At the trial of indictments for sexual abuse of children, hearsay evidence of statements of one of the alleged victims, made while the child was sleeping ("sleep talk") were not admissible for any purpose, where the statements were not shown to be reliable; where the child had no capacity while asleep to have conscious thought; and where, in any event, the statements were so lacking in rational support that they were not probative of any material fact: in the circumstances, the defendant was entitled to a new trial. [719-721]
COMPLAINT received and sworn to in the Milford Division of the District Court Department on September 16, 1996.
On transfer to the Worcester Division, the case was tried before John S. McCann, J.
The Supreme Judicial Court on its own initiative transferred the case from the Appeals Court.
Meletios D. Chacharone for the defendant.
Sandra P. Wysocki Capplis, Special Assistant District Attorney, for the Commonwealth.
IRELAND, J. On appeal from his convictions of two counts of indecent assault and battery on a child under the age of fourteen years, the defendant raises the question whether out-of-court statements made by a child complainant while sleeping were properly admissible at trial. We transferred the case to this court on our own motion. Because the statements lack the necessary indicia of reliability, we find that their admission constituted reversible error. Accordingly, a new trial is required.
On September 16, 1996, a criminal complaint was filed charging the defendant with six counts of indecent assault and battery
on a child under the age of fourteen years (G. L. c. 265, s. 13B), four counts of assault and battery (G. L. c. 265, s. 13A), and one count of indecent exposure (G. L. c. 272, s. 53). On July 17, 1998, a jury in the District Court acquitted him of all but two charges of indecent assault and battery on a child under the age of fourteen years.
The record reveals the following facts. The defendant, Jorge R. Almeida, and a parent, J.D., were close friends and neighbors for approximately one year. During that time, the defendant and his wife babysat for J.D.'s children, whom we call Erica and Edward. In September, 1996, Erica's friend, whom we call Amy, stayed with Erica and her family for a few days. On September 7, 1996, Erica and Amy, both ten years old at the time, went to play with the defendant's rabbits. After allowing the girls into the shed where he kept the rabbits, the defendant followed, closed the door, and began to touch the girls' chests and Amy's vaginal area [Note 1]. The girls ran from the shed to Erica's porch, where they encountered two boys from their neighborhood. The girls told them that the defendant "was feeling them up." One of the boys claimed to have witnessed Jorge "attempting to touch . . . one of the girls," and informed Erica's father who, in turn, asked the girls what had happened. The girls repeated that the defendant had "felt them up." That night, Amy slept at J.D.'s house. In the middle of the night, J.D. awoke and heard Amy say in her sleep, "Jorge, get off me. Jorge, get off me."
At trial, Amy, Erica, and J.D. testified. In his capacity as a fresh complaint witness, J.D. testified to his conversation with the girls on the afternoon of the incident and the contents of Amy's "sleep talk." The defendant objected and, at sidebar, argued that the sleep talk was prejudicial and irrelevant. The prosecutor argued for the admission of the statements as follows: "[First, t]he witness would be unavailable because she would be asleep. . . . [N]obody is going to remember what they say in their sleep. Secondly, it is probable, [sic] what
happened. It is common that children act out like that and [will] be talking in their sleep and be showing the fear. It bolsters . . . it's post known in a fresh complaint." The judge admitted the statements in evidence but did not explain his grounds for doing so. It is difficult to glean from this exchange the specific grounds of admissibility argued to the judge. This uncertainty is compounded by the absence of a contemporaneous limiting instruction alerting the jury of the evidentiary purpose of the proffered sleep talk.
Admitting hearsay evidence of statements made while a person is sleeping, so-called "sleep talk," would run counter to one of the central principles governing the admissibility of evidence, namely, that the proffered material is reliable. See P.J. Liacos, Massachusetts Evidence s. 8.4.1, at 477 (7th ed. 1999) (exception to hearsay based on "a guarantee of trustworthiness in the circumstances surrounding the making of the particular declaration for which an exception is created.") Here, the circumstances surrounding Amy's declaration plainly lack the requisite degree of trustworthiness [Note 2]. See Commonwealth v. Rosario, 430 Mass. 505 , 508 n.3 (1999), quoting Commonwealth v. Trigones, 397 Mass. 633 , 637 (1986) (proffered hearsay statements subject to greater scrutiny in criminal trials; admissible "only if the statements bear 'adequate indicia of reliability'"); Duchanne v. Hyundai Motor America, 45 Mass. App. Ct. 401 , 407 (1998) (judge properly excluded "purely subjective opinion. . . based on nothing more than conjecture, surmise, and speculation"); Commonwealth v. Burgos, 36 Mass. App. Ct. 903 , 903 (1994) (proper to exclude statement where "corroborative foundation was so shaky that the hearsay was not trustworthy").
As the Commonwealth concedes, Amy cannot recall having made the statements because she did so without the filter of conscious thought. Common sense dictates that while asleep, one cannot distinguish between reality and fiction, a touchstone of admissibility [Note 3]. Cf. Commonwealth v. Murphy, 48 Mass. App. Ct. 143 , 145 (1999) ("Lawyers who present a child as a witness must cope with the threshold problem of establishing that the child is able to differentiate fact from fiction . . ."); Commonwealth v. O'Brien, 35 Mass. App. Ct. 827 , 828-829 (1994), citing Commonwealth v. Brusgulis, 398 Mass. 325 , 329-330 (1986) (judge must inquire into child witness's capacity to understand difference between truth and falsehood). Given that Amy did not have the capacity to evaluate the veracity of her statements, we cannot permit a fact finder to consider them for any purpose.
The statements also lacked the necessary foundation to be probative of any material fact in the case. See Commonwealth v. Woods, 414 Mass. 343 , 355, cert. denied, 510 U.S. 815 (1993) (upholding exclusion of evidence of limited probative value); Boston Edison Co. v. Assessors of Watertown, 387 Mass. 298 , 308 (1982) ("testimony [may be] so lacking in rational support that it has no probative force"). That Amy was having some form of dream about the defendant, the contents of which are entirely unknown, sheds no light on whether she had actually been sexually assaulted by him. Furthermore, the prejudicial nature of such evidence far outweighs its minimal probative value. See Commonwealth v. Martinez, 431 Mass. 168 , 174 (2000) (where prejudicial effect outweighed probative value, admission of threat testimony was error, but not prejudicial). The suggestion that Amy was recounting the afternoon's incident in her sleep would evoke strong sympathy for the child, while offering nothing of genuine probative value. See Lally v.
Volkswagen Aktiengesellschaft, 45 Mass. App. Ct. 317 , 326 n.14 (1998) ("Cases involving seriously injured children present perhaps the greatest risk that jurors will improperly be influenced by emotion and sympathy"). We also note that the defendant was convicted only on the charges stemming from the day that the father said he heard her talking in her sleep, and that the jury did not find the children's allegations of other incidents, which were not supported by any sleep talk, sufficiently credible to sustain a guilty verdict. Thus, we are not "sure that the error did not influence the jury, or had but very slight effect." Commonwealth v. Vinnie, 428 Mass. 161 , 163, cert. denied, 525 U.S. 1007 (1998), quoting Commonwealth v. Flebotte, 417 Mass. 348 , 353 (1994). Given this context, the judge's decision to admit the sleep talk was an abuse of discretion and a new trial is warranted [Note 4].
Accordingly, the judgments are reversed, the verdicts set aside, and the case is remanded for a new trial.
[Note 1] Both girls testified that there had been several incidents of molestation throughout the year. The defendant was acquitted of all charges stemming from these alleged earlier incidents. The two convictions were based on the alleged September 7, 1996, assaults.
[Note 2] Among the States that have addressed the admissibility of statements made while a person is sleeping, there is no uniformity with respect to either outcome or analytical approach. A number of States have found "sleep talk" evidence inadmissible due to its lack of reliability. See, e.g., State v. Zimmerman, 121 Idaho 971, 976 (1992) (sleep talk inadmissible; not relevant due to lack of probative value regarding actual events and inherent unreliability); State v. Presley, 108 Or. App. 149, 152 (1991) (relevance of sleep talk not established because of failure to demonstrate nexus between child's vocalization and alleged incidents).
[Note 3] In State v. Posten, 302 N.W.2d 638, 641 (Minn. 1981), the court reasoned that the real issue was trustworthiness and that, in that instance, it was not dealing with a conniving person but rather with a child who had obviously suffered. That court qualified the admission of the sleep talk by noting that it was not the only evidence against the defendant and that it was being used primarily for corroborative purposes. We disagree as to the trustworthiness of sleep statements. In addition, we do not believe the sleep talk at issue here was used for corroborative purposes.
[Note 4] We have reviewed the other evidentiary bases advanced by the Commonwealth (e.g., spontaneous utterance, fresh complaint, nonhearsay) and find they are without merit.