“Great Certainty” Is an Unconstitutional Reasonable-Doubt Definition: Plain Error and Partial Overruling of State v. Levitt

1. Introduction

State v. Kolts arises from delayed disclosure allegations of child sexual abuse. In 2018, N.H. reported that her uncle, Rein Kolts, sexually assaulted her repeatedly during childhood. The State charged two counts of aggravated sexual assault under 13 V.S.A. § 3253(a)(8), requiring proof of sexual assault and that the victim was under thirteen at the time of the act. A jury convicted Kolts in May 2024.

On appeal, Kolts raised three issues: (1) sufficiency of the evidence on count two (penetrative sexual contact), especially as to N.H.’s age; (2) a federal due-process challenge to the reasonable-doubt instruction; and (3) admission of prior-bad-act evidence. The Vermont Supreme Court affirmed the sufficiency ruling but reversed for a new trial due to an unconstitutional reasonable-doubt definition, expressly overruling State v. Levitt, 2016 VT 60, “to the extent it conflicts.”

2. Summary of the Opinion

A. Sufficiency of the evidence (Count Two)

Applying the deferential acquittal standard, the Court held the State presented “any evidence” from which a rational juror could find, beyond a reasonable doubt, that N.H. was under thirteen at the time of the barn assault. The Court emphasized permissible use of circumstantial evidence—particularly N.H.’s testimony that assaults in the barn occurred when she was “eleven and under” and that they “stopped” when she was “eleven or twelve,” as well as inferences from typical academic progression.

B. Jury instruction (Reasonable doubt)

The Court held it is constitutional error—and plain error—to define “beyond a reasonable doubt” as being “convinced of it with great certainty.” The definition impermissibly lowered the State’s burden below the federal due-process benchmark articulated in In re Winship. Because the error was obvious under prevailing federal law, affected substantial rights, and threatened the fairness and public reputation of proceedings, the Court reversed and remanded for a new trial. In doing so, it overruled State v. Levitt insofar as Levitt approved the “great certainty” phrasing when considered within the charge as a whole.

C. Prior bad acts

The Court did not reach the prior-bad-act evidentiary claim because reversal was required on the constitutional instruction error.

3. Analysis

3.1 Precedents Cited

A. Vermont sufficiency and jury-function cases

• State v. Anderkin, 145 Vt. 240 (1984): Reaffirmed the foundational rule that the State must prove each element beyond a reasonable doubt—framing the age-element dispute in count two.
• State v. Davis, 2018 VT 33: Supplied the acquittal standard—viewing evidence in the light most favorable to the State and asking whether it could fairly and reasonably convince a jury.
• State v. Cameron, 2016 VT 134: Reinforced deference to juries in close, fact-dependent cases; used to underscore the limited judicial role on acquittal review.
• State v. Perrault, 2017 VT 67 and State v. Durenleau, 163 Vt. 8 (1994): Together mark the line between rational inference (permitted) and conjecture/mere suspicion (insufficient).
• State v. Messier, 146 Vt. 145 (1985): Supported the proposition that the State need not exclude every hypothesis of innocence; the jury need not elevate alternative explanations to reasonable doubt.
• State v. Tenney, 143 Vt. 213 (1983), State v. Hammond, 2012 VT 48, and State v. Hinchliffe, 2009 VT 111: Anchored the rule that contradictions and credibility are for the jury, enabling the Court to treat N.H.’s varying age references as a jury question.

B. Vermont plain-error framework

• State v. Blanchard, 2021 VT 13: Confirmed unpreserved instruction challenges are reviewed for plain error.
• State v. Welch, 2020 VT 74 and State v. Alzaga, 2019 VT 75: Provided Vermont’s four-factor plain-error test and the “undermines our confidence in the verdict” formulation for instructional error.

C. Federal constitutional control of reasonable-doubt instructions

• In re Winship, 397 U.S. 358 (1970): The Court treated Winship as establishing the constitutional benchmark requiring guilt be proven with “utmost certainty.”
• Cage v. Louisiana, 498 U.S. 39 (1990): Used for its method—examining the ordinary meaning of instruction words (“grave,” “substantial”) and invalidating language that raises the degree of doubt for acquittal or lowers certainty for conviction. Kolts analogized “great certainty” to Cage’s condemned terms.
• Victor v. Nebraska, 511 U.S. 1 (1994) and Estelle v. McGuire, 502 U.S. 62 (1991): Supplied the governing test: whether there is a “reasonable likelihood” the jury applied the instruction to permit conviction on insufficient proof. Victor also illustrated that problematic words can sometimes be cured by context or an alternative “hesitate to act” definition—features missing here.
• Sullivan v. Louisiana, 508 U.S. 275 (1993): Cited for the principle that a defective reasonable-doubt instruction can vitiate the jury’s findings because it corrupts the standard of decision itself.
• Jackson v. Virginia, 443 U.S. 307 (1979): Appeared in the Winship/Cage chain as part of the due-process requirement for proof beyond a reasonable doubt.
• United States v. Partin, 552 F.2d 621 (5th Cir. 1977) and Cool v. United States, 409 U.S. 100 (1972): Supported the broader constitutional theme: diminishing the burden of proof conflicts with the presumption of innocence.
• Brown v. Greene, 577 F.3d 107 (2d Cir. 2009): Reinforced the “reasonable likelihood” framing in a modern federal appellate articulation.

D. The status of defining reasonable doubt (cautionary authorities)

• State v. McMahon, 158 Vt. 640 (1992), State v. Francis, 151 Vt. 296 (1989), State v. Haskins, 2016 VT 79, and State v. Redmond, 2020 VT 36: Used to reiterate Vermont’s long-standing warning that defining reasonable doubt is “hazardous.”
• Miles v. United States, 103 U.S. 304 (1880) and Holland v. United States, 348 U.S. 121 (1954): Provided parallel federal skepticism about attempts to define the standard.

E. Plainness/obviousness and federal supremacy

• State v. Koons, 2011 VT 22 and U.S. v. Olano, 507 U.S. 725 (1993): Defined “plain” as “clear” or “obvious” under existing law.
• State v. Scampini, 77 Vt. 92 (1904) and State v. Prouty, 94 Vt. 359 (1920): Established the Vermont Supreme Court’s obligation to follow U.S. Supreme Court decisions on federal constitutional questions.
• State v. Judkins, 161 Vt. 593 (1993) (mem.) and State v. Blow, 157 Vt. 513 (1991): Used as an “obviousness” comparator: unlike Judkins (where most jurisdictions went the other way), here federal law strongly dictated the result.

F. Additional authorities bearing on impact and correction

• State v. Barrett, 132 Vt. 369 (1974) and State v. Webster, 2017 VT 98: Supported the presumption that juries follow instructions—critical to rejecting Levitt’s “repetition cures” rationale.
• United States v. Marcus, 560 U.S. 258 (2010): Cited in the fairness/integrity analysis under the fourth plain-error factor.
• State v. Shores, 2025 VT 62: Provided the Vermont standard for overruling precedent—caution, but not “slavish” adherence.
• White v. American Airlines, Inc., 915 F.2d 1414 (10th Cir. 1990): Notable comparative usage: “great certainty” appears as a descriptor associated with clear-and-convincing evidence, underscoring that it is below beyond-a-reasonable-doubt.

3.2 Legal Reasoning

A. Sufficiency: rational inference versus conjecture

The Court’s sufficiency analysis is orthodox and highly deferential. It relied on two key evidentiary pathways: (i) direct testimony tying the barn assaults to “eleven and under” / “eleven or twelve,” and (ii) circumstantial inference that “sixth grade” aligns with being under thirteen, supported by evidence of a typical school trajectory (high school graduation at eighteen; college four years later; PhD study). Importantly, the Court treated inconsistencies (“twelve or thirteen”) as credibility/contradiction questions reserved to the jury under Tenney and Hammond.

B. The core holding: “great certainty” lowers the constitutional burden

The opinion’s central move is linguistic and constitutional: it treats Winship’s “utmost certainty” as the due-process benchmark and concludes that “great certainty” is materially less demanding—“remarkable” rather than “greatest/highest.” This difference, though seemingly subtle, is dispositive because the burden of proof is the constitutional “prime instrument” for reducing wrongful convictions (Winship).

Using Cage as a template, the Court treats the adjective (“great”) as an impermissible signal that the jury may convict with less certainty than the Constitution demands. Unlike Victor, where questionable wording was arguably contextualized (and also paired with the “hesitate to act” formulation), the Kolts instruction contained no alternative definition or contextual limiter that would recalibrate “great certainty” upward to the Winship level.

C. Why the rest of the charge does not cure the defect (and why Levitt is overruled)

The Court rejected Levitt’s balancing logic (presumption of innocence + repetition of the standard + admonitions about conjecture). Once the judge defines the term wrongly, repeating the term repeats the wrong definition. Two presumptions drive this conclusion: (i) jurors follow instructions (Barrett, Webster), and (ii) federal law asks whether there is a “reasonable likelihood” the jury applied the challenged definition to convict on insufficient proof (Estelle, Victor). Under those presumptions, Levitt’s premise that surrounding language “balances” the definitional defect is untenable.

D. Plain error: error, obviousness, prejudice, and institutional integrity

The Court applied Vermont’s four-factor test (Welch) under V.R.Cr.P. 52(b):

• Error: “Great certainty” is below Winship’s constitutional level.
• Obviousness: Federal precedent is controlling and firm; Levitt itself warned against definitional experiments; no broader jurisprudential support existed for “great certainty.”
• Prejudice: A lowered burden “unconstitutionally affected and interfered” with due-process rights—the standard governing guilt determination.
• Fairness/Integrity: Allowing convictions under a diminished burden would damage the legitimacy of criminal adjudication.

Although the defendant argued “structural error,” the Court did not decide that question because it found reversal warranted on plain-error grounds.

3.3 Impact

A. Immediate doctrinal change: Vermont disallows “great certainty” definitions

The decision establishes a clear statewide rule: defining “beyond a reasonable doubt” as “great certainty” is unconstitutional and constitutes plain error. Practically, trial courts must avoid that language; the safer course—consistent with Vermont’s repeated warnings (Francis, McMahon)—is to avoid defining the term at all, or to use constitutionally vetted formulations if a definition is given.

B. Appellate consequences: expanded unpreserved-instruction exposure

By labeling the issue plain error, the Court lowers procedural barriers to relief where the “great certainty” formulation was used without objection. The opinion thus increases the likelihood of reversals (or at least serious appellate risk) in any pending or future appeals involving the same language.

C. Pattern instructions and trial practice

Vermont criminal jury instructions will need to be audited for this phrasing. Prosecutors and defense counsel should scrutinize proposed reasonable-doubt language as a high-stakes issue; trial judges should treat the matter as federal-constitutional and non-negotiable.

D. Substantive criminal litigation is unchanged on sufficiency; procedure is changed on burden

Kolts leaves Vermont’s sufficiency jurisprudence intact—deference to juries, rational inferences, and credibility determinations remain central. The major shift is procedural-constitutional: the permissible content of jury guidance on the most consequential standard in criminal law.

4. Complex Concepts Simplified

• “Beyond a reasonable doubt”: The highest burden of proof in law, required for criminal conviction. It does not mean absolute certainty, but it requires a very high level of confidence before the State may take a person’s liberty.
• Why defining it is risky: Jurors often understand the phrase intuitively; attempts to “clarify” can unintentionally weaken the State’s burden or confuse jurors—hence Vermont’s repeated warning that definitions are “hazardous.”
• Plain error (V.R.Cr.P. 52(b)): An appellate safety valve allowing reversal even without an objection when an obvious error affects substantial rights and threatens the fairness and integrity of the proceedings.
• Circumstantial evidence and inference: Proof from surrounding facts (e.g., typical grade-to-age progression) from which jurors may logically infer the element in question; it is allowed so long as it is rational and not mere speculation.
• Motion for acquittal: A request for the judge to remove the case from the jury because the State’s evidence is legally insufficient. Courts must view the evidence favorably to the State and leave close factual calls to the jury.
• Structural error (not decided here): A category of constitutional errors that may require automatic reversal because they affect the framework of the trial. The Court reversed on plain-error grounds and did not reach the structural-error question.

5. Conclusion

State v. Kolts draws a bright constitutional line in Vermont criminal procedure: courts may not define “beyond a reasonable doubt” as “great certainty.” The decision treats the burden of proof as a federal due-process floor set by In re Winship and holds that “great certainty” falls below it. In doing so, the Court corrects—and partially overrules—State v. Levitt, rejecting the idea that surrounding instructions can “balance out” a defective definition. The immediate significance is practical and systemic: a single phrase in a jury charge can mandate retrial, even absent an objection, because it risks conviction on less than the constitutionally required standard.