Crawford’s Conundrum: What Has Changed Two Decades Later? 

Dylan Thorson* 

Twenty years ago, the landmark case Crawford v. Washington sought to revolutionize Confrontation Clause analysis. While Crawford has certainly altered the landscape of Sixth Amendment interpretation, the last two decades have seen cases stray from Crawford’s supposed ideal. This article will examine, first, how the Crawford Court’s failure to define “testimonial” left its rule with ambiguity. Second, how the subsequent “Primary Purpose Test,” compromised the Sixth Amendment right as understood in Crawford. Finally, this article will discuss how existing exceptions reflect the exceptions used in the old Roberts test.  

1. Crawford’s Failure to Define “Testimonial” Left Its Rule with Ambiguity 

The Sixth Amendment requires that “in all criminal prosecutions, the accused shall enjoy the right to … be confronted with the witnesses against him.”¹ Prior to Crawford, the Supreme Court in Ohio v. Roberts interpreted statements to be excepted from the Sixth Amendment when a declarant was unavailable and the statement otherwise had the “indicia of reliability.”² Reliability could be inferred in two ways. The first was to fall within any hearsay exception; the second was “a showing of particularized guarantees of trustworthiness.”³ Practically, the Roberts test meant two things. First, if a statement fell into a hearsay rule, there was no need for independent Sixth Amendment analysis. Second, without a clear standard of what would constitute a “showing of particularized guarantees of trustworthiness,” judges had a wide degree of latitude to determine for themselves which statements were excepted from the Sixth Amendment.  

Justice Scalia, writing in Crawford, decried the Roberts test for, among other things, being too narrow. “[I]t admits statements that do consist of ex parte testimony upon a mere finding of reliability.”⁴ Further, he criticized the use of reliability as “an amorphous, if not entirely subjective concept. There are countless factors bearing on whether a statement is reliable.”⁵ Some jurisdictions would use balancing tests as high as eight or nine factors⁶ and judges would “attach the same significance to the opposite facts.”⁷ In deciding the Crawford opinion, the Court would no longer “leave the Sixth Amendment’s protection to the vagaries of the rules of evidence, much less to the amorphous notions of reliability.”⁸ 

The Court imposed a new rule that centered on the testimonial nature of a statement and the chance for prior cross-examination. The Sixth Amendment would not allow into the record “testimonial statements of a witness who did not appear at trial unless he was unavailable to testify, and the defendant had a prior opportunity for cross-examination.”⁹ The text of the Sixth Amendment “does not suggest any open-ended exceptions” to this rule.¹⁰ 

Crawford was seen as a revolutionary moment in Sixth Amendment interpretation.¹¹ However, what the Court hoped would be a bright-line rule to redefine the Sixth Amendment fell victim to the same “vagaries” the Court accused the Roberts test of encouraging. This is because the Court “[left] for another day any effort to spell out a comprehensive definition of testimonial.” The new rule the Court imposed soon became “notoriously ambiguous” itself.¹²  

It is Crawford’s failure to define “testimonial” that seeded ambiguity in Sixth Amendment analysis. The Court threw out the amorphous “reliability” test of Roberts in favor of another amorphous “testimonial” test. From the very first case in the Crawford line, there was a large, undefined exception to the requirement of subjecting testimony to the crucible of cross-examination.  

2. The “Primary Purpose Test” Compromises the Sixth Amendment Right as Understood in Crawford. 

The blanks left in the Crawford analysis were soon filled by subsequent cases. Davis v. Washington brought with it the “Primary Purpose Test” whereby statements made to the police are not testimonial if made “under circumstances objectively indicating that the primary purpose of the interrogation is to establish police assistance to an ongoing emergency.”¹³ Michigan v. Bryant expanded this to statements made to police about past events when the circumstances indicate that a public emergency is still ongoing.¹⁴ What little the Primary Purpose Test does tell us about testimonial statements betrays the reasoning of Crawford. 

The “Primary Purpose Test” betrays the reasoning in Crawford because, first, it allows the jury to hear testimonial evidence not checked by cross-examination, and second, it replaces cross-examination with a judge’s totality determination. Crawford stood for the principle that testimonial statements must not be admitted unless the declarant is unavailable and the defendant had prior opportunity for cross-examination; to allow otherwise would violate the Sixth Amendment. The Primary Purpose Test defines nontestimonial in a way that includes testimonial statements. Even if a statement is unambiguously given to the police with the intention that it will be used as testimony against a defendant, the Primary Purpose Test holds that it is nontestimonial as long as testimony was not the main reason for which it was given. Thus, contrary to what Crawford claimed, testimonial statements can now be given to the jury absent cross-examination. The Crawford Court criticized Roberts for admitting ex parte statements simply because they were reliable.¹⁵ The Primary Purpose Test now expressly allows the admission of testimonial, ex parte statements simply because testimony is not the primary reason the statements are given.  

The method of determining what is nontestimonial similarly betrays Crawford. Crawford found that “admitting statements deemed reliable by a judge is fundamentally at odds with the right of confrontation” because the Sixth Amendment demands reliability to be seen through cross-examination.¹⁶ Though the Primary Purpose Test is not aimed directly at determining reliability, its requirement for judges to weigh and admit statements without cross-examination is reminiscent of the Roberts test. This similarity is only exacerbated by the factors the judge can look at to weigh statements. In Davis, Scalia allowed judges to consider what the “reasonable listener” would believe the statements are for.¹⁷ Bryant allows judges to consider factors like the informality of questions, the circumstances of the interrogation, and the time since the accident.¹⁸ All of these are reminiscent of the multi-factor balancing tests used in the Roberts era, and Scalia notes as much in his Bryant dissent.¹⁹  

One could argue that the Primary Purpose Test fits properly within the Crawford framework because Crawford clearly left open an exception for nontestimonial statements and the Primary Purpose Test outlines that exception. This argument fails because the Primary Purpose Test takes advantage of Crawford’s failure to define “testimonial” in a way that undermines Crawford’s purpose. The Crawford test was supposed to prevent testimonial statements from being heard by the jury absent cross-examination and unavailability. In defining nontestimonial in a way that allows the jury to hear testimonial statements without cross-examination, the Primary Purpose Test does not square with Crawford’s principles.  

3. The Nontestimonial Exceptions Closely Resemble the Reliability Exceptions in Ohio v. Roberts. 

The nontestimonial exceptions provided in the cases subsequent²⁰ Crawford do not provide a sufficient explanation as to why they should be treated differently than Crawford. In Davis, for example, statements made seeking help during an ongoing emergency were admitted into evidence.²¹ This is because a 911 call is “not designed primarily to ‘establish or prove’ some past fact, but to describe current circumstances.”²² This may be true, but why a “primary purpose” being directed at an emergency should affect the Sixth Amendment differently here than in Crawford is not directly stated by the Court. Both cases involve statements with testimonial intent entered against a defendant.  

The Davis Court does not properly distinguish Crawford. The Court notes the difference between the 911 call in Davis and the interrogation in Crawford.²³ The Crawford opinion, however, says that an interrogation is the minimum level, not the only level, that the Confrontation Clause protects.²⁴ Thus, it is inappropriate to distinguish a police call from an interrogation and declare that this is sufficient to warrant an exception to confrontation under Crawford. 

The exceptions of Crawford’s progenies retreat to the same boundaries of the Roberts test, where falling under a hearsay exception or having sufficient guarantees of trustworthiness would except a statement from the Confrontation Clause.²⁵ The emergency phone call in Davis reflects an excited utterance.²⁶ This is supported by the multiple factors the Court considers—the time since the event and what the average listener would think was happening.²⁷ Similarly, the statements taken in Bryant could be justified under either the excited utterance exception or the dying declaration exception.²⁸ In 2022, another post-Crawford case, Williams v. Illinois admitted a blind-tested DNA report without cross-examining the expert who prepared it because “the primary purpose … was not to accuse petitioner … but to catch a dangerous rapist who was still at large.”²⁹ The Court took assurance in the fact that this was allowed under rule 703 of the Federal Rules of Evidence.³⁰ This is in contrast to Crawford’s criticism of the Roberts test resting on the “vagaries of the rules of evidence.”³¹ The cases establishing the exceptions after Crawford have used the Primary Purpose Test to create outcomes similar to the Roberts test.³² 

The Court in Crawford acknowledged that many cases decided under the testimonial analysis would overlap with the outcomes of the Roberts analysis.³³ However, the reason Crawford adopted a testimonial test is because the methodology of the Roberts test was incorrect.³⁴ The Primary Purpose Test too closely returns to the methodology of Roberts in the factors that it weighs and the discretion it gives to the judge. This has resulted in exceptions that mimic that of a Roberts analysis.  

4. Conclusion. 

Twenty years ago, Crawford pronounced a new, ambiguous “testimonial” standard to use in Sixth Amendment jurisprudence. Today, that standard has come to adopt many of the characteristics it was meant to replace.   

*J.D. Candidate at the University of Saint Thomas School of Law, Minnesota.

1. U.S. Const. amend. XI.   ↩︎
2. 448 U.S. 56, 66 (1980). ↩︎
3. Id. ↩︎
4. Crawford v. Washington, 541 U.S. 36, 60 (2004) (emphasis in original). ↩︎
5. Id. at 63. ↩︎
6. Id. ↩︎
7. Id. ↩︎
8. Id. at 61. ↩︎
9. Id. at 53–54. ↩︎
10. Id. at 54. ↩︎
11. Josephine Ross, Crawford’s Short-Lived Revolution: How Davis v. Washington Reins in Crawford’s Reach, 83 N.D. L. Rev. 387, 387 (2007). ↩︎
12. Id. at 388. ↩︎
13. Davis v. Washington, 547 U.S. 813, 822 (2006). ↩︎
14. Michigan v. Bryant, 562 U.S. 344, 373 (2011). ↩︎
15. Crawford, 541 U.S. at 60. ↩︎
16. Crawford, 541 U.S. at 61. ↩︎
17. Davis, 547 U.S. at 827. ↩︎
18. See Bryant, 562 U.S. 344, 371–373. ↩︎
19. Bryant, 562 U.S. at 393 (Scalia, J.,  dissenting). ↩︎
20. This is distinguished from the exceptions provided in Crawford itself: statements made to an acquaintance in passing and statements entered in a forfeiture by wrongdoing. Crawford, 541 U.S. at 51, 60. ↩︎
21. Davis, 547 U.S. at 827. ↩︎
22. Id. ↩︎
23. Id. at 826. ↩︎
24. Crawford, 541 U.S. at 68. ↩︎
25. Roberts, 448 U.S. at 66. ↩︎
26. Davis, 547 U.S. at 821. ↩︎
27. See id. at 828–830. ↩︎
28. Bryant, 562 U.S. at 351, n. 1. ↩︎
29. Erin Sheley, The Dignitary Confrontation Clause, 97 Wash. L. Rev. 207, 221 (2022). ↩︎
30. Id. A plurality did note that this was not meant to justify a “backdoor” use of testimonial reports Id. ↩︎
31. Crawford, 541 U.S. at 61. ↩︎
32. Outcomes of the two tests are certainly not always the same. When a statement is made to someone not involved in law enforcement, for example, there is rarely a question under the Crawford analysis about whether it was intended for litigation. Under the Roberts analysis, though, there would be a question of its reliability. This does not change the fact that, in some circumstances, the Crawford standard has adopted exceptions that mimic those of the exceptions in Roberts. ↩︎
33. Crawford, 541 U.S. at 57. (noting that case law has been “largely consistent” with the principles of the Confrontation Clause). ↩︎
34. Id. at 60. ↩︎