The issue is whether the People met their burden to show by a preponderance of the evidence that defendant was competent to stand trial. It is an issue of a kind that ordinarily implicates a mixture of factual and legal findings in its determination, and, accordingly, is rarely jurisdictionally appropriate for our review. This, however, is a most unusual case.
On September 22, 2004, defendant, convinced that his wife had been unfaithful to him and heedless of previously issued orders of protection obtained by her against him, accosted her in the lobby of her apartment building and stabbed her repeatedly. The incident was recorded by a surveillance video camera and defendant’s commission of the stabbing has never been disputed.
Based on the stabbing and other related incidents of alleged spousal abuse committed in violation of outstanding orders of *520protection, defendant was indicted for numerous crimes, most notably attempted murder, and, as an incident of the prosecution thus commenced, defendant was, pursuant to CPL 730.30 (1), ordered examined to determine whether he was fit to answer the charges against him. He was thereupon admitted to Bellevue Hospital and examined there in November and December of 2004 by two qualified psychiatrists and other health care professionals. His attorney was present during one evaluative session and his interaction with her was observed by the doctors.
Both psychiatrists concluded that defendant was unfit. They noted that defendant had a history of microvascular disease, and, some six years before, had suffered a series of strokes leaving him with extensive brain damage. CT and MRI scans performed during the Bellevue hospitalization disclosed a large lesion on the left side of defendant’s brain and defendant exhibited halting and dysarthric speech, poor concentration, poor short-term memory and significant right-side weakness, all of which the doctors understood to be attributable to the lesion. Psychological testing, also performed during the Bellevue hospitalization, confirmed the existence of significant cognitive disabilities consistent with defendant’s neurologic history. Defendant scored in the impaired range on both of the standard neurologic batteries administered to him: the testing showed significant impairment in defendant’s memory (both immediate and delayed), attention, and abstract reasoning, and notable, although less severe, impairment of his executive functioning. Although he demonstrated some superficial familiarity with the roles of trial participants, his evident inability to think abstractly, even on a basic level, was found to preclude him from assisting his attorney in his defense. This conclusion was reached after both doctors had observed defendant’s lawyer attempt to discuss with him his legal options.1 Dr. Manuel Lopez-Leon wrote,
“[d]uring this interview [the interview conducted with defendant’s attorney present] it became clear *521that Mr. Phillips concrete reasoning, speech impediments, and other symptoms of dementia severely impaired his ability to understand and rationally manipulate information pertaining to his case. He was not able to retain and repeat information pertinent to potential legal strategies, and did not grasp the magnitude of the consequences derived from those strategies.”
And, the second examiner, Dr. Doonam Kim, similarly noted
“Mr. Phillips demonstrated much difficulty in using information provided by his lawyer to appropriately discuss legal options in regards to his case . . . Even when discussing issues unrelated to his legal case, he was unable to answer questions without difficulty or an extreme amount of effort given his physical and cognitive limitations.”
In receipt of these competency assessments, Justice Soloff adjudged defendant an incapacitated person within the meaning of CPL 730.10 (1), i.e., “a defendant who as a result of mental disease or defect lacks capacity to understand the proceedings against him or to assist in his own defense,” and in January of 2005 remanded him to the custody of the Commissioner of Mental Health pursuant to CPL 730.50 (1).
Some six months later, defendant’s “restoration of fitness” treatment team at Kirby Forensic Psychiatric Center announced that defendant was now fit to be tried. The treatment team’s conclusion as to the efficacy of its restorative efforts, however, was not concurred in by the Hospital Forensics Committee,2 which, in accordance with the evaluation and recommendation of committee member Daniel Capruso, Ph.D., a diplómate in clinical neuropsychology, unanimously found that defendant remained unfit.
In his neuropsychological assessment and accompanying competency evaluation, and in his ensuing testimony at defendant’s competency hearing, Capruso stated that defendant suffered from transcortical motor aphasia, a diagnosis he had carried since his series of strokes in the late 1990s. This was a *522permanent and irreversible condition following from and neurologically correlating with his stroke-induced brain damage. Capruso reviewed in considerable detail defendant’s MRI brain scans, pointing out the damaged areas of defendant’s left, and, to a lesser extent, right brain. He explained that the left hemisphere lesion, ventricular expansion and concomitant loss of brain mass could be expected to compromise brain function, particularly with respect to the transmission of impulses necessary to the use of language. Citing an authoritative text, Capruso stated that the vast majority of those who suffered left hemisphere injuries such as defendant’s were left with language comprehension deficits. That defendant was not a fortunate exception was borne out in Capruso’s neuropsychological testing. That testing, employing standard batteries scaled to account for social and educational factors, disclosed extraordinary language deficits. In fluency, naming, repetition and comprehension defendant, then 59 years of age, performed below the level of a six year old. The “problem” said Capruso, “is generating and understanding words, language.” Although defendant was capable of responding to simple inquiries, he was demonstrably unable to comprehend or reliably respond to even slightly more complex statements and/or questions. In his neuropsychological evaluation report Capruso wrote:
“Although it is true that the patient is capable of comprehending simple words and phrases, he can not reliably comprehend and execute complex commands. Thus while the patient can understand relatively simple legal concepts, such as the charges against him, legal information of slightly greater complexity is beyond his comprehension. He would be unable to adequately follow the discourse in a typical court proceeding, which often involves complex questions, elaborated answers, and objections from opposing attorneys. Although his visual memory may be adequate under some circumstances, his verbal memory is deficient, and he remembers less than half of what he has been told, even when he needs only to recognize the material, instead of having to recall it himself. His verbal fluency is so limited that he could not possibly take the stand and testify in his own defense.”
*523In his CPL article 730 fitness evaluation,3 Dr. Capruso noted, in response to the specifically framed inquiries on the standard form, that defendant by reason of his aphasia suffered from moderate impairments of perception and memory, and severe impairment of his ability to think and communicate. While Capruso detected that defendant had some rudimentary understanding of what happened at a trial, he found, as had the Bellevue evaluators, that defendant’s ability to establish a working relationship with an attorney was severely impaired, as was his ability to listen to the advice of counsel and consider the risks and benefits of alternative courses of legal action. When Capruso asked defendant how he could help his lawyer defend him, defendant replied, “[t]he doctor put a wire in my heart.”
In light of the sharply differing recommendations of the treatment team and the Hospital Forensics Committee, Dr. James Hicks, a board-certified psychiatrist and the Associate Clinical Director of Kirby Forensic Psychiatric Center, was called upon to review and resolve the matter.4 Hicks found defendant fit. In his CPL article 730 report to the court and subsquent hearing testimony he acknowledged that defendant had a large left hemisphere brain lesion, which he referred to as a “big hole,” and that his condition could not improve and indeed could only worsen.5 He, however, was of the view that defendant’s cognitive ability—at least as much of it as was necessary to comprehend trial proceedings—had not been significantly impaired. He noted that he had, by asking defendant yes-or-no questions or questions requiring only simple one-word answers, been able to gather that defendant understood what he had been accused of; that he was aware that a jury might have “mixed feelings”6 about his behavior towards his wife; that he understood the *524roles of the trial participants; that, if convicted, he faced a sentence of 25 years (which could well mean that he would die in prison); and, that he could receive a lesser sentence by entering a plea.
Hicks, however, did acknowledge significant impairment of defendant’s ability to express himself and that his aphasia posed an obvious challenge to attorney-client communication. In finding defendant fit, he had apparently assumed that defendant would plead guilty; although he thought that defendant could be tried, he conceded that “[a] trial would be problematic,” and that it would “requiret ] some thought by the people who know the court more than I do in terms of what exactly would be demanded of [defendant].” Hicks admitted during his hearing testimony that he had not previously considered whether defendant would be able to testify in his own behalf. Considering the prospect for the first time, he owned that testifying would be defendant’s “biggest challenge.” Defendant, he observed, could not respond to open-ended questions and, if he testified, would have to be led.
In addition to the testimony and evaluation of Dr. Hicks, the People relied at the competency hearing upon the testimony of Cynthia Scheuer, Ph.D., a psychologist and a member of defendant’s treatment team at Kirby. Scheuer was quite clear that she was not defendant’s therapist—at least not in any commonly understood sense. She never met with defendant individually. Rather, her observations and occasional interaction with him occurred when, once or twice a week, he along with 15 to 20 other patients attended “restoration of fitness” groups, which she led. The purpose of these meetings, explained Scheuer, was to educate unfit defendants respecting criminal justice issues so that they might eventually evince sufficient comprehension of criminal proceedings to be deemed fit and returned to court. Scheuer was quite certain that defendant had no cognitive deficits. She was also sure that he had made significant strides during his Kirby hospitalization and repeatedly noted in his chart that he had improved. She allowed that he had difficulty verbalizing and that he said very little, but insisted, without explanation, that “he just seemed to understand what was going on.” She found it telling that he seemed to respond to her humor.
*525The defense witnesses at the competency hearing, in addition to Dr. Capruso, included Dr. Katherine Henry and defendant himself.
Doctor Henry, the Director of the Department of Neurology at Bellevue, testified that she had interviewed defendant at Bellevue7 and, among other things, had elicited from him several sets of completely inconsistent responses to simple questions, which although differently worded, sought the same information. Based on her clinical findings, and after reviewing defendant’s medical records, his brain scans, and the neuropsychological testing performed by Dr. Capruso, she concurred in the diagnosis of transcortical motor aphasia and agreed with Dr. Capruso that defendant’s aphasia involved not only the compromise of his ability to use language expressively but also signficant impairment of his ability to think and understand. Defendant, she said, was capable of following simple directions and giving simple responses, but could not execute commands involving several steps—even when they were very straightforward—and was incapable of reliably responding to open-ended (i.e., non-yes-or-no) questions. She reviewed his brain scans in considerable detail and explained that the radiologically confirmed damage to his left frontal lobe correlated well to his expressive and cognitive deficiencies; the damage affected areas of the brain through which neural impulses necessary to those faculties had to pass. It was, she noted, “a big lesion” affecting “[eloquence, speech, your ability to communicate, to be verbal, to function in life, as it were, to process information.”
Obliquely addressing Dr. Scheuer’s impressions, Dr. Henry explained that aphasics commonly tried to cooperate and furnish what they believed were socially appropriate responses, and that by means of parroting, facial expressions and short answers they frequently succeeded in producing in inexpert observers a false impression of comprehension. Even in clinical settings, much would be read into their answers and behavior that simply could not be there given the actual level of impairment disclosed in neuropsychological testing. Dr. Henry noted that defendant had, since his debilitating strokes in the late 1990s, been perceived to be significantly more comprehending than he was; *526there had not infrequently been a striking disparity between the clinical impressions of those untrained in neurology or speech pathology and defendant’s expert neuropsychological evaluations and testing, which had consistently disclosed dramatic deficits in memory and comprehension. “Because of the expressive piece,” explained Dr. Henry, “you can’t tell. And if you can’t tell, you can’t tell. So somebody observing a patient in a ward will fill in the gaps if they need to. I mean, you’ll take it at face value . . . Even a professional because it’s natural to do that.” Of course, defendant did understand some things, but the difficulty was that it was often very hard to ascertain precisely what he did and did not grasp. “The trouble,” observed Dr. Henry, “is, you don’t know where his capability would stop and his inability would step in ... It would always be an assumption on somebody’s part . . . [T]he damage to his brain is going to keep him from processing things coming in and expressing things coming out.” Dr. Henry was of the opinion that defendant was not capable of informed consent in the medical setting and that his deficits would make his participation in his defense impracticable: “I don’t think his working memory is working well enough to hear a piece of information here, store it, bring it back, observe what’s going on and apply it to be able to participate actively in his own defense.” In her view, there was no special accommodation that could compensate for these fundamental cognitive deficiencies.
Defendant’s competency hearing testimony consisted largely of his response to a standard hypothetical evidently frequently used to assess a defendant’s ability to assist in his or her defense. Defendant’s answers, taken at face value, indicated that he did not understand what a guilty plea was or what it entailed,8 and, when he was questioned respecting the hypothetical—a simple scenario involving a fight in a bar—his responses demonstrated, not merely incomprehension, but characteristically paraphasic interpolations of at best irrelevant and at worst potentially incriminating detail. Defendant, for example, indicated that the most important fact to tell his attorney was that his hypothetical victim was bleeding when, in fact, there had been no mention at all of any such circumstance.
At the conclusion of the competency hearing, defendant’s attorney implored the court to find defendant incapacitated, *527noting on the basis of her already lengthy experience representing him that her client’s verbiage was “at best confusing, certainly unreliable, [and] sometimes inculpatory.” She asserted that he could not testify and could not understand the theory upon which a trial defense would have to rest, namely, lack of responsibility by reason of mental defect. She stated, “I am telling you that I believe that Mr. Phillips is incapable of assisting me in his defense. I cannot get him to a point where I feel comfortable where I feel that he is able to assist me.”
The court nonetheless found defendant fit. After a lengthy review of the evidence, it concluded, “[although [defendant] suffers from transcortical motor aphasia, he does not exhibit any severe mental impairment or imbalance in his current mental state.” In reaching this conclusion, the court attached little weight to the neuropsychological testing adduced by the defense and elected to rely most heavily on the observations of defendant by Dr. Scheuer, which she felt corresponded to her own observations of him in court. Defendant seemed to her adequately to understand what was transpiring even if he could not always verbalize appropriate responses. She acknowledged the significant limitations in defendant’s ability to express himself, but was of the opinion that various accommodations would enable defendant to participate sufficiently in his defense. These included shortened court sessions with frequent breaks and recesses to allow for more extended lawyer/client consultation; open file discovery; encouragement of questions that would elicit short, unlayered responses, and, if defendant testified, the restatement of questions in different ways “to assure that he has used the word intended in responding”; daily copies of the trial transcript; and the court’s willingness to entertain applications for the recall of witnesses. The court also indicated that, in the event of a conviction, it would review the proceedings, applying the criteria set forth in People v Francabandera (33 NY2d 429 [1974]) and Wilson v United States (391 F2d 460 [DC Cir 1968]) to determine whether, in light of his disability, defendant had received a fair trial. The court eventually did perform this retrospective analysis over the objections of both sides9 and found that the trial had been fair.
*528The Appellate Division affirmed defendant’s judgment of conviction in a brief memorandum (68 AD3d 541 [2009]). It found no basis to disturb the trial court’s weighing of the conflicting testimony. It agreed with the trial court that standardized tests were of “limited value” in determining legal competency and that Dr. Scheuer’s testimony was “very significant” because of her extended contact with defendant (id. at 542).
It is fundamentally incompatible with due process to try an incompetent defendant (Cooper v Oklahoma, 517 US 348, 364 [1996]). If, then, a defendant is to be tried he or she must have the basic capacity to exercise those rights upon which the fairness of a trial has been understood to depend, including the right to effective representation, to confront and cross-examine witnesses, and to elect to testify or to refrain from doing so without penalty. Minimally, a defendant must be able to understand the proceedings against him and possess “sufficient present ability to consult with his lawyer with a reasonable degree of rational understanding” (Dusky v United States, 362 US 402, 402 [1960]). Accordingly, in New York a defendant may not be tried and will be deemed an “incapacitated person” if “as a result of mental disease or defect [he or she] lacks capacity to understand the proceedings against him [or her] or to assist in his [or her] own defense” (CPL 730.10 [1]). Once a bona fide issue has been raised respecting a defendant’s fitness, it is the prosecution’s burden to demonstrate fitness by a preponderance (People v Mendez, 1 NY3d 15, 19 [2003]). Our review of a fitness determination is limited to whether the evidence is legally sufficient to support the determination (id. at 20). This does not mean, however, that a determination of fitness must be upheld because there is some evidence to support it. In People v Jordan (35 NY2d 577 [1974]), for example, we acknowledged that there was some medical evidence to support the view that defendant was fit but nonetheless found that “the proof [fell] short, as a matter of law, of meeting the requirement that defendant be able to comprehend his predicament and be capable of participating rationally in his own defense” (id. at 581). It is, of course, true that a fitness determination is ultimately a legal, and not a medical, finding (see People v Tortorici, 92 NY2d 757 *529[1999]), but like any other legal finding it must rest upon an adequate factual foundation. This one did not.
In passing upon the sufficiency of the People’s showing of defendant’s fitness, it is useful to note that most of the material facts and circumstances are undisputed. Defendant suffers from aphasia secondary to stroke-induced brain damage. He is consequently largely bereft of the ability to express himself in words; he has no fluency, must labor to say even simple things and is prone to paraphasic intrusions (i.e., expressions that have no logical relevance but appear in speech simply as artifacts of neurologic disorder). Psychological testing has repeatedly shown that defendant’s aphasia implicates not only his ability to communicate but his ability to think and understand. After evaluating defendant and observing his interaction with his attorney, two qualified examiners found him unfit to be tried upon the pending indictment because, in their judgment, his physical and cognitive limitations precluded him from effectively assisting his attorney in his defense. There is not and never has been a contention that defendant is malingering; his brain damage and its aphasic expression are demonstrably real, extensive and irremediable. All the medical experts agreed that defendant’s condition cannot improve and will likely worsen. Indeed, despite Dr. Scheuer’s many notations in the hospital chart to the effect that she had observed improvement in defendant, the medical evidence is overwhelming that defendant’s basic debilitating condition could not have improved a whit during his stay at Kirby.10
The question which inevitably arises is how defendant could have been found fit by his “restoration of fitness” treatment team when he had only months before been found unfit by reason of the unalterable sequellae of irreversible brain damage. One can of course hypothesize that the extent of defendant’s impairment was misjudged by his Bellevue evaluators—that although he manifested undeniable speech pathology he had, as Dr. Scheuer ventured, no cognitive deficits, or at least none that would prevent him from being deemed fit for trial. There is, however, no competent medical evidence to support such a theory and much to refute it.
As the trial court recognized, and as is in any case clear, defendant suffered and claimed to be incapacitated not from a *530psychiatric illness but from a neurologic injury and consequent neurologic disorder. The only neurologic experts to testify at defendant’s competency hearing were Dr. Capruso, who was qualified both as an expert in forensic psychology and neuropsychology, and Dr. Henry, a neurologist of evident eminence. Only these witnesses were qualified to offer expert opinion testimony as to the nature and extent of defendant’s neurologic disorder. The People’s witnesses, Drs. Hicks and Scheuer, although undoubtedly possessing expertise in forensic psychiatry and psychology, respectively, were never qualified as neurological experts. Indeed, the court stated quite clearly on the record that Dr. Hicks was not being qualified to testify as an expert in neurology and he repeatedly indicated that neurology and neuropsychology were not his areas of expertise. He, in fact, stated that he generally relied on Dr. Capruso for neuropsychological findings. As for Dr. Scheuer, there was not the slightest suggestion that she was qualified to offer a view as to the nature or extent of defendant’s disorder. While Scheuer and Hicks could of course testify respecting their personal interactions with and observations of defendant, their opinions as to the severity of his underlying condition and the level of intellectual functioning it could be reasonably expected to permit, were not appropriately relied upon by the court.
While fitness to stand trial may be, and often is, established notwithstanding a defendant’s affliction with a neurological disorder, it is obvious that there may come a point at which such a disorder will so compromise thought and judgment as to preclude a finding that he or she is capable of understanding the proceedings and assisting his or her attorney in presenting a defense. As this record compellingly demonstrates, whether that point has been reached is not always reliably ascertained on the basis of a standard competency exam. Here, it was manifest that defendant’s ability to speak had been devastated by a catastrophic series of neurologic events. Whether, despite his substantial loss of expressive language, defendant retained sufficient cognitive capacity to understand the proceedings against him was an inquiry which an interview seeking to elicit information in spoken form was ill-suited to answer. Indeed, as Dr. Henry testified, it is treacherously difficult to gauge the extent of an aphasic’s intellectual functioning and persons inexpert in neurology and language pathology—even otherwise learned health care professionals—are prone to err, often with the kindest of intentions, in ascribing to aphasics greater *531comprehension than they actually possess. In this context particularly, science was not to be shunned in favor of what amounted to little more than anecdotal accounts of what the defendant on one occasion or another seemed to understand. Standardized neuropsychological testing was in defendant’s case not, as the trial court and Appellate Division thought, of “limited value,” it was the only scientifically reliable means of judging defendant’s cognitive capacity.
As noted, that testing, both at Bellevue and as performed by Dr. Capruso at Kirby Forensic Psychiatric Center, showed that defendant was significantly impaired—that his basic ability to use language and think had been so compromised that in critical areas of cognitive performance he was functioning on the level of a six year old. While the threshold for fitness is not high and borderline intellectual functioning is ordinarily not preclusive of a fitness finding, the deficits disclosed by defendant’s testing were profound, in numerous categories placing him in the very lowest percentiles, and, particularly when coupled with his expressive disability, raised a profound question as to whether he could in fact comprehend the proceedings and assist his attorney. Inasmuch as there was no competent countervailing proof that the defendant’s neurologic injury was not so severe as to cast his fitness in doubt, it is plain that the People did not meet their burden to demonstrate fitness by a preponderance and that defendant should have been deemed an incapacitated person.
While it is true that numerous persons with medical and psychological expertise testified at defendant’s competency hearing, and that they did not always agree, the proceeding is not therefore accurately pigeonholed as a “battle of the experts” properly and unreviewably resolved by the hearing court’s crediting of the experts of one side. As noted, upon the crucial issue in this competency proceeding—the extent to which defendant’s neurological defect would interfere with his ability to comprehend and participate in the trial—there was no “battle” at all. The qualified opinion testimony was unanimous that defendant’s cognitive impairment was incompatible with comprehension of a proceeding as inevitably complex as a criminal trial.
Even if the expert testimony bearing upon the magnitude of defendant’s neurologic injury and its cognitive sequellae were discounted, and consideration were limited to the information elicited on the standard competency exams, the evidence would *532still be insufficient to support a fitness finding. Leaving aside the issue of how much defendant was capable of understanding, it was absolutely clear to all of the doctors who assessed his essentially static aphasic condition that defendant was severely impeded in communicating—that he could summon and use words only with the greatest difficulty and that, for the most part, he could communicate, even urgent needs, only by means of dialogue involving the skilled use of yes-or-no questions. The two Bellevue evaluators who had the benefit of seeing defendant interact with his attorney were quick to understand that he could not effectively assist her in responding to the serious charges against him, as was Dr. Capruso. The impracticability of effective trial representation was, of course, eloquently confirmed in court by defendant’s attorney, whose view, particularly after having represented and intensively dealt with defendant for a year, was entitled to serious consideration (see Medina v California, 505 US 437, 450 [1992] [“defense counsel will often have the best-informed view of the defendant’s ability to participate in his defense”]; United States ex rel. Roth v Zelker, 455 F2d 1105, 1108 [2d Cir 1972] [“The opinion of a defendant’s attorney as to his ability to understand the nature of the proceedings and to cooperate in the preparation of his defense, is indeed significant and probative”]; People v Gelikkaya, 84 NY2d 456, 460 [1994] [defense counsel “was in the best position to assess defendant’s capacity”]; see also People v Jordan, 35 NY2d at 580), not only because of her extensive contact with defendant but because of the enormous practical and ethical difficulty of discharging the representational obligations that would be entailed by a finding that her profoundly disabled client was fit.11 It was confirmed as well by defendant’s hearing testimony, which demonstrated, fairly dramatically, that defendant would not be able usefully to communicate with his counsel about legal matters.
Only Dr. Hicks proposed as a viable option that defendant collaborate with his attorney by means of yes-or-no exchanges. He, however, admitted that he had been thinking in terms of a plea and had not really considered how this device might be accommodated in the context of a rapidly unfolding trial. “A trial,” he *533said, would be “problematic” and testimony by the defendant in his own defense, even more so; defendant, he suggested, would have to be constantly led in his testimony and the discourse of the trial kept simple. Hicks, who acknowledged that he did not know much about trials, perhaps can be forgiven for proposing such a manifestly unworkable accommodation, but judges should know better. It was simply unrealistic to suppose that the legal, procedural and testimonial complexities of a criminal trial could be usefully discussed by a lawyer with his or her client in the form of serial yes-or-no questions, much less that such an unwieldy means of conferring—made all the more so by the need, recognized by Dr. Hicks, to check and recheck all of defendant’s replies by other yes-or-no questions12—could be availing amid the press of trial proceedings. Given all that Hicks did not consider and the contrary conclusions of all of the other professionals who had formally evaluated defendant,13 it would be exceedingly difficult to say “on the whole record in this case” (People v Jordan, 35 NY2d at 581) that defendant’s fitness had been made out.
The People’s other witness, Dr. Scheuer, it should be noted, never formally interviewed defendant or evaluated his competency. That her manifestly erroneous and highly subjective impressions of defendant were accorded practically dispositive significance on the fundamental issue of competency is a matter of concern. A determination as to a defendant’s mental competency, while ultimately made by a judge according to a legal standard, must, when bona fide questions have been raised as to a defendant’s fitness, be premised on qualified expert opinion. Indeed, it is a central purpose of article 730 of the Criminal Procedure Law to assure that competency determinations have a sound medical basis. The article, with its extensive provision for medical guidance of judicial decision making, implicitly recognizes that mental competency, when genuinely at issue, is not, like credibility, a quality that may be judged simply on the basis of the court’s unaided observations, or in exclusive reliance upon similarly inexpert impressions of others.
*534In the end, the trial court was apparently torn over her decision. While finding defendant fit, she prescribed numerous accommodations which themselves raise questions as to defendant’s actual competency. For example, although the court found that defendant did not exhibit mental impairment, she required that where possible the attorneys should ask questions designed to elicit simple, unlayered responses. It is puzzling why such forced simplicity at odds with basic precepts governing the elicitation of trial testimony should be required to accommodate a defendant without mental impairment. But it was not only in her list of accommodations that the court can be understood to have expressed her ambivalence over her competency decision; she also undertook to review the fairness of the completed trial in light of defendant’s disability. Both the defense and the prosecution objected to this review, albeit for different reasons.
It is on many levels wrong to subject an incapacitated person to a criminal trial. Fitness, then, is a matter that must be determined before trial; the subsequent course of any ensuing prosecutorial proceedings is not relevant to its adjudication. The cases relied upon by the trial court in proposing and conducting its post hoc review, Wilson v United States (391 F2d 460 [1968]) and our decision in People v Francabandera (33 NY2d 429 [1974]), address the issue of what should be done to assure that a defendant who suffers from amnesia as to the events upon which his prosecution is based has received a fair trial. We made clear in Francabandera that we did not view the discreet mnemonic deficiency there at issue as one raising a question as to the defendant’s essential fitness, and, that being that case, there could have been no objection to a retrospective assessment to determine whether the deficiency did, in fact, operate to deprive the defendant of a fair trial; there was no possibility that an unfit defendant was being put through a trial to test the hypothesis that he was fit. Here, however, fitness was the issue at the beginning of the trial and, realistically, remained the issue at its end. The question was not whether defendant had some non-global deficiency that might, but would not necessarily, interfere with his right to a fair trial, but whether he had the basic capacity to understand the proceedings and assist his attorney. That issue was not appropriately reserved for final determination at the trial’s end. If the court had such grave doubts about defendant’s fitness, those doubts should, given the burden placed upon the People and the fundamental importance of a criminal defendant’s competency *535to the integrity of the proceedings against him, have been resolved against subjecting defendant to a trial.
Accordingly, while I agree with the majority that the trial court was extraordinarily painstaking in her review of this very unusual and difficult matter, I cannot agree that her legal conclusion as to defendant’s fitness was, as a matter of law, cor^ rect.
Judges Ciparick, Graffeo, Read, Smith and Pigott concur with Judge Jones; Chief Judge Lippman dissents and votes to reverse in a separate opinion.
Order affirmed.
. Lest there should be any question as to whether the futility of this exercise was attributable to some flaw in the attorney’s approach to the representation, it should be noted that the relationship was, apart from its apparently organically fated inutility, on a personal level, excellent. It is, moreover, clear from the record that defendant’s attorney was an uncommonly gifted legal advocate and uniquely well-qualified for this particular representation by reason of her prior career as a psychologist.
. Hospital forensics committees are required in state psychiatric facilities under the Office of Mental Health’s regulations (see 14 NYCRR 541.3). These committees, composed of licensed mental health clinicians (one of whom must be a board-certified psychiatrist) (id.) must be consulted in connection with all forensically related decisions made respecting patients committed via provisions of the Criminal Procedure Law (see 14 NYCRR 541.9).
. Contrary to the majority’s suggestion, Dr. Capruso’s conclusion as to defendant’s fitness was not based simply on a series of neurological examinations.
. Hicks explained in his testimony that the defendant’s case was initially referred to him for an opinion as to whether, in light of the untreatable nature of his disability, defendant was a candidate for Jackson relief, i.e., the dismissal of the indictment followed by either release or civil commitment (see Jackson v Indiana, 406 US 715, 738 [1972]).
. Although it was Hicks’s view that defendant was fit, he also was of the view that, if defendant was ultimately found unfit, he should be afforded Jackson relief since his condition was intransigent.
. These were Hicks’s words, not defendant’s. Hicks, it should be noted, kept no notes of his interview with defendant and his testimony did not purport to convey with literal exactitude what was said. Defendant during his *524competency hearing testimony, denied that he had ever met with Hicks privately. He said that they spoke in the “gym.”
. Defendant was readmitted to Bellevue from Hikers Island (to which he had been returned following his discharge from Kirby) during the competency hearing (which took place over seven days spread out over some six months) to ascertain whether he had had yet another stroke. It was during the course of that hospitalization that he was interviewed and examined by Dr. Henry.
. This testimony is hard to reconcile with Dr. Hicks’s account of defendant’s knowledge of pleas. Hicks was apparently the only examiner favored by defendant with a display of his conversancy in the lore of pleas.
. Although defendant’s counsel did in the end ask that the verdict be set aside as the product of an unfair trial, counsel had consistently opposed the retrospective evaluation proposed by the court, maintaining that no level of accommodation could compensate for her client’s essential lack of capacity to *528understand the proceedings and assist in his defense. The prosecution, on the other hand, opposed the post-trial evaluation on the ground that, since defendant had been deemed fit, there should be no residual issue as to whether he had by reason of disability been deprived of a fair trial.
. Dr. Hicks testified that he found these notes “a little puzzling” given the fundamentally irreversible nature of defendant’s brain damage and consequent aphasia.
. This, it should be stressed, was not, as the majority suggests in its citation to People v Morgan (87 NY2d 878, 880 [1995]), a situation in which the court had before it only the bare “observations and representations” of counsel. The record in this case overwhelmingly demonstrated that counsel’s representations were authentically premised.
. The need to do this was recognized by the court as well when it required in its list of accommodations that, should defendant testify, the “attorneys should restate their questions to him in different ways, to assure that he has used the word intended in responding to their questions.”
. Although there is reference made in the hearing and medical records to other doctors concurring in Dr. Hicks’s fitness finding, there is no indication that any of these physicians ever actually independently assessed defendant’s competency.