessie , 19 N.Y.3d 147, 156, 947 N.Y.S.2d 357, 970 N.E.2d 380 [2012] ; see People v. Powell , 37 N.Y.3d 476, 490, 162 N.Y.S.3d 235, 182 N.E.3d 1028 [2021] ; People v. Evans , 141 A.D.3d 120, 128, 32 N.Y.S.3d 119 [1st Dept. 2016], appeal dismissed 28 N.Y.3d 1101, 45 N.Y.S.3d 362, 68 N.E.3d 89 [2016] ). "This, however, does not mean that expert testimony on the theories behind the reasons for false confessions is rendered unnecessary. Certain aspects of the scientific study of the phenomenon might well be outside the ken of the typical juror" ( People v. Powell , 37 N.Y.3d at 490, 162 N.Y.S.3d 235, 182 N.E.3d 1028, citing People v. Bedessie , 19 N.Y.3d at 156, 947 N.Y.S.2d 357, 970 N.E.2d 380 ; see People v. Lee , 96 N.Y.2d at 162, 726 N.Y.S.2d 361, 750 N.E.2d 63 ; People v. Churaman , 184 A.D.3d 852, 853, 126 N.Y.S.3d 487 [2d Dept. 2020] ; People v. Days , 131 A.D.3d at 979, 15 N.Y.S.3d 823 ), and "there is no doubt that experts in such disciplines as psychiatry and psychology or the social sciences may offer valuable testimony to educate a jury about those factors of personality and situation that the relevant scientific community considers to be associated with false confessions" ( People v. Bedessie , 19 N.Y.3d at 161, 947 N.Y.S.2d 357, 970 N.E.2d 380 ; see People v. Hansson , 162 A.D.3d 1234, 1237, 79 N.Y.S.3d 341 [3d Dept. 2018], lv. denied 32 N.Y.3d 1004, 86 N.Y.S.3d 762, 111 N.E.3d 1118 [2018] ; People v. Evans , 141 A.D.3d at 129, 32 N.Y.S.3d 119 ; People v. Days , 131 A.D.3d at 979, 15 N.Y.S.3d 823 ).
Expert testimony "is meant to be used as an informational tool to educate the jury on the causal connection between relevant factors and false confessions outside their ken in order to address common misconceptions about a person making a false admission of criminal conduct" ( People v. Powell , 37 N.Y.3d at 490-491, 162 N.Y.S.3d 235, 182 N.E.3d 1028 ; see People v. Evans , 141 A.D.3d at 129, 32 N.Y.S.3d 119 ). However, to be admissible, the expert's proposed testimony must be "relevant to the defendant and interrogation before the court" ( People v. Bedessie , 19 N.Y.3d at 161, 947 N.Y.S.2d 357, 970 N.E.2d 380 ; see People v. Hansson , 162 A.D.3d at 1237, 79 N.Y.S.3d 341 ; People v. Jeremiah , 147 A.D.3d 1199, 1204-1205, 47 N.Y.S.3d 490 [3d Dept. 2017], lv. denied 29 N.Y.3d 1033, 62 N.Y.S.3d 302, 84 N.E.3d 974 [2017] ; People v. Boone , 146 A.D.3d 458, 459, 46 N.Y.S.3d 520 [1st Dept. 2017], lv. denied 29 N.Y.3d 1029, 62 N.Y.S.3d 298, 84 N.E.3d 970 [2017] ; People v. Lippe , 145 A.D.3d 1035, 1036, 44 N.Y.S.3d 199 [2d Dept. 2016], lv. denied 30 N.Y.3d 1020, 70 N.Y.S.3d 453, 93 N.E.3d 1217 [2017] ; People v. Evans , 141 A.D.3d at 129, 32 N.Y.S.3d 119 ; People v. Days , 131 A.D.3d at 979, 15 N.Y.S.3d 823 ), i.e., "the specific circumstances of [the] case" ( People v. Dockery , 186 A.D.3d 1710, 1711, 129 N.Y.S.3d 829 [2d Dept. 2020] ; People v. Iqbal , 147 A.D.3d 782, 783, 45 N.Y.S.3d 580 [2d Dept. 2017], lv. denied 29 N.Y.3d 1092, 63 N.Y.S.3d 8, 85 N.E.3d 103 [2017] ; People v. Joubert , 125 A.D.3d 686, 686, 999 N.Y.S.2d 552 [2d Dept. 2015], lv. denied 25 N.Y.3d 1166, 15 N.Y.S.3d 298, 36 N.E.3d 101 [2015] ). In determining the relevancy of the proposed expert testimony, the trial court must examine " ‘the nature of the interrogation, the applicability of the science of false confessions to the defendant and the extent to which the People's case relie[s] on the confession’ " ( People v. Jeremiah , 147 A.D.3d at 1205, 47 N.Y.S.3d 490, quoting People v. Evans , 141 A.D.3d at 126, 32 N.Y.S.3d 119 ; see People v. Hansson , 162 A.D.3d at 1237, 79 N.Y.S.3d 341 ; People v. Days , 131 A.D.3d at 979-981, 15 N.Y.S.3d 823 ).
In the instant case, Dr. Scherr has opined that certain conditions of the defendant's interrogation suggest that he could have been induced to confess falsely and that scientific research has shown that the techniques employed by the investigators are linked to eliciting false confessions. While there is other evidence tying Defendant to the charged crimes, "the confession [is] a central component of the People's case, and thus does not undermine the usefulness of expert testimony on the issue of false confessions" ( People v. Evans , 141 A.D.3d at 126, 32 N.Y.S.3d 119, citing People v. Days , 131 A.D.3d at 981, 15 N.Y.S.3d 823 ). Accordingly, the Court will review the conditions and risk factors that were allegedly present during Defendant's interrogation and which Dr. Scherr has identified as being associated with false confessions, with an understanding that "not all categories of such testimony are applicable or relevant in every case" ( People v. Berry , 27 N.Y.3d 10, 20, 29 N.Y.S.3d 234, 49 N.E.3d 703 [2016], quoting People v. LeGrand , 8 N.Y.3d at 459, 835 N.Y.S.2d 523, 867 N.E.2d 374 ; see People v. Austin , 46 A.D.3d 195, 200, 845 N.Y.S.2d 315 [1st Dept. 2007], lv. denied 9 N.Y.3d 1031, 852 N.Y.S.2d 16, 881 N.E.2d 1203 [2008] ).
While the Court appreciates the perceived incompatibility between Defendant's argument that his alleged confession was "a police-drafted statement that he was coerced to sign" and his argument that the statement was an induced false confession, because he allegedly adopted the statement under either posited situation by signing and initialing every page thereof under the penalty of perjury, the Court will consider the risk factors addressed by Dr. Scherr.
Guilt Presumptive Interrogation
Dr. Scherr testified that guilt presumption on the part of an interrogator will motivate the approach taken in questioning a suspect, often resulting in aggressive interrogations and high-pressure interactions to elicit a confession. However, as it pertains to the existent circumstances of the instant case, while Dr. Scherr testified that refusals to believe Defendant's denials and the use of the Arther technique demonstrate "high pressure" or "high confrontation," he acknowledged that there is nothing in the record or the provided materials that show that the interrogation tactics used by the investigators were "aggressive" and, in fact, the materials indicate that the defendant was not yelled at or the subject of physical abuse, and that he was given food, drink, and the opportunity to smoke a cigarette. In fact, when the Court specifically asked Dr. Scherr if there was anything in the record or the materials that suggested an "aggressive" or "high-pressure interaction" between Defendant and the investigators, he responded that there was not.
In Dr. Scherr's report, he stated that the purpose of the Arther technique "is to elicit a confession from a suspect" and that "research has deemed [the technique as] highly biased against the innocent." Specifically, Dr. Scherr stated that, "when used, it is not employed as an established credibility assessment, but rather as an additional tactic to extract a confession."
Additionally, Dr. Scherr testified that an interrogator's presumption of a suspect's guilt will often lead to tunnel vision, which results in failures to examine alibis, corroborate statements, and to investigate exculpatory evidence. He also admitted, however, that the lack of follow-up activity by the investigators is an after-the-fact matter that does not induce false confessions. Accordingly, because no evidence or circumstances have been presented to the Court linking the investigators’ presumption of Defendant's guilt with the alleged elicitation of a false confession from him, "any expert testimony on that score would be speculative and delivered in a factual vacuum" ( People v. Powell , 37 N.Y.3d at 492-493, 162 N.Y.S.3d 235, 182 N.E.3d 1028 ; see People v. Ignatyev , 147 A.D.3d 489, 491, 47 N.Y.S.3d 285 [1st Dept. 2017], lv. denied 29 N.Y.3d 1033, 62 N.Y.S.3d 302, 84 N.E.3d 974 [2017] ) and is, thus, inadmissible.
Promises of Leniency
Dr. Scherr testified that promises of leniency are a type of minimization and that there is evidence that such took place in this case. Specifically, Dr. Scherr referred to Defendant's statement in his post confession interview that he confessed because he believed that he could plead to something less severe if he cooperated, as indicative of such. Dr. Scherr also referred to Question 7 of that interview which asks, "What did I say that finally convinced you to tell the truth?," and Defendant's response that, "They told me if it wasn't just murder, I could plead to something else," as another example, and to Investigator Stephens’ statement to Defendant that, "the [polygraph] machine couldn't tell if it was intentional" or "accidental," as an example of a statement that implied that a less severe consequence would result from an admission.
As Dr. Scherr has demonstrated that scientific research recognizes "Promises of Leniency" and "Minimization" as techniques that are linked to eliciting false confessions, and as the examples above are indicative that these techniques were utilized with the defendant, Dr. Scherr will be permitted to testify regarding this particular risk factor.
Contamination
Dr. Scherr testified that contamination is the likelihood that information already known to police is "fed" to a suspect, which the suspect then offers as their confession and admission of guilt. According to Dr. Scherr, the effect of the leaked information is that it becomes very hard to distinguish between the information that is true or false in a statement, and that it makes the confession more believable because the suspect allegedly has non-public knowledge about the crime.
In support thereof, Dr. Scherr cited to the research studies and findings of Brandon Garrett.
Dr. Scherr stated in his report that, "[i]n many proven cases involving false confessions, the narrative admission that the innocent suspect offered was often compelling and persuasive but not because the innocent person had guilty knowledge. Rather, the innocent confessor's narrative contained non-public and guilty knowledge because that information was learned during the interrogation through, for example, evidence being shown to them or the content of the questions themselves."
It was Dr. Scherr's testimony that when the investigator showed Defendant "the van" parked outside of a police department window and told him that, "it was going to be a problem," this statement suggested to the defendant that the vehicle was connected to the alleged crimes. Dr. Scherr further testified that after Defendant was told that three witnesses implicated him by saying that he was in the van with the victim, such information was included in his statement. However, Dr. Scherr acknowledged that there was nothing in the record indicating that the investigators gave the defendant any factual information about what the witnesses alleged had occurred, the forensic or physical evidence from the scene where the victim's remains were found, or about the specific pieces of jewelry that were allegedly found in the van.
The Court recognizes some apparent inconsistencies in Dr. Scherr's testimony. For instance, on cross-examination, Dr. Scherr agreed that contamination does not cause a false confession and stated, "I don't think that claim has ever been made." However, on re-direct, Dr. Scherr acknowledged that contamination is a factor that research shows increases the risk of false confessions and that witness statements can also be contaminated by the police.
Dr. Scherr also testified that Investigator Castaldo either intentionally or unintentionally may have conveyed information about the crime scene to Defendant and Denise Rose, as both indicated in their statements that the victim's hands were tied in front, which was eventually shown not to be the case but was Investigator Castaldo's alleged theory at the time.
As Dr. Scherr has shown that scientific research recognizes "Contamination" as a technique that is linked to eliciting false confessions, and as the examples above are indicative that these techniques were utilized with the defendant, Dr. Scherr will be permitted to testify regarding this particular risk factor.
Duration
In his testimony, Dr. Scherr stated that the evidence and facts that he was provided indicate that Defendant was interrogated for over seven hours. Dr. Scherr also testified that surveys of law enforcement agencies indicate that, on average, their interrogations last between thirty minutes to two hours, and that one of the top training manuals for interrogations submits that interrogations should rarely, if ever, go more than four hours. Thus, according to Dr. Scherr, the duration of Defendant's interrogation was longer than suggested by the best practices of law enforcement.
Dr. Scherr appears to have conflated "the period of arrest and the period of interrogation" (People v. Powell , 37 N.Y.3d at 493, 162 N.Y.S.3d 235, 182 N.E.3d 1028 ; see Miranda v. Arizona , 384 U.S. 436, 86 S.Ct. 1602, 16 L.Ed.2d 694 [1966] ), when he stated, "I think the fact that he was in custody means that he was being interrogated."
Dr. Scherr referenced a study conducted by Mindthoff titled, "A survey of potential jurors’ perceptions of interrogations and confessions." [2018], in which jurors stated, on average, that an interrogation lasting eight (8) hours is necessary to get a statement from a suspect, which goes against the best practice in many of the interrogation manuals.
Dr. Scherr further asserted in his report that interrogations that continue for extended durations are problematic because, "[a]cross time, innocent suspects’ repeated denials are continually met with (and often times more intensely met with) refusals to believe the denials," resulting in the interrogation becoming more confrontational, the suspect's will to resist being broken down, and a depletion of the suspect's cognitive resources. Consequently, the suspect comes to realize that the only way out of this interrogation, i.e., "their only escape hatch," is through a confession and confirming the interrogator's expectation of their guilt. As it pertains to the instant case, Dr. Scherr found that Defendant had continually denied his involvement from the commencement of the questioning by Investigators Castaldo and Quick through the administration of the polygraph examination with Investigator Stephens. Dr. Scherr further concluded that Defendant's will to resist was exhausted and depleted after the passage of several hours and after he was told about the alleged results of the polygraph examination, which Dr. Scherr described as the "inflection point."
However, Dr. Scherr also admitted that there was nothing in the record or information that indicated "an intensity of an approach or a question, or a tactic" by Investigators Castaldo and Quick, and that the proposition that investigators do so when a suspect repeatedly denies involvement is a general research finding. Dr. Scherr also acknowledged that there was nothing in the record to suggest that Defendant was physically isolated or that he sought the company of those who would be supportive of him.
While Courts have recognized that a "lengthy interrogation" is a condition that "suggest[s] that [a] defendant could have been induced to confess falsely" (emphasis added) ( People v. Evans , 141 A.D.3d at 125, 32 N.Y.S.3d 119, citing People v. Days , 131 A.D.3d at 979-980, 15 N.Y.S.3d 823 ), in this case, no evidence was presented that the length of interrogation resulted in an intensity of an approach, the questioning, or a tactic by Investigators Castaldo and/or Quick. Therefore, because no evidence or circumstances were presented to the Court linking the length of Defendant's interrogation with the alleged elicitation of a false confession from him, "any expert testimony on that score would be speculative and delivered in a factual vacuum" ( People v. Powell , 37 N.Y.3d at 492-493, 162 N.Y.S.3d 235, 182 N.E.3d 1028 ; see People v. Ignatyev , 147 A.D.3d at 491, 47 N.Y.S.3d 285 ) and is, thus, inadmissible.
Error Correction Tactic
Dr. Scherr indicated in his report that "[s]ometimes errors are intentionally inserted into suspects’ and witnesses’ written and recorded confession statements [and] [i]n the face of no electronic recording, the investigators may have thought this to be a sufficient substitute to demonstrate voluntariness and credibility." Dr. Scherr testified that this approach or "trick" is "used to bolster the credibility and voluntariness of those suspects giving these statements and offering this information."
Dr. Scherr testified that upon reviewing the statement attributed to Defendant, he observed that there were errors that were crossed out and initialed, and that such were also present in the statements of alleged witnesses Adam Wilson and William MacGregor. While Dr. Scherr acknowledged that the striking and initialing of actual errors is a proper technique when they are present in a statement, he also warned that a risk of impropriety arises from the frequency that this technique is used among different people. Dr. Scherr further testified that during Investigator Castaldo's deposition, he indicated that some of the errors in Defendant's statement were "errors put in there and actually crossed out by Mr. Quick, who would then point to them and have the witnesses or the person of interest initial as a correction."
However, as the error correction technique, when improperly used, bolsters the purported credibility and voluntariness of a confession but does not induce a false one, Dr. Scherr cannot testify that it is a risk factor that can lead to false confessions.
Notwithstanding the above and because the issue of the statement's voluntariness is a factual issue to be determined by the jury, if there is evidence presented that Investigators Castaldo and/or Quick directed Defendant or the other witnesses to strike and initial portions of their statements where no errors existed, the Court will entertain an application to allow Dr. Scherr to opine on the underlying reason to do so.
Failure to Corroborate
According to Dr. Scherr, there was a failure to corroborate the information that Defendant allegedly provided in his statement. For example, Dr. Scherr testified that the Reid Manuel references a variety of ways to corroborate information such as "go[ing] out and assess[ing] whether that was true information," and "try[ing] to find corroborating evidence that it was genuine."
However, Dr. Scherr acknowledged that corroboration is a follow-up activity that the investigators should have undertaken after Defendant's alleged statement and that, as an after-the-fact event, the failure to do so does not induce false confessions.
Accordingly, as a failure to corroborate information that was allegedly provided does not induce false confessions, Dr. Scherr cannot testify that it is a risk factor that can lead to such.
Failure to record
According to Dr. Scherr, the failure to record can induce a false confession because research has shown that when interrogators know that they are not, or believe that they are not being recorded, they engage in more confrontational tactics. As a result, the people being questioned feel and report experiencing more pressure and, since there is not a precise and objective memorialization of the suspect's account, the reliability of such is based upon the police reports and the statement allegedly taken by the police. Thus, Dr. Scherr concluded that there is an underreporting of confrontation and investigators are more likely to use this technique and understate the importance of it.
Dr. Scherr referenced the following field studies: "Does video recording alter the behavior of police during interrogation? A mock crime-and-investigation study." (Kassin, 2014); "Police reports of mock suspect interrogations: A test of accuracy and perception." (Kassin, 2017); "Does video recording inhibit crime suspects? Evidence from a fully randomized field experiment." (Kassin et al., 2019).
While Courts have acknowledged that "the fact that no one had videotaped the hours of the interrogation that had been conducted before the [alleged] confession was made raises significant concerns," (see People v. Evans , 141 A.D.3d at 125, 32 N.Y.S.3d 119, quoting People v. Days , 131 A.D.3d at 981, 15 N.Y.S.3d 823 ), the Court of Appeals has held that " ‘the neglect to record is not a factor or circumstance that might induce a false confession’ " ( People v. Powell , 37 N.Y.3d at 493, 162 N.Y.S.3d 235, 182 N.E.3d 1028, quoting People v. Bedessie , 19 N.Y.3d at 158, 947 N.Y.S.2d 357, 970 N.E.2d 380 ). Accordingly, Dr. Scherr cannot testify that the failure to record is a risk factor that can lead to a false confession.
Misinformation
According to Dr. Scherr, misinformation is fabricated or artificial "information evidence" that is presented to the suspect and that can influence their behavior, beliefs and perceptions. Dr. Scherr testified that misinformation confuses a suspect, especially an "innocent" suspect, about their alleged involvement in a case. For example, Dr. Scherr cited to research which demonstrated that innocent people who were shown fabricated surveillance video suggesting their involvement in a crime were more willing to confess and admit their involvement.
According to Dr. Scherr, prior to making his alleged statement, Defendant was misinformed by Investigator Stephens that he had failed the polygraph examination, and that he had been implicated by three other individuals in the commission of the crime.
As to the polygraph, Dr. Scherr testified that Defendant very much believed in the "diagnosticity" of the polygraph examination and its ability to demonstrate his innocence. In support of that argument, Dr. Scherr pointed to Defendant's responses to Questions 1 and 2 of his post confession interview. In response to Question 1, which asks, "Why did you even come here to be examined," Defendant stated, "I wanted to prove I didn't do anything." When asked in Question 2, "After I told you that you didn't have to take the test, why did you continue," Defendant responded, "I didn't think it would prove that I was lying." Thus, Dr. Scherr concluded that the polygraph environment was a factor that increased Defendant's willingness and risk to falsely confess.
According to Dr. Scherr, data from the National Registry of Exonerations shows that in cases involving a false confession, fourteen percent of those were also accompanied by a polygraph examination.
While Dr. Scherr acknowledged that New York courts prohibit the admission of polygraph examinations and their results, he still considered the report of polygrapher Mark Handler, who reviewed the polygraph administration and determined that Defendant had "actually passed" the polygraph examination, in reaching his conclusion. Due to the recognized unreliability of polygraph examinations and their results, Dr. Scherr opined that any information provided to a suspect regarding their performance on a polygraph examination would be misinformation.
By Decision and Order (Zuckerman, J.) dated December 8, 2021, the Court precluded the proffered trial testimony of Defendant's proposed expert witness, Mark Handler, regarding polygraph examinations.
Dr. Scherr further testified that after his initial report, he received a "memo" provided and submitted by the People, in which it became known that there was an expert aside from Mr. Handler who had noted that Defendant had "passed" the polygraph examination.
Polygraph examinations and their results are inadmissible during trials as they are not considered competent or reliable (see People v. Angelo , 88 N.Y.2d 217, 223, 644 N.Y.S.2d 460, 666 N.E.2d 1333 [1996] ; People v. Shedrick, 66 N.Y.2d 1015, 1018, 499 N.Y.S.2d 388, 489 N.E.2d 1290 [1985], rearg. denied 67 N.Y.2d 758, 500 N.Y.S.2d 1028, 490 N.E.2d 1234 [1986] ; People v. Brewer , 196 A.D.3d 1172, 1174, 151 N.Y.S.3d 784 [4th Dept. 2021], lv. denied 37 N.Y.3d 1095, 156 N.Y.S.3d 782, 178 N.E.3d 429 [2021] ; People v. Rodriguez , 135 A.D.3d 1181, 1184, 23 N.Y.S.3d 692 [3d Dept. 2016], lv. denied 28 N.Y.3d 936, 40 N.Y.S.3d 364, 63 N.E.3d 84 [2016] ; People v. De Lorenzo , 45 A.D.3d 1402, 1402, 845 N.Y.S.2d 652 [4th Dept. 2007], lv. denied 10 N.Y.3d 763, 854 N.Y.S.2d 325, 883 N.E.2d 1260 [2008] ; People v. Weber , 40 A.D.3d 1267, 1267, 836 N.Y.S.2d 327 [3d Dept. 2007], lv. denied 9 N.Y.3d 927, 844 N.Y.S.2d 182, 875 N.E.2d 901 [2007] ).
Moreover, in previous Decisions and Orders, the Court held that Defendant could not introduce evidence regarding the administration of the polygraph examination (see Decisions and Orders (Zuckerman, J.) dated April 16, 2021 and December 8, 2021). Judge Zuckerman further held that Defendant could only question Investigator Stephens as to whether a polygraph examination was administered and whether he told Defendant he failed the examination, but not about how the polygraph was administered (see Decision and Order (Zuckerman, J.) dated April 16, 2021).
The Court will provide a limiting instruction to the jury that the statement of Investigator Stephens that Defendant failed the polygraph examination may not be considered on the issue of guilt or innocence but only on the issue of the voluntariness of Defendant's alleged confession.
Therefore, because polygraph examinations and their results are not considered competent or reliable, Dr. Scherr may not characterize the statement of Investigator Stephens to Defendant that he had failed the polygraph examination as misinformation.
However, if advising a suspect of the result of a polygraph examination, by itself, is a recognized factor or circumstance for inducing false confessions, then the Court will entertain an application to allow such testimony.
Dr. Scherr also noted in his report that Defendant "was confronted with misinformation that others, including the co-defendant and other friends, told investigators that Krivak had sex with the victim and Krivak was the last person to be seen with the victim." Dr. Scherr further noted that, "[e]qually important, Krivak was confronted with misinformation that both Adam Wilson and William MacGregor indicated that they were present during the crime and implicated Krivak." According to Dr. Scherr, at the time Defendant was given this information by the investigators, it was known that Wilson had already recanted.
Because providing misinformation is a recognized factor that can influence a false confession, Dr. Scherr will be permitted to testify regarding any information that was provided to Defendant prior to his statement, which the investigators knew at that time was false or had been recanted.
Conclusion
For the foregoing reasons, the People's motion to preclude Dr. Scherr is granted as limited to the above.
DR. ROBERT A. LEONARD
The Law
" ‘The admissibility of expert testimony is dependent on whether [it] would help to clarify an issue calling for professional or technical knowledge, possessed by the expert and beyond the ken of the typical juror’ " ( People v. Weinstein , 207 A.D.3d 33, 58, 170 N.Y.S.3d 33 [1st Dept. 2022], quoting People v. Williams , 20 N.Y.3d 579, 583-584, 964 N.Y.S.2d 483, 987 N.E.2d 260 [2013] ; see People v. Taylor , 75 N.Y.2d 277, 288, 552 N.Y.S.2d 883, 552 N.E.2d 131 [1990] ). "It is for the trial court in the first instance to determine when jurors are able to draw conclusions from the evidence based on their day-to-day experience, their common observation and their knowledge, and when they would be benefited by the specialized knowledge of an expert witness" ( People v. Powell , 37 N.Y.3d at 490, 162 N.Y.S.3d 235, 182 N.E.3d 1028, quoting People v. Lee , 96 N.Y.2d at 162, 726 N.Y.S.2d 361, 750 N.E.2d 63 ; see People v. LeGrand , 8 N.Y.3d at 455, 835 N.Y.S.2d 523, 867 N.E.2d 374 ; People v. Taylor , 75 N.Y.2d at 288, 552 N.Y.S.2d 883, 552 N.E.2d 131 ; People v. Cronin, 60 N.Y.2d 430, 433, 470 N.Y.S.2d 110, 458 N.E.2d 351 [1983] ; People v. Godallah , 132 A.D.3d at 1150, 19 N.Y.S.3d 119 ; People v. Harris , 74 A.D.3d 984, 985, 902 N.Y.S.2d 190 [2d Dept. 2010], lv. denied 15 N.Y.3d 920, 913 N.Y.S.2d 647, 939 N.E.2d 813 [2010] ). This determination has been "characterized as an evaluation of ‘whether the testimony would aid a lay jury in reaching a verdict’ " ( People v. Powell , 37 N.Y.3d at 490, 162 N.Y.S.3d 235, 182 N.E.3d 1028, quoting People v. Lee , 96 N.Y.2d at 162, 726 N.Y.S.2d 361, 750 N.E.2d 63 ; see People v. Anderson , 36 N.Y.3d 1109, 1111, 144 N.Y.S.3d 678, 168 N.E.3d 851 [2021] ; People v. Bedessie , 19 N.Y.3d at 156, 947 N.Y.S.2d 357, 970 N.E.2d 380 ; People v. Evans , 141 A.D.3d at 124, 32 N.Y.S.3d 119 ; People v. Days , 131 A.D.3d at 978, 15 N.Y.S.3d 823 ).
While "an expert's testimony, by its very nature, always to ‘some degree invades the jury's province,’ and this circumstance alone is not an adequate basis for rejecting expert testimony" ( People v. Bedessie , 19 N.Y.3d at 157, 947 N.Y.S.2d 357, 970 N.E.2d 380, quoting People v. Lee , 96 N.Y.2d at 162, 726 N.Y.S.2d 361, 750 N.E.2d 63 ; see People v. Days , 131 A.D.3d at 978, 15 N.Y.S.3d 823), the jury may not be displaced of its fact-finding function by expert testimony where there is no "reason to suppose that such testimony will elucidate some material aspect of the case that would otherwise resist comprehension by jurors of ordinary training and intelligence" ( People v. Inoa , 25 N.Y.3d 466, 472, 13 N.Y.S.3d 329, 34 N.E.3d 839 [2015] ; see People v. Cronin , 60 N.Y.2d at 432, 470 N.Y.S.2d 110, 458 N.E.2d 351 ).
Testimony of Dr. Leonard
According to Dr. Leonard, linguistics is the science of language analysis. Dr. Leonard testified that he was retained by the defense to analyze the language of a police statement purportedly dictated by Defendant and then transcribed "verbatim" by Investigator Quick. Dr. Leonard was also asked to compare the language of other purported witnesses’ statements, which were alleged by the prosecution to be a verbatim record of what was dictated by each individual and written down by Investigator Quick. Finally, Dr. Leonard was asked to determine whether the language contained in Defendant's purported confession is consistent with a verbatim record of spoken, spontaneous language produced by the defendant himself, as claimed by law enforcement. In connection therewith, Dr. Leonard produced a report dated December 15, 2021, and an amended report on June 9, 2022.
A PowerPoint presentation of ninety-eight (98) slides was utilized as a demonstrative aid during Dr. Leonard's testimony.
The amended report was in response to Dr. Schilling's report dated May 18, 2022.
In his report, Dr. Leonard developed two competing hypotheses. First, that the language patterns in the purported confession are consistent with a verbatim transcription of spoken, spontaneous language produced by Defendant. Second, that that the language patterns in the purported confession are not consistent with spoken language, spontaneous language, or the language patterns of Defendant.
According to Dr. Leonard, his analysis centered around the use of the word "verbatim," because Investigator Quick characterized the statement attributed to Defendant as being such during the Huntley hearing. Dr. Leonard testified that according to dictionaries and to databases of use, the word "verbatim" means, "word for word, precise, every word from source becomes verbatim, those exact words, nothing more, nothing less." Dr. Leonard also testified that the police claimed that the statement attributed to Defendant was "spontaneous spoken language." Dr. Leonard referred to "spontaneous speech" in his report as "a type of natural and authentic speech produced without any external influence, out of a momentary impulse of the speaker, in contrast to speech that is, e.g., scripted or rehearsed." Examples of spontaneous speech referred to by Dr. Leonard are "fillers," such as "um," or "discourse markers," such as "so," "you know," "right," and "like."
Dr. Leonard testified that, while he "did not exhaustively look at all the instances of verbatim," he did look in the "one billion word Corpus of Contemporary American English," which does contain this definition.
In this case, the Court found that there were external influences, including the police environment and the questions posed by Investigators Castaldo and Quick.
To resolve the competing hypotheses, Dr. Leonard analyzed the 1996 police statements of Defendant, Adam Wilson, William MacGregor, and Denise Rose. Dr. Leonard also reviewed, "1) Rose (3), a third police statement purportedly produced by Denise Rose on April 10, 1996; 2) Rose's courtroom testimony from the years 1997, 2012, and 2016; 3) four letters known to be authored by Andrew Krivak from the years 1998, 1999 (1); 1999 (2) and 2000; 4) the 1996 pre-trial courtroom testimonies of Investigator Quick, Investigator Castaldo, and Investigator Stephens and; 5) the 2019 deposition testimonies of Investigator Quick and Investigator Castaldo."
All four letters were written by Defendant to his sister while he was incarcerated.
Dr. Leonard testified that he used two modes of analysis to test his hypotheses. First, authorship analysis, which seeks "to find matches or mismatches of patterns in a questioned document and in documents of known authorship." Second, narrative analysis, which he described in his report as "a long-standing area of intensive study within the science of linguistics," which "allows for an accurate comparison of structural elements across narratives produced by different speakers as well as of retellings of a story produced by a single speaker on different occasions."
The conclusion reached by Dr. Leonard to a reasonable degree of scientific linguistic certainty was that "the language patterns in the purported confession are better explained as not being consistent with a verbatim transcription of spoken language and the language patterns of Andrew Krivak, [and that] similar instances of language use found in Krivak's purported confession are also found in other statements which are purported to have been independently produced by other parties." The similar instances of language use found by Dr. Leonard were characterized as "similar narrative structures," "n-grams and near n-grams across narrative episodes," "lack of markers of spoken language," "lack of features of spontaneous language," "post-subject position ‘then,’ " "time adjunct ‘At this time,’ " "full names for repeated referents," and "complementizer insertion/deletion."
A considerable amount of time was spent by Dr. Leonard explaining how one expects that different narrators telling their own personal versions of a shared experience will include both similar and different "episodes." According to Dr. Leonard, it is less expected that narratives of personal experience will include virtually the exact same "episodes" in virtually the same order, which is what he submits happened in this case. For example, Dr. Leonard referenced the "time adjunct " ‘At this time,’ " and stated that pursuant to the Corpus of Contemporary American English (COCA), such construction is rare both in written and spoken English. However, Dr. Leonard found that this construction was present in the purported statements of Defendant, Rose and Wilson. Thus, Dr. Leonard submits that his testimony is necessary to explain the "data patterns."
According to Dr. Leonard, in COCA's 127-million-word subcorpus of spoken language, this "n-gram" appears about 2,800 times.
The Court disagrees. Since Rose, Wilson and MacGregor, the very people upon whose statements Dr. Leonard's comparisons were based, are scheduled to testify in this matter, the defendant can establish through direct and/or cross examination whether the written statements they provided were verbatim. The Court further finds that Dr. Leonard's testimony is not necessary because the jury can compare the written statements allegedly provided in this case and make its own determination, as the triers of fact, whether they include virtually the exact same "episodes" in virtually the same order.
The case of Ragbir v. United States , 2019 WL 9522419 [D.N.J. 2019], in which "Dr. Leonard's report analyzed whether the written confession was ‘a verbatim transcript of [Defendant's] exact words’ to the police," is factually very similar to this matter. In Ragbir , like here, Dr. Leonard compared the defendant's alleged confession with a source of his known language and noted that the alleged written confession contained no "disfluencies" or "discourse markers," i.e., informal transition words like "well" or "see." While acknowledging that the purported confession was signed by the defendant at the bottom of each page and the possibility that the detective who characterized the statement as "verbatim" in that case might have a slightly different definition of the word than a professor of linguistics, Dr. Leonard reached the same conclusion that he reached in this case, to wit, that the confession was not a "verbatim" transcription of the defendant's words to the police.
The Court in Ragbir noted that because "Dr. Leonard seems to suggest that the written confession does not accurately reflect the way that anyone spontaneously speaks [a] lay juror could easily grasp that proposition without expert help. Everyone knows that unprompted speech contains fillers, pauses, and false starts; anyone could read the confession and see it contains none of these. By the same token, however, lay people would understand why a police officer taking down a report of a confession might not include them."
This Court agrees with the holding in Ragbir and finds that it is within the ken of adult, English-speaking lay jurors to see that the statements allegedly provided by Defendant, Wilson, Rose and MacGregor do not contain the disfluencies, discourse markers, fillers, pauses, and false starts that are naturally present in verbatim statements.
This Court also shares the concern of the Ragbir Court that "[a] jury could feel they were being condescended to or misdirected from the pertinent issue (whether the confession was accurate) to one less pertinent (whether it was verbatim)." Moreover, the Court finds that the terms used by Dr. Leonard to explain his methodology and conclusion, such as "n-grams," "near n-grams," "post-subject position ‘then,’ " "time adjunct ‘At this time,’ " and "complementizer insertion/deletion," would confuse, rather than help the jury to understand the issue at hand.
During the hearing, Dr. Leonard testified considerably about "near N-Grams," which he defined as "an almost identical series of words found in a string," and concluded could be seen across the narrative episodes of the defendant and the witnesses at issue. As an example, Dr. Leonard testified that the partial phrase "it was about 6:30" that is contained in Rose's first statement, and the partial phrase "at about 6:00 pm" that is contained in her second statement, are "near N-grams." These phrases, like the others cited by Dr. Leonard as examples of "near N-grams," can be easily examined and compared by a lay jury in determining whether they are almost identical and what, if any, weight to give that determination in deciding whether Defendant's alleged statement was verbatim. Moreover, there is nothing scientific about this analysis as it can be replicated by defense counsel through exhibiting and comparing the statements at trial and examining the witnesses who allegedly provided them.
Dr. Leonard devoted ten (10) slides of his PowerPoint presentation to this issue, which consisted of tables comparing the similarities amongst statements attributed to Defendant, Wilson, Rose and MacGregor.
Additionally, the Court finds that the portion of Dr. Leonard's report regarding "near N-grams" is not based upon generally accepted principles and methods. Dr. Schilling testified that a "near N-gram" is not a term that she has encountered in research or publications in the field of linguistics, peer-reviewed research, or published journals, and that she is unaware of any accepted definition for what constitutes a "near N-gram." Dr. Schilling further testified that the term "has no meaning in terms of conducting a linguistics analysis," and the Court credits such testimony. Accordingly, as that portion of Dr. Leonard's opinion which is based upon his comparison of "near N-grams" is not sufficiently related to existing data and its connection to such data is "only by the ipse dixit of the expert" ( People v. Brooks , 31 N.Y.3d at 941, 73 N.Y.S.3d 110, 96 N.E.3d 206, citing General Elec. Co. v. Joiner , 522 U.S. 136, 146, 118 S.Ct. 512, 139 L.Ed.2d 508 [1997] ; see D.H. Pace Co. v. Aaron Overhead Door Atlanta LLC , 526 F. Supp. 3d 1360, 1372 [N.D. Ga. 2021] ); Guerra v. Ditta , 185 A.D.3d 667, 669, 127 N.Y.S.3d 148 [2d Dept. 2020] ), Dr. Leonard's testimony is likewise precluded on this ground.
The only evidence of the use of such term that defense counsel elicited was a single reference within a chapter of a book titled "Challenging the Myth of Monolingual Corpora," that was published by Brill in the Netherlands in 2017, wherein the author stated, "To identify copied strings of words, in a first step n-grams and what could be called ‘near n-grams’ were identified" (emphasis added).
According to Dr. Schilling, Dr. Leonard also violated the principle of accountability, which she testified is a foundational principle and central tenet of sociolinguistics. Dr. Schilling testified that "the idea of [that] principle is that analysts should record all occurrences of a variable feature rather than selecting those variants of a variable that tend to confirm their argument while ignoring others that do not." Dr. Schilling further testified that "[t]he linguist should count both occurrences of features of interest and cases where the feature could have occurred but did not," [since] "[i]n this way, the research[er] can calculate the percentage of occurrence of a particular feature."
In response, Dr. Leonard acknowledged that "Dr. Schilling identifie[d] instances of pre-subject position ‘then’ that were not included in [his] Report," but asserted that "[t]hey were not meant to be included [as his] report aimed simply to demonstrate the mere presence of instances of post-subject ‘then’ constructions; pre-subject ‘then’ constructions are not instances of this."
Dr. Leonard also stated that "Dr. Schilling claims that my analysis ‘fails to include some phrases that differ in wording across statements,’ however, [my report] aimed to include similarities which are unexpected (or deviations from an expected norm), rather than differences."
Lastly, while Dr. Schilling disagreed with the methodology employed by Dr. Leonard, she did agree that the statement attributed to Defendant was not linguistically verbatim and, therefore, there is no active dispute among the parties regarding this issue. According to Dr. Schilling, the linguistic community very likely has a different definition of "verbatim" than the investigators in this case do.
Dr. Leonard also acknowledged that there are different understandings of the word verbatim. For example, Black's Law Dictionary (11th ed. 2019) defines "verbatim" as simply "Word for word," but specifically states that, "Courts have repeatedly held that, in the context of the requirement that a trial record must be ‘verbatim,’ absolute word-for-word accuracy is not necessary - and insubstantial omissions do not make a transcript ‘nonverbatim.’ " Finally, the Court notes that Black's Law Dictionary defines "verbatim ac litteratim " as "Word for word and letter for letter."
Conclusion
Since the determination as to whether the purported statement of Defendant is "verbatim" is within the ken of the typical juror, and the jurors will be able to draw conclusions from the evidence based upon their day-to-day experience, their common observation and their knowledge, the People's motion to preclude the testimony of Dr. Leonard is granted.
The foregoing constitutes the Decision and Order of this Court.