State v. Krause: Victim-Interconnected “Reporting Context” and Overlapping Witnesses Can Justify Denial of Severance Despite Limited Cross-Admissibility of Underlying Sexual-Assault Details

I. Introduction

State v. Krause (Wash. Mar. 26, 2026) addresses when a trial court must sever multiple sexual-offense counts under CrR 4.4(b), which requires severance when it will “promote a fair determination of the defendant’s guilt or innocence of each offense.” Cole Edward Krause was tried in a single proceeding on four rape counts involving three complainants (A.C.S., A.L., and A.C.). A central feature of the trial record was that the complainants’ decisions to report were “intertwined”: one complainant’s experience and aftermath affected another’s willingness to report; later reports were made partly to support earlier complainants.

Krause moved repeatedly to sever some or all counts into separate trials. The trial court denied severance pretrial and again midtrial, reasoning that the cases were related, certain evidence was “cross-admissible,” defenses were largely the same (consent), and the jury could be instructed to consider each count separately. The Court of Appeals reversed (unpublished), but the Washington Supreme Court reversed that decision, holding the trial court acted within its discretion.

Key issues included: (1) the proper standard of appellate review of severance rulings; (2) application of the four severance factors (strength, defenses, instructions/compartmentalization, cross-admissibility); and (3) whether, and how, the “benefits of joinder” may outweigh prejudice where the detailed facts of each alleged sexual assault would not be admissible in separate trials.

II. Summary of the Opinion

The Supreme Court held the trial court did not manifestly abuse its discretion in denying severance. Although the State conceded (and the majority accepted) that the detailed “rape facts” for each count would not be cross-admissible under ER 404(b) in separate trials, the Court concluded that (a) some “reporting-context” evidence would be admissible across trials to explain credibility, motive, and state of mind, (b) the jury was instructed to decide each count separately, (c) defenses were not meaningfully conflicting, and (d) the benefits of joinder—especially substantial overlap in witnesses and avoiding repeated victim testimony—outweighed the prejudice from non-cross-admissible details.

The Court also rejected arguments for de novo review and applied the usual abuse of discretion standard. It further indicated that the absence of an explicit on-the-record ER 404(b) analysis did not, by itself, convert review to de novo in the context of a CrR 4.4(b) severance ruling.

III. Analysis

A. Precedents Cited (and Their Role)

  • State v. Russell, 125 Wn.2d 24 (1994): The backbone for severance doctrine. The Court relied on Russell for (1) the four-factor severance framework (strength, defenses, instructions, cross-admissibility), (2) the rule that identical defenses reduce confusion, and (3) the core severance concerns (cumulation of evidence; inference of criminal disposition). The Court also cited State v. Russell, 171 Wn.2d 118 (2011) for the principle that failure to request a limiting instruction generally defeats a claim of instructional error.
  • State v. Rice, 110 Wn.2d 577 (1988): Cited (through Russell) for the definition of abuse of discretion—no reasonable judge would decide as the trial court did.
  • State v. Bythrow, 114 Wn.2d 713 (1990): Used for two key propositions: (1) the defendant must show “specific prejudice,” and (2) the “strength of evidence” factor is designed to prevent the State from using a strong count to bolster weak counts. Bythrow also frames the role of jury compartmentalization (with reference to federal authority).
  • State v. McCabe, 26 Wn. App. 2d 86 (2023): Cited to validate the severance procedure as iterative: a preliminary pretrial ruling based on anticipated evidence, then reassessment after the trial evidence unfolds. The majority used this to justify deference to the trial court’s on-the-record reevaluation after hearing testimony.
  • State v. Grisby, 97 Wn.2d 493 (1982): Cited for Washington’s general preference against separate trials—an interpretive “tilt” toward joinder absent demonstrated unfairness.
  • State v. Slater, 197 Wn.2d 660 (2021): Provided the two-step structure emphasized by the majority: (1) assess prejudice via the four factors; (2) if prejudice does not plainly deprive a fair trial, weigh prejudice against benefits of joinder. The Court distinguished Slater because its joined charges were “not connected or related in any way” and involved different witnesses.
  • State v. Bluford, 188 Wn.2d 298 (2017): Central comparison point. The Court distinguished Bluford because it involved joining sex offenses with unrelated robberies and little/no evidentiary overlap. Bluford’s warning that judicial economy cannot outweigh a fair trial is acknowledged, but the majority treated this case as presenting materially greater “benefits of joinder,” chiefly overlapping witnesses and interconnected reporting evidence.
  • State v. Brunn, 145 Wash. 435 (1927): Quoted (via Bluford) for the long-standing principle that multiple charges cannot be used to undermine the accused by presenting a “multiplicity of unrelated and unconnected charges.” The majority used this to reinforce that these counts were not “unrelated and unconnected” given their interrelationships.
  • State v. Watkins, 53 Wn. App. 264 (1989) and State v. Hernandez, 58 Wn. App. 793 (1990): Used to support the idea that where defenses are identical or largely aligned, confusion risks are low, and a defendant must make a “convincing showing” of a strong need not to testify to justify severance.
  • United States v. Johnson, 820 F.2d 1065 (9th Cir. 1987) and Drew v. United States, 331 F.2d 85 (D.C. Cir. 1964): Referenced through Bythrow/Bluford line to support the concept that juries can compartmentalize evidence with instructions, mitigating prejudice from joinder.
  • State v. Riley, 12 Wn. App. 2d 714 (2020): Raised by Krause to argue for de novo review when ER 404(b) analysis is not documented. The Supreme Court declined to extend Riley, stating Krause had not shown an on-the-record ER 404(b) analysis is required for deciding a CrR 4.4(b) severance motion.
  • State v. Saltarelli, 98 Wn.2d 358 (1982) and State v. DeVincentis, 150 Wn.2d 11 (2003): Invoked for the heightened prejudice risk of prior sexual misconduct evidence. The majority limited Saltarelli’s force here by emphasizing the difference between (a) using “prior rape” evidence to show propensity and (b) trying multiple charged counts together, where the evidence is offered to prove each charge’s elements.
  • State v. Miller, 33 Wn. App. 2d 560 (2025): Cited to illustrate the permissible use of multiple-acts evidence to meet the State’s burden rather than for propensity.
  • State v. Markle, 118 Wn.2d 424 (1992) and State v. Kalakosky, 121 Wn.2d 525 (1993): Used to show Washington has previously permitted joinder of sex offenses against multiple victims under comparable circumstances, and that lack of full cross-admissibility does not automatically bar a joint trial.
  • Samia v. United States, 599 U.S. 635 (2023): Quoted (via amicus) for the systemic value of joint trials and the victim-centered benefit of avoiding repeated trauma from multiple proceedings.
  • State v. W.R., 181 Wn.2d 757 (2014) (Owens, J., dissenting) and State v. Morley, 46 Wn. App. 156 (1986): Cited to contextualize the State’s interest in encouraging rape reporting and recognizing barriers to reporting, supporting the Court’s view of joinder’s benefits in sexual-violence prosecutions.

B. Legal Reasoning

1. Standard of review: discretion remains the rule

The Court reaffirmed that severance rulings are reviewed for manifest abuse of discretion (from State v. Russell). It rejected Krause’s bid for de novo review based on (i) the trial judge’s alleged reliance on a prior judge’s ruling and (ii) the lack of an explicit ER 404(b) record. The majority emphasized that the trial court made an independent judgment and engaged in the CrR 4.4 structure—initial assessment plus reconsideration after hearing evidence (citing State v. McCabe).

2. Step One (four-factor prejudice assessment): only partial support for severance

  • Strength of the evidence: The Court deferred to the trial judge’s view that the State’s case was similarly strong on each count despite different reporting timelines, reducing the concern that a strong case would “prop up” weaker counts (Bythrow).
  • Clarity of defenses: The defense was broadly “consent” across counts, with one affirmative defense nuance on the incapacitation theory. The Court concluded the difference was unlikely to confuse the jury or impair presentation of defenses and rejected speculative claims that separate trials would have allowed selective silence.
  • Jury instructions/compartmentalization: The jury received the standard instruction to decide each count separately. The Court noted Krause did not request an ER 404(b) limiting instruction and relied on State v. Russell (2011) and ER 105 to treat that omission as significant.
  • Cross-admissibility: This was the key friction point. The Court accepted that the graphic “details of each victims’ rape” would not be admissible in separate trials (as the State conceded), but held that cross-admissibility exists in an important subset of evidence: the complainants’ relationships, communications, and reporting decisions, which the Court characterized as relevant to credibility, motive, and state of mind. The Court then reiterated that lack of complete cross-admissibility does not automatically require severance (State v. Bluford; State v. Bythrow), especially where the jury can compartmentalize and where some inter-case evidence would recur anyway.

3. Step Two (weighing prejudice against benefits of joinder): “overlapping witnesses” and victim burden become central benefits

Having found prejudice did not plainly deprive Krause of a fair trial, the Court weighed the risk of improper propensity inference from non-cross-admissible details against the “benefits of joinder.” The decisive benefit was that separate trials would “potentially have all of the same witnesses,” including complainants testifying about the reporting context and their interrelationships. This, the Court reasoned, makes the case unlike State v. Bluford and State v. Slater, where witnesses were different and overlap was minimal.

Importantly, the Court expressly elevated victim-facing considerations as part of the “benefits of joinder” analysis: avoiding repeated trauma from multiple trials (quoting Samia v. United States), aligning with Washington’s statutory policy of treating victims with dignity (RCW 7.69.010), and acknowledging low reporting rates and barriers to rape reporting (RCW 7.105.900(3)(b)). The Court used record facts showing reluctance to participate due to “time and stress” and fear of “years” of court proceedings as concrete reasons joinder served the administration of justice.

The Court also rejected the argument that the prosecutor’s “one verdict” rhetoric in closing necessarily converted the case into propensity reasoning, distinguishing the “repeated focus on ... propensity” condemned in State v. Slater.

C. Impact

Doctrinal impact: The decision substantially clarifies how Washington courts may justify denying severance in multi-complainant sexual-assault prosecutions: even when the detailed facts of each alleged assault are not cross-admissible, severance is not required if the trial court reasonably finds (i) meaningful overlap in admissible “reporting-context” evidence, (ii) juror compartmentalization is likely (and is supported by instructions and trial observation), and (iii) the benefits of joinder are strong due to overlapping witnesses and the real burden of repeated testimony.

Practical impact: Trial courts now have more explicit Supreme Court support for treating overlapping witness lists and victim burden/trauma from repeated trials as legitimate “benefits of joinder” at step two—particularly where complainants’ reporting decisions are factually interdependent. Conversely, defendants can expect that arguing “sex crimes are inherently prejudicial” will not suffice without a concrete showing that joinder likely caused improper cumulation or disposition-based reasoning, especially when defenses are straightforward and the jury is instructed to keep counts separate.

Appellate review impact: The Court’s refusal to convert severance review to de novo absent an explicit ER 404(b) record reinforces deference to trial judges who make fact-intensive, on-the-record severance determinations and revisit them after testimony.

Fault line (dissent): Justice Gordon McCloud’s dissent characterizes the majority’s approach as “unprecedented,” asserting it flips prior teaching that sex-crime joinder is especially prejudicial (State v. Bluford; State v. Bythrow; State v. Saltarelli; State v. DeVincentis). That dissent signals future litigation over whether “victim burden” can outweigh the prejudice from admitting multiple non-cross-admissible sexual-assault narratives in one trial, and how to reconcile Krause with the “judicial economy can never outweigh a fair trial” maxim quoted in Bluford.

IV. Complex Concepts Simplified

  • Joinder (CrR 4.3(a)) vs. Severance (CrR 4.4(b)): Joinder is the State charging multiple counts in one case; severance is splitting those counts into separate trials. Even if joinder is allowed, severance may still be required to ensure fairness.
  • “Manifest abuse of discretion”: An appellate court will not reverse just because it would have decided differently; it reverses only if the ruling is outside the range of reasonable decisions.
  • Cross-admissibility: Whether evidence from Count A would be admissible in a separate trial on Count B. If yes, joinder is less prejudicial; if no, joinder can risk the jury using evidence improperly.
  • ER 404(b) and “propensity”: ER 404(b) generally forbids using other bad acts to argue “he did it before, so he likely did it again.” That is “propensity” reasoning. In a joint trial, the danger is the jury may treat multiple allegations as proof of a criminal disposition rather than proving each count independently.
  • Compartmentalization: The idea that jurors can keep evidence for each count separate—helped by clear instructions to decide each count independently.
  • Affirmative defense & preponderance: On one count involving alleged incapacitation, the jury was instructed that Krause had to prove, by a “preponderance” (more likely than not), that he reasonably believed the complainant was not incapacitated.

V. Conclusion

State v. Krause holds that a trial court may, within its discretion, deny severance of multiple sexual-assault counts involving different complainants even when the detailed assault narratives are not cross-admissible—so long as the court reasonably finds the case’s “reporting context” and witness overlap create substantial benefits of joinder, the defenses are not meaningfully conflicting, the evidence strength is not lopsided, and the jury can be instructed (and observed) to decide each count separately. The decision concretely legitimizes overlapping witnesses and the burden of repeated victim testimony as weighty “benefits of joinder,” while preserving the formal two-step severance framework from Russell/Slater/Bluford.