Because, in my view, sound and longstanding precedent requires that we reverse the judgment of conviction and grant a new trial, I respectfully dissent. Although not every impermissible word or phrase in a jury instruction violates due process, here County Court’s reference *928to "substantial doubt” in the jury instruction on reasonable doubt, when read in context, resulted in " 'a reasonable likelihood that the jury * * * applied the challenged instruction in a way’ that violates the Constitution” (Estelle v McGuire, 502 US 62, 72, quoting Boyde v California, 494 US 370, 380, reh denied 495 US 924).
In two recent cases we have held that the court’s use of the substantial doubt language in a jury instruction on reasonable doubt that is almost identical to the jury charge at issue here impermissibly reduced the People’s burden of proof and violated the defendants’ Constitutional right to due process of law (see, People v Branch, 224 AD2d 926; People v Moore, 216 AD2d 902, lv denied 87 NY2d 905). The majority holds that Victor v Nebraska (511 US 1, reh denied sub nom. Sandoval v California, 511 US 1101) requires that we change our position. I disagree. In Victor v Nebraska (511 US 1, 19, supra), the Supreme Court observed that the use of the phrase " 'substantial doubt’ ” in a jury instruction on reasonable doubt is "somewhat problematic” because, although the jury may interpret that phrase to mean " 'not seeming or imaginary’ ”, i.e., something more than a speculative doubt, that phrase may also be understood to mean a doubt " 'specified to a large degree’ ”, which implies a quantum of proof greater than that required for proof beyond a reasonable doubt. The Court made the distinction that, where the term "substantial doubt” is employed not to suggest the quantum of proof but, rather, to illustrate that reasonable doubt is not predicated on " 'mere possibility * * * bare imagination, or * * * fanciful conjecture’ ”, reversal is not required (Victor v Nebraska, supra, at 20).
We have repeatedly condemned the use of the phrase "substantial” in defining reasonable doubt (see, People v Jackson, 199 AD2d 1002, lv denied 83 NY2d 872; People v Freeman, 193 AD2d 1117, lv denied 82 NY2d 718; People v Payne, 192 AD2d 1117; People v Frank, 186 AD2d 977) and, as previously stated, we have twice reversed upon that ground for language in a jury instruction almost identical to the one at issue (see, People v Branch, supra; People v Moore, supra). We have also noted that the CJI charge on reasonable doubt does not use the phrase "substantial doubt” in defining reasonable doubt (1 CJI[NY] 6.20). The Committee on Criminal Jury Instructions has recommended that "words and phrases that should be avoided when charging on reasonable doubt are: 'good, sound substantial reasons,’ 'reasonable certainty’, 'substantial reason’ [and] 'substantial doubt’ ” (1 CJI Newsletter No. 2, at 2 [Winter 1993]).
*929Our decisions in Branch and Moore (supra) were based upon the careful and deliberate judgment that there is a reasonable likelihood a jury would not understand the words "substantial doubt” as referring only to a doubt that has substance and reality but, rather, to the meaning that is generally understood, i.e., "abundant”, "hefty”, "significant”, "sizeable” or "weighty” (Kipfer, Roget’s 21st Century Thesaurus, at 805).
The Supreme Court, in arriving at its conclusion, found it necessary painstakingly to parse the context in which the words "substantial doubt” were used. "[T]he context makes clear”, the Court concluded, "that 'substantial’ is used in the sense of existence rather than magnitude of the doubt” (Victor v Nebraska, supra, at 20). In determining whether the People’s burden of proof meets the same standard in every trial, appellate courts should not be obliged to undertake a semantic analysis regarding a jury’s understanding of the words used in a jury instruction. Such an analysis would compel appellate courts to make what is at best a guess whether the jury took the words "substantial doubt” to mean that a reasonable doubt must be based upon a substantial reason or, instead, whether the jury took the words to mean that a reasonable doubt requires a "sizeable” or "weighty” doubt.
The dissent in Victor v Nebraska (supra, at 32-33 [Blackmun and Souter, JJ., dissenting]) in emphasizing the danger of the use of the words "substantial doubt”, even when used in tandem with the concept of doubt predicated on imagination and conjecture, pointed out that "the reason the Court condemned the 'substantial doubt’ language in Cage [v Louisiana (498 US 39)] had nothing to do with the absence of appropriate contrasting language; rather, the Court condemned the language for precisely the reason it gave: '[T]he words 'substantial’ and 'grave’, as they are commonly understood, suggest a higher degree of doubt than is required for acquittal under the reasonable doubt standard’ [Cage v Louisiana, supra], at 41.” If the Justices of the Supreme Court may themselves differ with respect to the effect and meaning of the phrase "substantial doubt”, it is unreasonable to expect lay jurors to understand when a doubt is considered reasonable in the context of language that may confuse rather than clarify. As Justice Kennedy observed in his concurring opinion regarding the use of the words " 'moral evidence’ ”, an observation that is equally applicable to the words "substantial doubt”, "for jurors who have not had the benefit of the Court’s research, the words will do nothing but baffle” (Victor v Nebraska, supra, at 23 [Kennedy, J., concurring]). There is at the very least a *930reasonable likelihood that a jury will misapply the challenged jury instruction, notwithstanding the semantic analysis in Victor v Nebraska (supra).
Further, Victor v Nebraska (supra) does not compel us to change our position. Branch and Moore (supra) remain properly decided under the greater protections afforded by the New York State Constitution. "[Tjhere can be no doubt in New York that the fairness of a criminal proceeding is of particular State concern, and New York historically has provided various protections in this area above the Federal constitutional minimum” (People v Alvarez, 70 NY2d 375, 379; see, People v Bigelow, 66 NY2d 417, 426-427; People v Johnson, 66 NY2d 398, 407; People v Isaacson, 44 NY2d 511, 519-520, rearg denied 45 NY2d 776; People v Hobson, 39 NY2d 479, 483-484). In applying that principle, the courts of New York have "sought to provide and maintain 'bright line’ rules.to guide the decisions of law enforcement and judicial personnel who must understand and implement our decisions in their day-to-day operations in the field” (People v P. J. Video, 68 NY2d 296, 305, cert denied 479 US 1091). Adopting a bright line rule that requires reversal when a court uses the "substantial doubt” language sends a clear message to trial courts to refrain from using such language and fosters conformity and predictability in the exacting standard that must be applied before a defendant may be found guilty of the crime charged.
I disagree with the majority that what we said in People v Towndrow (187 AD2d 194, 196, lv dismissed 81 NY2d 1021) must guide us here. That case concerned a jury instruction that a reasonable doubt must be " 'based upon some good sound substantial reason’ ” and did not, as here, define a reasonable doubt as the equivalent of a " 'substantial doubt’ ” (People v Towndrow, supra, at 195). The language in Towndrow, although objectionable, did not, as here, directly suggest the quantum of proof required for conviction. Further, we there reminded trial courts that "[w]e have condemned that formulation repeatedly, in language that leaves no doubt about its impropriety” (People v Towndrow, supra, at 195).
The gravity of the danger presented by a jury’s misinterpretation of the quantum of proof necessary to convict a defendant in a criminal trial cannot be overstated. "To be a meaningful safeguard, the reasonable-doubt standard must have a tangible meaning that is capable of being understood by those who are required to apply it. It must be stated accurately and with the precision owed to those whose liberty or life is at risk. Because of the extraordinarily high stakes in criminal tri*931ais, '[i]t is critical that the moral force of the criminal law not be diluted by a standard of proof that leaves people in doubt whether innocent men are being condemned’ ” (Victor v Nebraska, supra, at 29 [Blackmun and Souter, JJ., dissenting], quoting In re Winship, 397 US 358, 364).
The reasonable doubt standard of proof provides the protection owed to a criminal defendant only to the extent that the jury comprehends its meaning. Because the court’s reference to "substantial doubt” in the instruction presents a reasonable likelihood that the jury may have misapprehended the concept of reasonable doubt, I would adhere to the precedent that we have established by reversing the judgment and granting defendant a new trial. (Appeal from Judgment of Jefferson County Court, Clary, J.—Robbery, 2nd Degree.) Present—Den-man, P. J., Green, Wesley, Balio and Boehm, JJ.