SJC hears dispute over 92a accommodations after trial judge found witness testimony flawed

The Supreme Judicial Court on appeal heard argument in Commonwealth v. Sammy D. Lozada over whether a trial judge’s finding of noncompliance with “92a paragraph 3” — and related concerns about the accuracy of a deaf, linguistically deprived witness’s testimony — requires a new trial or was waived by defense counsel.

The question matters because the witness, identified in the trial record as Maria Samat, communicates by visual gestural systems rather than full American Sign Language, and attorneys and experts who later reviewed the trial record told the court that ordinary sign-language interpreters and courtroom procedures may have been inadequate.

Travis Lynch, attorney for the Commonwealth, told the justices he wanted to “cut right to the prejudice issue” and quoted the trial judge’s written finding that “while unlikely it’s possible that if this statute had been complied with … the witness would have been forbidden from testifying.” Lynch argued the judge had credited an interpreter’s testimony that she was “99% competent” and that the record did not show that the trial interpretation was actually inaccurate.

Tony Fellows, an attorney who argued for the party raising the complaint about the testimonial accommodations, described Samat as “severely language deprived” and said she effectively has “a bag of words,” meaning a limited set of gestures and signs developed with caregivers that can mean many different things depending on context. Fellows and his experts told the court that accurate communication for someone with Samat’s limitations requires visual props — figurines, color charts, calendars or models — and specialized practitioners in “visual gestural interpretation.”

Fellows urged the court to consider that appellate review is handicapped by a record that is largely verbal: “every deaf person who testifies in court, their testimony should be video recorded as well as audio recorded,” he said, relaying the view of experts who worked on the case.

Several justices and counsel debated whether the trial judge framed the issue as an admissibility problem — concluding the witness should not have been allowed to testify because the 92a procedure was not followed — or whether the issue was instead one of ineffective assistance of counsel and waiver because no timely objection was made at trial. One commentator during argument observed that the judge both found a statutory violation and, separately, ineffective assistance of counsel.

Counsel and the court also discussed a practical difficulty: visual-gestural interpreters of the sort the experts described are rare, lack a formal certification process, and typically develop communication with a particular individual over extended time. Fellows said his experts “are not certified because there’s no such thing,” and described experienced visual-gestural practitioners as those who have worked with language-deprived people for months or years.

The justices probed whether a fuller pretrial or evidentiary hearing using the right tools and interpreters could have yielded more reliable testimony without excluding the witness entirely, and whether the trial judge erred by treating the matter as one of automatic inadmissibility when, the Commonwealth argued, a more tailored accommodation plan might have preserved admissible evidence. Lynch acknowledged the judge’s concern but emphasized that the judge credited the interpreter who testified and that the record did not show that the interpreter’s work was demonstrably inaccurate.

No final ruling was issued at oral argument. The court’s decision will determine whether the trial judge’s rulings — including the finding of statutory noncompliance and any order of a new trial tied to ineffective assistance concerns — are upheld or reversed and will clarify procedural and remedial expectations when visually based communicative systems must be accommodated in criminal proceedings.

For appellate courts and trial judges, the case spotlights three recurring issues: (1) how to evaluate testimonial accuracy when the underlying communication is visual and not captured in a standard transcript; (2) what procedures trial courts must follow under 92a paragraph 3 when a witness is deaf or linguistically deprived; and (3) how courts should handle the practical scarcity and lack of formal certification for experienced visual-gestural interpreters. The SJC’s ruling is likely to shape whether trial courts must do more to preserve a record (including recommending audiovisual recording) and how much deference appellate courts should give trial judges who observed the testimony in real time.