Prosecutorial Use of Forensic Science at Trial: When Is a Lab Report Testimonial?

University of Minnesota Law School

Scholarship Repository

Minnesota Law Review

2009

Prosecutorial Use of Forensic Science at Trial:

When Is a Lab Report Testimonial?

Joe Bourne

Follow this and additional works at: h7ps://scholarship.law.umn.edu/mlr

Part of the Law Commons

6is Article is brought to you for free and open access by the University of Minnesota Law School. It has been accepted for inclusion in Minnesota Law

Review collection by an authorized administrator of the Scholarship Repository. For more information, please contact [email protected].

Recommended Citation

Bourne, Joe, "Prosecutorial Use of Forensic Science at Trial: When Is a Lab Report Testimonial?" (2009). Minnesota Law Review. 540.

h7ps://scholarship.law.umn.edu/mlr/540

1058

Note

Prosecutorial Use of Forensic Science at Trial:

When Is a Lab Report Testimonial?

Joe Bourne∗

Scientific evidence creates a unique opportunity for prose-

cutorial abuse.

Investigations of police crime laboratories have

revealed widespread error and sloppiness—even corruption.

Even well-intentioned, conscientious crime-laboratory workers

may be subject to “subconscious . . . pro-prosecution bias” and

“confirmation bias,” which can color their interpretations or re-

porting of results.

The reports these forensic scientists gener-

ate play a powerful role in criminal prosecutions. According to

the “CSI effect” theory, juror determinations increasingly rely

on misunderstandings of the nature of scientific evidence

spawned by forensic science television shows such as CSI:

Crime Scene Investigation, which “portray[] forensic science as

high-tech magic.”

At the same time, the use of scientific evi-

dence at trials has increased significantly in recent years.

∗ J.D. Candidate 2009, University of Minnesota Law School; B.A. 2005,

Emory University. I would like to thank Professor Stephen Cribari for assis-

tance in formulating a topic and for feedback and encouragement along the

way. I also thank the board and staff of Volumes 92 and 93 of the Minnesota

Law Review, in particular Nick Smith, Jenni Vainik, Mike Schoepf, Liz Borer,

and Kyle Hawkins for their time, commentary, and support.

1. Cf. Pamela R. Metzger, Cheating the Constitution, 59 V

AND. L. REV.

475, 499 (2006) (describing the case of a West Virginia state trooper and foren-

sic serologist who “engage[d] in long-term, systematic, and deliberate falsifica-

tion of evidence in criminal cases”).

2. See generally id. at 491–500 (discussing sources of error).

3. Id. at 496–97.

4. N.J. Schweitzer & Michael J. Saks, The CSI Effect: Popular Fiction

About Forensic Science Affects the Public’s Expectations About Real Forensic

Science, 47

JURIMETRICS J. 357, 358 (2007). For a detailed treatment of the

CSI effect, see generally Tom R. Tyler, Viewing CSI and the Threshold of

Guilt: Managing Truth and Justice in Reality and Fiction, 115 Y

ALE L.J. 1050

(2006) (discussing the evidence for and against a “CSI effect” and exploring

the potential implications of such an effect). Jurors, disappointed by evidence

failing to live up to the standards developed from shows like CSI, could be

more likely to find reasonable doubt. Id. at 1052. Or, they could be more likely

2009] FORENSIC SCIENCE AT TRIAL 1059

Meanwhile, constitutional law pertaining to the use of this

science is in a state of turmoil. In 2004, the Supreme Court in

Crawford v. Washington

dramatically reworked its Confronta-

tion Clause

analysis, holding that testimonial hearsay is in-

admissible unless the witness is unavailable and the defendant

had a prior opportunity to cross-examine that witness.

The

Court expressly did not define “testimonial” beyond setting a

minimum baseline;

it also failed to explicitly delineate a clear

analytical framework.

The Court again failed to set forth a

comprehensive definition of “testimonial” in Davis v. Washing-

ton.

Thus, while the confrontation right would offer a defen-

dant meaningful protection from an analyst’s scientific report if

that report is testimonial, the Court did not clearly set forth the

concept of testimoniality.

Because both Crawford and Davis were decided in the con-

text of statements made to police officers or agents of the police

by persons who had directly witnessed the crimes at issue,

lower courts have had to fend for themselves in attempting to

figure out what to do with scientific evidence such as laboratory

reports and coroner’s reports. Under cases such as Ohio v. Ro-

to find the defendant guilty due to the combination of an overbelief in the

probative value of science and the well-documented psychological desire to see

wrongdoers punished. See id. at 1063–76.

5. See John M. Spires, Note, Testimonial or Nontestimonial? The Admis-

sibility of Forensic Evidence After Crawford v. Washington, 94 K

Y. L.J. 187,

187 (2005).

6. Crawford v. Washington, 541 U.S. 36 (2004).

7. U.S.

CONST. amend. VI.

8. Crawford, 541 U.S. at 68.

9. See id. (“[Testimonial] applies at a minimum to prior testimony at a

preliminary hearing, before a grand jury, or at a former trial; and to police in-

terrogations.”).

10. See id. (“We leave for another day any effort to spell out a comprehen-

sive definition of ‘testimonial.’”).

11. See Davis v. Washington, 547 U.S. 813, 813 (2006) (noting that the

Court was not “attempting to produce an exhaustive classification of all con-

ceivable statements [that are] . . . testimonial”).

12. See id.; cf. California v. Green, 399 U.S. 149, 192 (1970) (Brennan, J.,

dissenting) (“There is no way to test the recollection and sift the conscience of

a witness regarding the facts of an alleged offense if he is unwilling or unable

to be questioned about them; defense counsel cannot probe the story of a silent

witness and attempt to expose facts that qualify or discredit it.”).

13. Crawford, 541 U.S. at 38 (“At [the defendant’s] trial, the State played

for the jury [his wife’s] tape-recorded statement to the police . . . .”); Davis, 547

U.S. at 813 (“[A] 911 operator ascertained from [the victim] that she had been

assaulted by her former boyfriend . . . [and] the court admitted the 911 record-

ing . . . .”).

1060 MINNESOTA LAW REVIEW [93:1058

berts,

which considered factors such as the reliability of the

evidence and articulated traditional constitutional balancing

tests, such evidence typically withstood Confrontation Clause

scrutiny.

State courts and the lower federal courts are cur-

rently struggling to determine if, and explain why, they are

dealing with testimonial or nontestimonial evidence.

Courts

are split on how to handle these cases, not due to distinguisha-

ble factual circumstances, but due to widely variant under-

standings of what the Supreme Court meant by “testimonial.”

Part I of this Note explicates the Supreme Court’s Confron-

tation Clause doctrine and the shift from Roberts to Crawford

and Davis. Part II looks at the scientific-evidence question and

examines how the lower courts have addressed it. This Part al-

so considers, and rejects, the developing case law holding that

scientific evidence, if the product of a machine-based process, is

not a statement of any person and therefore cannot constitute

testimonial hearsay. Part III offers an analysis for courts to fol-

low. In order to be faithful to Crawford, courts should apply a

bright-line rule that covers the most common types of cases: lab

reports prepared by or for the police to further the investigation

or prosecution of a suspected crime are per se testimonial. Un-

der the limited facts falling outside of such a rule—including

cases involving autopsy reports, which present somewhat dif-

ferent constitutional considerations—courts should undertake

fact-intensive, case-by-case inquiries, considering the criteria

laid out more fully in Part III.

I. SUPREME COURT CONFRONTATION CLAUSE

JURISPRUDENCE

The Sixth Amendment to the U.S. Constitution provides:

“In all criminal prosecutions, the accused shall enjoy the

right . . . to be confronted with the witnesses against

him . .

. .”

Through the Fourteenth Amendment, the Confron-

tation Clause applies against the states.

Yet what it guaran-

14. Ohio v. Roberts, 448 U.S. 56 (1980).

15. Id. at 66 (“[A] statement is admissible only if it bears adequate ‘indicia

of reliability.’”).

16. See Cyrus P.W. Rieck, Note, How to Deal with Laboratory Reports un-

der Crawford v. Washington: A Question with No Good Answer, 62 U.

MIAMI L.

REV. 839, 907 (2008) (explaining that courts are “desperate for an answer”).

17. See id. (noting that courts are reaching different conclusions on “es-

sentially the same facts”).

18. U.S.

CONST. amend. VI.

19. Pointer v. Texas, 380 U.S. 400, 406 (1965).

2009] FORENSIC SCIENCE AT TRIAL 1061

tees in practice has been the subject of considerable debate, in

part, no doubt, because as the second Justice Harlan once

noted, the text on its face leaves itself open to widely variant

interpretations.

Judicial understanding of the Confrontation

Clause has changed significantly through the years. This Note

first considers the long-dominant Roberts paradigm, followed

by the sudden shift to Crawford analysis and the Davis Court’s

addition to Crawford’s guidance.

A. R

OBERTS RELIABILITY AND BALANCING

Following precedent, in 1980 the Roberts Court held that a

defendant’s confrontation right was not violated by the prose-

cution’s use of hearsay statements made by an unavailable

witness at a preliminary hearing at which the defendant func-

tionally cross-examined the witness.

Apart from this narrow

holding, the Court’s opinion was important for its reasoning,

which came to stand for two doctrinal propositions.

First, Roberts stood for the proposition that the Confronta-

tion Clause was subject to a balancing test. Acknowledging in

dicta that the Clause “reflects a preference for face-to-face con-

frontation at trial,”

the Court explained that “competing in-

terests, if ‘closely examined,’ may warrant dispensing with con-

frontation at trial.”

While the language of close examination

suggests a difficult threshold for the government to pass, the

Court noted that the government always has “strong” compet-

ing interests in two overarching areas: “effective law enforce-

ment” and “development and precise formulation of the rules of

evidence applicable in criminal proceedings.”

This could be

read as a signal to lower courts that the defendant’s confronta-

tion right is, on balance, really not all that weighty.

Second, the Court concluded that, for Confrontation Clause

purposes, admissibility of the hearsay testimony of an unavail-

able declarant depended on the reliability of that hearsay.

20. California v. Green, 399 U.S. 149, 175 (1970) (Harlan, J., concurring

in the judgment) (“[T]he clause may be read to confer nothing more than a

right to meet face to face all those who appear and give evidence at trial. . . .

[But] the clause is equally susceptible of being interpreted as a blanket prohi-

bition on the use of any hearsay testimony.”).

21. Ohio v. Roberts, 448 U.S. 56, 70–73 (1980).

22. Id. at 63.

23. Id. at 64 (quoting Chambers v. Mississippi, 410 U.S. 284, 295 (1973)).

24. Id.

25. See id. at 66 (noting that hearsay of an unavailable witness may be

admitted “if it bears adequate ‘indicia of reliability’”).

1062 MINNESOTA LAW REVIEW [93:1058

a per se matter, hearsay was deemed to exhibit sufficient “indi-

cia of reliability” if it matched a “firmly rooted hearsay excep-

tion.”

Otherwise the prosecution would have to show that the

hearsay bore “particularized guarantees of trustworthiness” to

get it admitted at trial against the defendant without allowing

the defendant a right to confront the declarant.

In other

words, the Roberts Court construed the Clause to procedurally

protect criminal defendants from substantively unreliable ac-

cusatory evidence.

B. R

EJECTING ROBERTS-TYPE ANALYSIS: CRAWFORD’S

“TESTIMONIAL” FRAMEWORK

In 2004 the Supreme Court overruled the Roberts test in

Crawford v. Washington, finding reliability analysis “so unpre-

dictable that it fails to provide meaningful protection from even

core confrontation violations.”

The Court also rejected the use

of a balancing test where a defendant’s confrontation right was

violated.

Crawford instead issued a bright-line command: tes-

timonial hearsay is only admissible if the witness is unavaila-

ble and the accused had “a prior opportunity for cross-

examination.”

Since it was not necessary on the facts of the

case to do so, the Court opted not to “spell out a comprehensive

definition of ‘testimonial.’”

Consequently, the considerations

that were important to the Crawford Court’s discussion of tes-

timonial hearsay must be looked to for any future Confronta-

tion Clause analysis.

The Court traced the common law origin of the confronta-

tion right back to the famous 1603 treason trial of Sir Walter

Raleigh, in which Raleigh was sentenced to death based on the

out-of-court, co-conspiratorial confession of Lord Cobham.

Ra-

leigh argued that Cobham was lying, and demanded that Cob-

26. Id.

27. Id.

28. But cf. Crawford v. Washington, 541 U.S. 36, 61 (2004) (explaining

that the framers created a “procedural rather than substantive” constitutional

protection in order “to ensure reliability of evidence,” and criticizing the Ro-

berts test for allowing juries to hear evidence untested by adversarial cross-

examination where a judge found the evidence to be substantively reliable).

29. Id. at 63.

30. See id. at 67–68 (“By replacing categorical constitutional guarantees

with open-ended balancing tests, we do violence to their design.”).

31. Id. at 68.

32. Id.

33. Id. at 43–44.

2009] FORENSIC SCIENCE AT TRIAL 1063

ham tell his story to Raleigh and the jury; the English court re-

fused to recognize such a confrontation right.

The First Con-

gress had in mind this type of abuse when it included the Con-

frontation Clause in the Sixth Amendment.

Crawford

establishes that the “principal evil” at which the Clause is

aimed is the “use of ex parte examinations as evidence against

the accused.”

The Court expressly rejected the theory that the constitu-

tional right to cross-examination might be coterminous with

hearsay rules:

An off-hand, overheard remark might be unreliable evidence and thus

a good candidate for exclusion under hearsay rules, but it bears little

resemblance to the civil-law abuses the Confrontation Clause tar-

geted. On the other hand, ex parte examinations might sometimes be

admissible under modern hearsay rules, but the Framers certainly

would not have condoned them.

Thus, Confrontation Clause analysis should not look to

hearsay law.

In dicta, however, the Court suggested that two

hearsay exceptions are particularly noteworthy.

First, be-

cause the “dying declarations” hearsay exception existed at

common law, it may have been incorporated into the Sixth

Amendment, even where testimonial dying declarations are at

issue.

Second, the “business records” exception was part of the

common law by 1791 when the Sixth Amendment was passed,

but significantly, these records “by their nature were not testi-

monial.”

The Court’s consideration of which exceptions were

allowed in criminal cases by 1791, and its general emphasis on

the historical backdrop to the Confrontation Clause, indicates

that courts should look to history.

34. Id.

35. See id. at 49 (noting how the First Congress responded to an Antifede-

ralist writing that “criticized the use of ‘written evidence’” without the “cross

examining [of] witnesses”).

36. Id. at 50.

37. Id. at 51.

38. See id. at 61 (“Where testimonial statements are involved, we do not

think the framers meant to leave the Sixth Amendment’s protection to the va-

garies of the rules of evidence, much less to amorphous notions of ‘reliabili-

ty.’”).

39. Id. at 56 & n.6.

40. Id. at 56 n.6. The Court noted that “[i]f this exception must be ac-

cepted on historical grounds, it is sui generis.” Id.

41. Id. at 56.

42. See id. at 43–50. Justice Scalia’s opinion for the seven-Justice majority

was a decidedly originalist opinion. Chief Justice Rehnquist, writing for him-

self and Justice O’Connor, while concurring in the overall judgment, dissented

1064 MINNESOTA LAW REVIEW [93:1058

Although the Court did not spell out a complete definition

of “testimonial,” it set a minimum baseline that includes “prior

testimony at a preliminary hearing, before a grand jury, or at a

former trial; and . . . police interrogations.”

Additionally, the

Court explained that statements made to police officers during

the course of interrogations are testimonial “under even a nar-

row standard” because they “bear a striking resemblance to ex-

aminations by justices of the peace in England,” making clear

that a statement is testimonial if it is the contemporary analo-

gue of an abuse about which the First Congress was con-

cerned.

The last piece of definitional guidance the Court of-

fered was three potential articulations of testimonial hearsay:

(1) “ex parte in-court testimony or its functional equivalent”; (2)

“formalized testimonial materials, such as affidavits, deposi-

tions, prior testimony, or confessions”; and (3) “statements that

were made under circumstances which would lead an objective

witness reasonably to believe that the statement would be

available for use at a later trial.”

Because it was not neces-

sary for the Court to adopt any of these various formulations, it

did not expressly do so—but notably, it also did not reject any

of them.

Lending support to the third, broadest formulation,

the Court noted that “[i]nvolvement of government officers in

the production of testimony with an eye toward trial presents

unique potential for prosecutorial abuse—a fact borne out time

and again throughout a history with which the Framers were

keenly familiar.”

In sum, the Confrontation Clause bars the admission of

testimonial hearsay in lieu of in-court confrontation either if

the declarant is available or if the defendant lacked a prior op-

portunity for cross-examination of an unavailable declarant.

In deciding whether a statement is testimonial, a court should

not look to modern hearsay rules, but rather to the historical

context of the Sixth Amendment while considering what would

be the contemporary analogues to the Framers’ concerns.

And

on prudential and stare decisis grounds from the Court’s decision to overrule

Roberts. See id. at 69 (Rehnquist, C.J., concurring in the judgment).

43. Id. at 68 (majority opinion).

44. See id. at 52.

45. Id. at 51–52.

46. See id. at 52 (“These formulations all share a common nucleus and

then define the clause’s coverage at various levels of abstraction around it.”).

47. Id. at 56 n.7.

48. Id. at 68.

49. Id. at 67–68 (discussing how it is unlikely that the framers would be

2009] FORENSIC SCIENCE AT TRIAL 1065

anything which looks too much like prosecutorially elicited ex

parte testimony will raise Confrontation Clause concerns.

C. T

HE DAVIS TIMING PRONG: ONGOING EMERGENCIES VS.

INVESTIGATION OF PAST CRIMES

Two years later, the Court in Davis again opted not to “at-

tempt[] to produce an exhaustive classification” of testimoniali-

ty.

It did, however, articulate an investigatory purpose test:

Statements are nontestimonial when made in the course of police in-

terrogation under circumstances objectively indicating that the pri-

mary purpose of the interrogation is to enable police assistance to

meet an ongoing emergency. They are testimonial when the circums-

tances objectively indicate that there is no such ongoing emergency,

and that the primary purpose of the interrogation is to establish or

prove past events potentially relevant to later criminal prosecution.

The Court was careful to note that the totality of the cir-

cumstances test it articulated was no broader than necessary to

resolve the cases before it.

Statements not made in response

to police interrogation may also be testimonial.

The Court al-

so noted that “even when interrogation exists,” ultimately “the

declarant’s statements, not the interrogator’s questions,” are

controlling for Confrontation Clause analysis.

In determining whether statements made during a 911 call

and a house call by the police in response to domestic violence

were testimonial hearsay, the Court considered the following

satisfied relying on modern “reliability factors”).

50. Cf. id. at 50 (“[T]he principal evil at which the Confrontation Clause

was directed was the civil-law mode of criminal procedure, and particularly its

use of ex parte examinations as evidence against the accused.”).

51. Davis v. Washington, 547 U.S. 813, 822 (2006).

52. Id. at 822. The Court was apparently unconcerned with the subjective

intention of the police officer doing the investigating or interrogating. Justice

Thomas, in a separate opinion, articulated the reason behind the objective na-

ture of the Court’s test as follows:

The Court’s repeated invocation of the word “objectiv[e]” to describe

its test . . . suggests that the Court may not mean to reference pur-

pose at all, but instead to inquire into the function served by the in-

terrogation. Certainly such a test would avoid the pitfalls that have

led us repeatedly to reject tests dependent on the subjective inten-

tions of police officers.

Id. at 839 (Thomas, J., concurring in the judgement and dissenting in part).

53. See id. at 822 n.1 (majority opinion) (explaining that the holding ap-

plies specifically to interrogations because the statements of the current case

were the “products of interrogations”).

54. Id. (“This is not to imply, however, that statements made in the ab-

sence of any interrogation are necessarily nontestimonial.”).

55. Id.

1066 MINNESOTA LAW REVIEW [93:1058

factors: contemporaneousness; the existence of an “ongoing

emergency”; “the nature of what was asked and answered”; and

formality or solemnity.

In the case of a 911 call where the dec-

larant identified her assailant while she faced an ongoing

emergency, the operator’s questions were objectively designed

to “elicit[] statements . . . necessary to be able to resolve the

present emergency,” and her answers were “frantic” and “pro-

vided over the phone” in a potentially dangerous situation, the

Court held that her statements were nontestimonial.

Essen-

tially, because “[n]o ‘witness’ goes into court to proclaim an

emergency and seek help,” her “emergency statement” was in-

sufficiently similar to the type of ex parte testimony on which

Raleigh was convicted to create a Sixth Amendment problem.

On different facts the Court reached the opposite conclu-

sion.

Where the police arrived at the declarant’s home, were

told by her “that things were fine,” she was not presently in

danger, and the officers questioned her in a separate room from

her husband for the purpose of obtaining better information

about the crime that had occurred, the Court determined that

the declarant’s statements were “an obvious substitute for live

testimony, because they do precisely what a witness does on di-

rect examination,” thus rendering the statements testimonial.

Such an inquiry is plainly fact-intensive. Further complicating

a Davis inquiry is the Court’s encouragement of the use of in

limine procedure to redact the testimonial portions of state-

ments that began as nontestimonial statements made in re-

sponse to an emergency.

Davis does not contemplate scientific evidence. However,

the analytical factors employed by the Court can be useful to

lower courts attempting to determine whether such evidence is

testimonial.

56. Id. at 827.

57. Id. at 814, 827.

58. Id. at 828.

59. Two different state supreme court cases were consolidated in Davis v.

Washington and thus the Court was able to apply its interpretation of the Con-

frontation Clause to two entirely different sets of facts. Id. at 817–21.

60. Id. at 830.

61. Id. at 829.

2009] FORENSIC SCIENCE AT TRIAL 1067

II. SCIENCE MEETS THE CONFRONTATION CLAUSE IN

THE LOWER COURTS

Cases involving scientific evidence commonly occur where

seized substances are tested for a composition analysis, where

blood samples of suspected intoxicated drivers are tested for al-

cohol and narcotics content, where DNA analysis is performed

in rape or homicide cases, and where medical examiners per-

form autopsies.

In such cases, Roberts reliability analysis was

fairly easy to perform. But courts have struggled to apply the

new testimoniality analysis prescribed by Crawford and Davis.

This Part examines the cases that consider evidence such as

laboratory reports to be per se testimonial, the cases that es-

sentially render ad hoc decisions, the post-Crawford cases that

are de facto Roberts analyses, and the machine-generated-

statements rationale currently being developed in at least three

of the circuits.

A. C

OURTS HOLDING THAT A LABORATORY REPORT IS ALWAYS

TESTIMONIAL

This Note now considers the various cases finding labora-

tory reports to be per se testimonial within the meaning of

Crawford and Davis. Typical of these cases is State v. Caul-

field,

in which the Minnesota Supreme Court held that a po-

lice laboratory report identifying a seized substance as cocaine

constituted testimonial hearsay where the analyst who pre-

pared the report did not testify at trial.

The court found that

the laboratory report fit under “each of the three generic de-

scriptions offered by the Supreme Court in Crawford” of what

“testimonial” might mean.

First, it “functioned as the equiva-

lent of testimony” by identifying the cocaine as cocaine.

Second, the report was an affidavit—a formalized testimonial

material specifically identified in Crawford.

Third, it “was

clearly prepared for litigation.”

The court considered the

last—preparation for use at trial—to be the “critical determina-

62. See, e.g., Prater v. State, 820 S.W.2d 429, 430 (Ark. 1991); State v.

Caulfield, 722 N.W.2d 304, 306–07 (Minn. 2006); State v. Kent, 918 A.2d 626,

628–29 (N.J. Super. Ct. App. Div. 2007).

63. Caulfield, 722 N.W.2d at 304.

64. Id. at 306–07.

65. Id. at 309.

66. Id.

67. Id.; Crawford v. Washinton, 541 U.S. 36, 51 (2004).

68. Caulfield, 722 N.W.2d at 309.

1068 MINNESOTA LAW REVIEW [93:1058

tive factor” in its analysis.

Anticipating Davis, the Minnesota

Supreme Court relied heavily on the circumstances surround-

ing the generation of the “statement” identifying the seized

substance, concluding that the sole purpose of generating the

laboratory report was to have it available for use at trial.

The

court’s determination was based on the facts that the police had

already arrested Caulfield and that they had preliminarily de-

termined that the seized substance was cocaine.

A New Jersey appellate court in State v. Kent

reached a

similar conclusion to Caulfield applying both Crawford and

Davis.

Adam J. Kent had crashed his car, and when a police

officer arrived at the scene at 1:40 a.m., he noted that Kent

smelled of alcohol.

Kent acknowledged that he had been

drinking.

The suspect was taken to a hospital, where the of-

ficer asked a nurse to draw a sample of his blood, which was

then taken to a police laboratory for toxicology and gas chroma-

tography analysis.

The laboratory report indicated that the

defendant’s blood alcohol content was above the legal limit.

While acknowledging that, under the New Jersey Rules of Evi-

dence, a police chemist’s laboratory report is a business record,

the Kent court nonetheless found that admission of the chem-

ist’s report where the chemist did not testify at trial violated

the defendant’s Sixth Amendment right to cross-examine him.

Performing a Davis analysis, the court concluded that the de-

fendant’s blood was not analyzed to deal with an ongoing emer-

gency, and that the primary purpose of the report was to prove

at trial the past event that the defendant’s blood alcohol con-

centration was high enough to expose him to criminal liabili-

ty.

The court also concluded that under Crawford the nurse’s

signed blood sample certification was the functional equivalent

69. See id. (“We have said the critical determinative factor assessing

whether a statement is testimonial is whether it was prepared for litigation.”).

70. See id. (“The . . . report was clearly prepared for litigation.”).

71. Id.

72. State v. Kent, 918 A.2d 626 (N.J. Super. Ct. App. Div. 2007).

73. Cf. id. at 639–40 (citing Caulfield with approval).

74. Id. at 628–29.

75. Id. at 629.

76. Id. at 629–31.

77. Id. at 631.

78. Id. at 636–40.

79. Id. at 637.

2009] FORENSIC SCIENCE AT TRIAL 1069

of an affidavit because the nurse knew that falsifying its con-

text would be unlawful, thereby rendering it testimonial.

Courts in other jurisdictions have reached the same con-

clusion as the Minnesota and New Jersey courts did in Caul-

field and Kent. The distilled analysis of these decisions consists

of two major, interrelated points. First, laboratory reports do

not fall within the business-records exception recognized in

1791, that is, the inherently nontestimonial sort of business

record contemplated by Justice Scalia’s dictum in Crawford.

As the District of Columbia Court of Appeals explained: “Tradi-

tionally, the historical business-records exception did not en-

compass records prepared for use in litigation, let alone records

produced ex parte by government agents for later use in crimi-

nal prosecution.”

The court explained that Crawford makes

clear that the accused’s confrontation right cannot have been

diminished by expansions of the business-records exception un-

der modern evidence law.

Second, courts emphasize that these reports are being gen-

erated for the primary (or sole) purpose of facilitating criminal

prosecution, and that as such they are testimonial statements

implicating the Confrontation Clause.

The United States Ar-

my Court of Criminal Appeals provides a good example of this

reasoning in United States v. Williamson.

The defendant’s

marijuana was seized, tested, and subsequently identified by a

senior forensic chemist as marijuana.

The court concluded

that because “the ‘statement’ [was] a post-apprehension labora-

tory report, requested after local police arrested [the defen-

dant],” the statement—that scientific examination indicated

that the seized substance was an illegal drug—was necessarily

testimonial.

Courts have also reached this conclusion in the

80. See id. at 637–39.

81. See Crawford v. Washington, 541 U.S. 36, 56 (2004); see also Thomas

v. United States, 914 A.2d 1, 13 (D.C. 2006) (“As an historical matter, the ex-

ception in 1791 was a very narrow one.”).

82. Thomas, 914 A.2d at 13.

83. Id. at 27.

84. See, e.g., id. at 14 (“[B]ecause DEA chemist’s reports are created ex-

pressly for use in criminal prosecutions as a substitute for live testimony

against the accused, such reports are testimonial, whether or not they happen

to meet this jurisdiction’s definition of a business record.”).

85. United States v. Williamson, 65 M.J. 706 (A. Ct. Crim. App. 2007).

86. Id. at 707–10.

87. Id. at 717–18.

1070 MINNESOTA LAW REVIEW [93:1058

context of laboratory reports identifying seized substances,

tests of a rape victim’s blood alcohol content where her intoxi-

cation level affected her ability to consent,

and tests of the

suspect’s blood in drunk driving cases such as Kent.

B. S

CIENTIFIC EVIDENCE IS NONTESTIMONIAL—BUT ONLY

SOMETIMES

While many courts have conducted fairly fact-intensive in-

quiries, a rare occurrence in which the outcome truly seems de-

pendent on the unique facts of the case is United States v. Ma-

gyari.

In Magyari, the United States Court of Appeals for the

Armed Forces held that a laboratory report identifying a posi-

tive test for methamphetamine during the course of regular,

randomized urinalysis screening was nontestimonial.

The

court admitted the technician’s report under a business-records

hearsay exception.

It emphasized that the urinalysis was

“routine” and randomized, and that the “vast majority” of such

tests are negative for illegal substances.

In the Magyari

court’s view, the technicians simply “were not engaged in a law

enforcement function, a search for evidence in anticipation of

prosecution or trial. . . . Because the lab technicians were mere-

ly cataloging the results of routine tests, the technicians could

not reasonably expect their data entries would ‘bear testimony’

against Appellant at his court-martial.”

88. See, e.g., Hinojos-Mendoza v. People, 169 P.3d 662, 667 (Colo. 2007)

(en banc) (“There can be no serious dispute that the sole purpose of the report

was to analyze the substance found in Hinojos-Mendoza’s vehicle in anticipa-

tion of criminal prosecution.”); State v. Laturner, 163 P.3d 367, 376 (Kan. Ct.

App. 2007) (“The forensic scientist who prepared Laturner’s lab report was a

witness; the statements in her lab report were testimony; and she knew when

preparing her report that it would be used by the State at Laturner’s trial to

prove he committed the crime of possessing methamphetamine.”); State v.

March, 216 S.W.3d 663, 666 (Mo. 2007) (“A laboratory report, like this one,

that was prepared solely for prosecution to prove an element of the crime

charged is ‘testimonial’ because it bears all the characteristics of an ex parte

affidavit.”).

89. See People v. Rogers, 780 N.Y.S.2d 393, 397 (N.Y. App. Div. 2004)

(“Because the test was initiated by the prosecution and generated by the de-

sire to discover evidence against defendant, the results were testimonial.”).

90. See State v. Kent, 918 A.2d 626, 637–40 (N.J. Super. Ct. App. Div.

2007).

91. United States v. Magyari, 63 M.J. 123 (C.A.A.F. 2006).

92. Id. at 124–25.

93. Id. at 127; see M

IL. R. EVID. 803(6).

94. Magyari, 63 M.J. at 126.

95. Id. at 126–27.

2009] FORENSIC SCIENCE AT TRIAL 1071

In concluding that the technicians could not have reasona-

bly expected the results of their tests to be used for prosecution,

the court also stressed the importance of the following facts:

many people tested Magyari’s urine sample and recorded data

entries in his records; his sample was distinguishable only by

an anonymous number; and there was no reason to think that

anyone was pressured to reach a particular conclusion.

How-

ever, the court expressly rejected the theory that laboratory re-

ports are never testimonial.

It was quick to point out that la-

boratory reports or similar records might be prepared “at the

behest of law enforcement in anticipation of prosecution,”

which would likely compel a finding of testimoniality.

It con-

cluded merely that the relevant Crawford considerations were

not in play under the unique facts with which it was pre-

sented.

Importantly, the situation lacked both the potential

and the incentive for prosecutorial abuse.

C. N

ONTESTIMONIALITY AND THE TENDENCY TOWARD ROBERTS

ANALYSIS DRESSED IN CRAWFORD AND DAVIS CLOTHING

This Note now examines the line of cases holding that la-

boratory reports, and similar examples of scientific evidence,

are not testimonial. Five considerations run thematically

throughout these cases: the contemporaneous recordation of a

presently observable event; the fit of a business-records or simi-

lar hearsay exception; the objective facticity of the evidence; the

value of cross-examination for the defendant; and society’s

competing interests. Notably, such considerations are all de-

rived from a misreading of Davis, a misunderstanding of Craw-

ford, or a latent Roberts analysis.

1. Contemporaneous Recordation of Observations

An example of misplaced reliance on contemporaneousness

is the California Supreme Court’s opinion in People v. Geier.

100

Geier involved a DNA report in a rape-homicide case.

101

The

96. Id. at 127.

97. Id.

98. Id.

99. Cf. id. (“This conclusion is consistent with the Crawford Court’s policy

concerns that might arise where government officers are involved ‘in the pro-

duction of testimony with an eye toward trial’ and where there is ‘unique po-

tential for prosecutorial abuse’ and overreaching.” (quoting Crawford v. Wash-

ington, 541 U.S. 36, 56 n.7 (2004))).

100. People v. Geier, 161 P.3d 104 (Cal. 2007).

101. Id. at 110, 131.

1072 MINNESOTA LAW REVIEW [93:1058

court concluded that, based on Crawford and Davis, a DNA re-

port would be testimonial only if it “describes a past fact re-

lated to criminal activity.”

102

That is, under the court’s inter-

pretation of Davis, “the crucial point is whether the statement

represents the contemporaneous recordation of observable

events.”

103

That the DNA report was requested by the police

and the testing scientist could have anticipated its later use at

a criminal trial was more or less irrelevant, since the court

rendered these factors insufficient to implicate the Confronta-

tion Clause.

104

The Geier court’s application of its rule to the facts before it

was straightforward: the state’s DNA analyst recorded her ob-

servations at the time they occurred.

105

The things she wrote in

her report therefore “constitute a contemporaneous recordation

of observable events rather than the documentation of past

events.”

106

The court accordingly held that the DNA report was

nontestimonial.

107

Other jurisdictions have undertaken similar inquiries. The

Supreme Court of New Hampshire in State v. O’Maley

108

agreed with Geier’s focus on “contemporaneous recordation,”

proclaiming it a “crucial point.”

109

At issue in O’Maley was a la-

boratory report of a drunk driving suspect’s blood alcohol con-

tent where the workers who actually collected and tested

O’Maley’s blood did not testify at trial.

110

The court held that

the blood sample collection form filled out by the blood-drawing

technician “constituted the technician’s contemporaneous re-

cordation of observable events” and did not “describe any of the

defendant’s past conduct.”

111

Rather than continuing this line

of thought, the court disposed of the question of the analyst’s

report on the theory that the report was never offered into evi-

dence.

112

This strain of analysis is fatally flawed. The argument that

a laboratory report used to prove a DNA match to prosecute a

102. Id. at 138.

103. Id. at 140.

104. See id. at 139.

105. See id.

106. Id.

107. Id. at 140.

108. State v. O’Maley, 932 A.2d 1 (N.H. 2007).

109. Id. at 11 (quoting Geier, 161 P.3d at 140).

110. Id. at 4.

111. Id. at 13.

112. See id. at 13–14.

2009] FORENSIC SCIENCE AT TRIAL 1073

rape-homicide is not used to prove past events and is merely a

present-sense recording is conceptually specious.

113

But apart

from the argument’s conceptual incoherence, it is constitution-

ally incorrect—it is based on a substantial misreading of Davis.

Davis was a fact-intensive inquiry made in the context of a po-

lice response to domestic violence.

114

The Court focused on con-

temporaneousness to determine whether the circumstances

surrounding the statements made to the police objectively indi-

cated that the primary purpose of the interrogation was to re-

spond to an ongoing emergency or whether it was to gather

evidence pertaining to the past event (domestic violence) to be

proved at a subsequent criminal trial.

115

In considering the con-

temporaneousness of the Davis declarant’s statements to the

911 operator, the court in Geier ignored the existence of the on-

going emergency in Davis which the declarant was contempo-

raneously describing to the police.

116

Geier’s DNA analyst was

certainly not analyzing DNA, and at the same time recording

that analysis, in response to any ongoing emergency; and re-

moved from the requirement of an ongoing emergency, all re-

cording of testimonial hearsay statements would qualify as con-

temporaneous recording.

2. Crawford’s Business-Records Exception

The vast majority of cases finding scientific evidence non-

testimonial misread the language in Crawford suggesting that

business records are inherently nontestimonial. The Second

Circuit, for example, held in United States v. Feliz

117

that a

business record within the meaning of the Federal Rules of

Evidence is “fundamentally inconsistent” with the Supreme

Court’s idea of testimonial hearsay.

118

The Second Circuit’s ra-

tionale was that since business records by definition are not

made in anticipation of litigation and do not include observa-

tions of police officers, they necessarily do not raise the sorts of

concerns the First Congress had in mind when it authored the

113. Cf. State v. Kent, 918 A.2d 626, 637 (N.J. Super. Ct. App. Div. 2007)

(“Nor can it reasonably be argued that the ‘primary purpose’ of the lab certifi-

cate was anything other than to prove past events, specifically defendant’s

blood alcohol concentration, relevant to his DWI prosecution.”).

114. See Davis v. Washington, 547 U.S. 813, 822 (2006).

115. See id. at 827.

116. See People v. Geier, 161 P.3d 104, 138 (Cal. 2007).

117. United States v. Feliz, 467 F.3d 227 (2d Cir. 2006).

118. Id. at 233–34; see F

ED. R. EVID. 803(6).

1074 MINNESOTA LAW REVIEW [93:1058

Sixth Amendment.

119

In Feliz, the court dealt with the admis-

sibility of autopsy reports of nine murder victims.

120

It ac-

knowledged that a reasonable medical examiner should antic-

ipate that an autopsy report might be used at trial, but stated,

“this practical expectation alone cannot be dispositive on the

issue of whether those reports are testimonial.”

121

It then held

on other grounds that autopsy reports are nontestimonial busi-

ness records.

122

Citing its own pre-Crawford precedent, the Massachusetts

high court in Commonwealth v. Verde

123

asserted that a chem-

ist’s report identifying a seized substance as cocaine was ad-

missible as prima facie evidence without the chemist’s testimo-

ny under the public-records hearsay exception.

124

The court

noted that the public-records exception was a “well-recognized”

one.

125

The court’s analysis ultimately came down to its view

that the laboratory report was nontestimonial because it ex-

pressed objective scientific facts rather than conclusory opi-

nions.

126

An Illinois appellate court voiced a similar theory, ex-

plaining, “Crawford specifically disclaims any intention to

restrict traditional hearsay exceptions.”

127

To the extent that courts rely on the theory that Crawford

permits the introduction of hearsay insofar as it falls within a

traditional hearsay exception, they seem to have confused cur-

rent Confrontation Clause jurisprudence—Crawford—with the

old rule—Roberts—which is no longer good law.

128

In other

119. Feliz, 467 F.3d at 234; see also Crawford v. Washington, 541 U.S. 36,

42–50 (2004) (discussing the original intent behind the Sixth Amendment’s

Confrontation Clause).

120. Feliz, 467 F.3d at 229.

121. Id. at 235.

122. See id. at 236–37. Those “other grounds” are competing interests

which effectively constitute a Roberts balancing test. See Roberts v. Ohio, 448

U.S. 56, 65–66 (1980).

123. Commonwealth v. Verde, 827 N.E.2d 701 (Mass. 2005). Verde was the

precursor to Commonwealth v. Melendez-Diaz, the latest Confrontation Clause

case taken up by the Supreme Court. See Commonwealth v. Melendez-Diaz,

870 N.E.2d 676 (Mass. App. Ct. 2007) (unpublished table decision), cert.

granted, 128 S.Ct. 1647 (Mar. 17, 2008) (No. 07-591).

124. Verde, 827 N.E.2d at 704–05.

125. Id. at 705 n.3.

126. See id. at 705–06.

127. People v. So Young Kim, 859 N.E.2d 92, 94 (Ill. App. Ct. 2006).

128. In 2007, Justice Alito, writing for a unanimous Court, explained that

“Roberts had held that the Confrontation Clause permitted the admission of a

hearsay statement made by a declarant who was unavailable to testify if the

statement bore sufficient indicia of reliability . . . because the statement fell

2009] FORENSIC SCIENCE AT TRIAL 1075

words, such opinions apply Roberts under the nominal guise of

applying Crawford.

The Second Circuit argued that the limitations on the

business-records exception function to weed out testimonial

evidence.

129

Certainly if a record is not testimonial in nature,

the Confrontation Clause does not restrict its admission—the

Clause only applies to testimonial statements.

130

But as the

New Hampshire Supreme Court explained, “[t]he Crawford dic-

ta related to the business record exception that existed when

the Federal Constitution was drafted, not that which currently

exists.”

131

The Second Circuit, after acknowledging a factor that

might render an autopsy report testimonial in spite of the

court’s interpretation of the hearsay exception—reasonable an-

ticipation of prosecutorial use—then failed to explain why the

evidence was nonetheless nontestimonial.

132

3. The Objective Facticity of Science

Some courts have concluded that objective facts in a scien-

tific report are nontestimonial. The Supreme Court of Kansas,

for example, drew a distinction in State v. Lackey

133

between

objective factual observations and opinions or disputed conclu-

sions. The “factual, routine, descriptive, and nonanalytical find-

ings” contained in an autopsy report are, according to the court,

not testimonial, whereas “conclusions drawn from the objective

findings” are testimonial.

134

Lackey rejected a distinction be-

tween autopsy reports prepared for homicide investigations and

those where criminal litigation was not expected,

135

instead

premising its rule on the theory that “routine and descriptive

observations” where the “examiner would have little incentive

within a firmly rooted hearsay exception . . . .” Whorton v. Bockting, 127 S. Ct.

1173, 1178 (2007). This approach is now defunct: “[t]he Crawford rule is flatly

inconsistent with the prior governing precedent, Roberts, which Crawford

overruled.” Id. at 1181.

129. See United States v. Feliz, 467 F.3d 227, 233–34 (2d Cir. 2006).

130. See Davis v. Washington, 547 U.S. 813, 821 (2006).

131. State v. O’Maley, 932 A.2d 1, 11 (N.H. 2007).

132. See Feliz, 467 F.3d at 234–36.

133. State v. Lackey, 120 P.3d 332 (Kan. 2005), overruled on other grounds

by State v. Davis, 158 P.3d 317, 321–23 (Kan. 2006). Davis overruled Lackey

with respect to hearsay evidence other than the autopsy report. See Davis, 158

P.3d at 321–23.

134. Lackey, 120 P.3d at 351.

135. Id. at 349.

1076 MINNESOTA LAW REVIEW [93:1058

to fabricate the results” do not implicate the concerns expressed

in Crawford.

136

The Massachusetts Supreme Judicial Court also drew a

line around records of “primary fact.”

137

It explained:

“[c]ertificates of chemical analysis are neither discretionary nor

based on opinion; rather, they merely state the results of a

well-recognized scientific test determining the composition and

quantity of the substance.”

138

Based on similar theories, other

courts have concluded that to the extent that reports contain

“objective” data, they cannot be testimonial because they are

not accusatory.

139

Such arguments are circular and depend on reliability con-

siderations.

140

Because “routine factual findings in an autopsy

report are generally reliable,” courts reason, “therefore, evi-

dence of routine factual findings is not testimonial.”

141

A Kan-

sas appellate court explained the potential consequences of

adopting a “scientific facts are nontestimonial” rule on the facts

of the narcotics possession case before it:

A sine qua non for guilt is the possession of methamphetamine, proof

of which is only found in a report, the accuracy of which has not been

tested in the courtroom. To overcome the presumption of innocence

and convict Laturner of this charge, the jury had to be convinced

beyond a reasonable doubt that the lab report was correct. Thus,

136. See id. at 351.

137. Commonwealth v. Verde, 827 N.E.2d 701, 705 (Mass. 2005) (quoting

Commonwealth v. Slavski, 140 N.E. 465, 469 (Mass. 1923)).

138. Id.

139. See, e.g., People v. Geier, 161 P.3d 104, 140 (Cal. 2007) (“Records of

laboratory protocols followed and the resulting raw data acquired are not ac-

cusatory.”); State v. O’Maley, 932 A.2d 1, 13 (N.H. 2007) (“The second factor

we believe is important is whether the statement is an accusation.”); see also

Michael H. Graham, Crawford/Davis “Testimonial” Interpreted, Removing the

Clutter, 62 U.

MIAMI L. REV. 811, 836–37 (2008) (arguing that forensic labora-

tory reports are nontestimonial because they “do not themselves accuse an

identified or identifiable person of having committed a crime”).

140. See, e.g., Hinojos-Mendoza v. People, 169 P.3d 662, 666 (Colo. 2007)

(en banc) (accusing courts of “erroneously focus[ing] on the reliability of [la-

boratory] reports”); State v. Caulfield, 722 N.W.2d 304, 309 (Minn. 2006) (en

banc) (“The state refers us to cases from other states that, after Crawford,

hold that lab reports are not testimonial. But these cases seem to wrongly fo-

cus on the reliability of such reports.”); State v. March, 216 S.W.3d 663, 665

(Mo. 2007) (“[G]enerally these cases seem to incorrectly focus on the reliability

of such reports.”); cf. Metzger, supra note 1, at 510 (“The misapprehension

about the testimonial nature of the crime laboratory reports seems to be this:

because the affidavits are ‘provided according to scientific procedures and

analysis,’ they are not testimonial.”).

141. State v. Laturner, 163 P.3d 367, 375 (Kan. Ct. App. 2007) (criticizing

this “circular” reasoning).

2009] FORENSIC SCIENCE AT TRIAL 1077

proof of guilt becomes ipse dixit: it is so because the State says it is

so.

142

Admission of otherwise testimonial statements merely be-

cause they consist of “neutral” facts—raw data rather than in-

terpreted data—flies in the face of the Court’s reasoning in

Crawford.

143

The “accusatory” angle, imagining that facts con-

tained in a laboratory or autopsy report do not accuse the de-

fendant of anything, also fails to withstand scrutiny—if a fact

is nonaccusatory, then in a typical DUI case, a defendant could

be convicted on the sole basis of two nonaccusatory, ex parte af-

fidavits: (1) a police officer’s statement that she saw the defen-

dant driving a car, then had the defendant’s blood drawn; and

(2) a laboratory technician’s statement that the defendant’s

blood contained a certain amount of an intoxicating substance.

Yet that would commit the “principal evil” that Crawford at-

tempts to eradicate.

144

4. Evidentiary Value to the Accused

In finding scientific evidence nontestimonial, courts tend to

point out that cross-examination would probably have been of

little value to the defendant in any event. In admitting a DNA

analysis report, the Supreme Court of California explained that

the analyst’s supervisor’s testimony “could reconstruct what the

analyst who processed the samples did at every step” because

laboratory regulations required the analyst to follow a set pro-

tocol.

145

The Supreme Court of New Hampshire noted that if

the technician had testified at trial, she merely would have said

that her report was accurate.

146

And in any event, “she ‘would

almost certainly not remember’” at trial how that exact test

had been performed.

147

The Second Circuit posited that the fact

that “‘medical examiners who regularly perform hundreds of

autopsies are unlikely to have any independent recollection of

the autopsy at issue’”

148

constitutes a “practical difficult[y]”

142. Id. at 376.

143. Cf. Crawford v. Washington, 541 U.S. 36, 66 (2004) (“The Framers

would be astounded to learn that ex parte testimony could be admitted against

a criminal defendant because it was elicited by ‘neutral’ government officers.”).

144. See id. at 50.

145. People v. Geier, 161 P.3d 104, 132 (Cal. 2007). The court also noted

that “DNA extraction is not a difficult procedure.” Id.

146. See State v. O’Maley, 932 A.2d 1, 13 (N.H. 2007).

147. Id. at 14 (quoting State v. Coombs, 821 A.2d 1030, 1033 (N.H. 2003)).

148. United States v. Feliz, 467 F.3d 227, 236 (2d Cir. 2006) (quoting

People v. Durio, 794 N.Y.S.2d 863, 869 (N.Y. Sup. Ct. 2005)).

1078 MINNESOTA LAW REVIEW [93:1058

avoided by its conclusion.

149

But what is the practical difficulty?

Putting the analyst on the stand to say “I don’t remember this

particular examination, but I always follow protocol and I stand

behind what I put in my report” allows the jury to make an in-

formed decision, and provides the defendant a fair opportunity

to inquire about bias and sloppy work habits.

Notably, the Second Circuit’s admonition about the likely

low value of cross-examination came in the context of its dis-

cussion of the dangers, from a policy perspective, of holding

that an autopsy report is a testimonial statement.

150

In other

words, value-to-the-defendant analysis is a factor in a judicial

balancing test, which the Supreme Court overruled in Craw-

ford.

151

Even if not used as part of a Roberts-type balancing

test, judicial inquiry into the usefulness of potential cross-

examination is inapposite.

152

Cross-examination is a “strategic

decision” that the defendant may choose to exercise where she

believes it will be useful or valuable

153

—such as where there is

reason to suspect there were problems with the testing.

154

5. The Competing Interests of Society

Courts worry that applying broadly the right to cross-

examination will hamper the efficient and effective administra-

tion of justice. The Supreme Court of Kansas expressed its fear

that requiring the in-court testimony of a medical examiner

who performed the autopsy relevant to a murder prosecution

would waste public resources and, as in the case before it, prec-

lude prosecutorially valuable evidence—or even prosecution it-

self—where the medical examiner had died or otherwise be-

come unavailable.

155

The court considered this to be a “harsh

and unnecessary result,” since autopsy reports are generally re-

149. Id.

150. See id.

151. See Crawford v. Washington, 541 U.S. 36, 67–68 (2004).

152. Cf. United States v. Washington, 498 F.3d 225, 235 (4th Cir. 2007)

(Michael, J., dissenting) (“A defendant’s right to confront witnesses against

him does not depend on whether a court believes that cross-examination would

be useful.”).

153. Id.

154. Cf. Metzger, supra note 1, at 499 (describing a case in which a state

crime laboratory worker “engage[d] in long-term, systematic, and deliberate

falsification of evidence in criminal cases”).

155. State v. Lackey, 120 P.3d 332, 351 (Kan. 2005), overruled on other

grounds by State v. Davis, 158 P.3d 317, 322 (Kan. 2006).

2009] FORENSIC SCIENCE AT TRIAL 1079

liable and probably not fabricated.

156

Other courts have voiced

similar concerns.

157

Practical consequences to the public, however, are no long-

er a legitimate judicial consideration in deciding Confrontation

Clause cases.

158

The Court’s emphatic, uncompromising lan-

guage in Crawford makes clear that the Clause conveys upon

the defendant an absolute right and that the determination of

whether a statement is testimonial cannot be driven by policy

considerations.

159

D. M

ACHINE WITNESSES: A DEVELOPING TWIST

A new Confrontation Clause rule is developing in some of

the U.S. Courts of Appeals. Already the Fourth, Seventh, and

Eleventh Circuits have held lab reports to be nontestimonial

statements of machines or computers. This section examines

this theory, ultimately rejecting it as inconsistent with Su-

preme Court Confrontation Clause jurisprudence.

In August 2007 a divided panel for the Fourth Circuit is-

sued an opinion deciding United States v. Washington,

160

a rou-

tine drunk driving case, on both typical and unique grounds.

The court issued three alternative holdings to dispense of the

defendant’s Confrontation Clause claim: (1) the “raw data” that

the expert witness relied upon or introduced were not state-

ments of the laboratory technicians; (2) the data were not hear-

say; and (3) the data were nontestimonial.

161

All three holdings

rely on the court’s conclusion that the reports were solely com-

puter-generated statements, which is an analytical angle from

156. Id.

157. See, e.g., United States v. Feliz, 467 F.3d 227, 236 (2d Cir. 2006) (not-

ing that an “‘autopsy cannot be replicated by another pathologist’” and indicat-

ing that it would be “‘against society’s interests to permit the unavailability of

the medical examiner . . . to preclude the prosecution of a homicide case’”

(quoting People v. Durio, 794 N.Y.S.2d 863, 869 (N.Y. Sup. Ct. 2005))); People

v. Geier, 161 P.3d 104, 136–37 (Cal. 2007) (quoting with approval the factors

that the Kansas Supreme Court assigned to society’s competing interests in

the balancing test performed in Lackey).

158. Compare Ohio v. Roberts, 448 U.S. 56, 64 (1980) (articulating a ba-

lancing test that would allow for competing interests such as “effective law en-

forcement” to outweigh the defendant’s right to confront witnesses at trial),

with Crawford v. Washington, 541 U.S. 36, 67–68 (2004) (overruling Roberts

and rejecting the use of a balancing test in place of the “constitutional guaran-

tee[ ]” of the Confrontation Clause).

159. See Crawford, 541 U.S. 36 passim.

160. United States v. Washington, 498 F.3d 225 (4th Cir. 2007).

161. Id. at 227–32.

1080 MINNESOTA LAW REVIEW [93:1058

which no other court had approached a Crawford issue pertain-

ing to forensic science.

The relevant facts follow: Dwonne A. Washington was con-

victed of driving under the influence after testing positive for

both phencyclidine (PCP) and alcohol above the legal limit.

162

The forensic laboratory’s machines produced data indicating

the amount of these substances in Washington’s blood.

163

The

laboratory’s manager, who did not operate the machines, testi-

fied at trial as an expert witness to the test results and the im-

plications for Washington’s driving.

164

In Washington’s view, he

was entitled to confront the lab technicians who actually saw

his blood and placed it in the testing machines.

165

The magi-

strate judge overruled Washington’s objections and admitted

the lab manager’s testimony.

166

The Fourth Circuit reasoned first that the data generated

by the laboratory’s machines were based on the machines’, ra-

ther than the technicians’, viewing and analysis of Washing-

ton’s blood sample.

167

“The raw data generated by the diagnos-

tic machines are the ‘statements’ of the machines themselves,

not their operators.”

168

Second, because Rule 801 of the Federal

Rules of Evidence defines hearsay as a “statement,” which “is

(1) an oral or written assertion or (2) nonverbal conduct of a

person,”

169

the court reasoned that there was no hearsay—

computers are not persons.

170

Third, “the reports generated by

the machines were not testimonial in that they were not relat-

ing past events but the current condition of the blood in the

machines.”

171

Since the machine had no way of knowing that

the blood sample it was analyzing was going to be used for evi-

dence at trial, the evidence could not be considered testimonial

under the Davis purpose test.

172

The court also noted that the

possibility of cross-examining the technicians was of “no value”

to the defendant, since the statements indicating that the blood

162. Id. at 227–28.

163. Id. at 228.

164. Id. at 228–29.

165. Id. at 229.

166. Id.

167. Id. at 230.

168. Id.

169. F

ED. R. EVID. 801(a), (c) (emphasis added).

170. Washington, 498 F.3d at 231.

171. Id. at 232.

172. Id.

2009] FORENSIC SCIENCE AT TRIAL 1081

tested positive for alcohol and PCP were statements of the ma-

chines, not of the technicians.

173

In January 2008 Chief Judge Easterbrook, writing for a

unanimous panel of the Seventh Circuit in United States v.

Moon, cited with approval the Fourth Circuit’s Washington de-

cision.

174

In the Seventh Circuit, “raw data produced by scien-

tific instruments” are now nontestimonial, “though the inter-

pretation of those data may be testimonial.”

175

The underlying

reasoning is much the same: “If the readings are ‘statements’

by a ‘witness against’ the defendants, then the machine must

be the declarant. Yet how could one cross-examine a gas chro-

matograph? Producing spectrographs, ovens, and centrifuges in

court would serve no one’s interests.”

176

The logic of Moon is self-defeating. The Seventh Circuit

conceded that “[a] chemist’s assertion that ‘this substance was

cocaine’” is testimonial within the meaning of Davis.

177

drawing its data-interpretation distinction, the court explained

by way of analogy that if a physician tests a patient’s blood, the

conclusion that the patient has diabetes would be testimonial,

but her insulin and blood sugar levels would be nontestimonial

“raw results.”

178

But in this example, the amount of insulin in

the patient’s blood would be analogous to the amount of cocaine

in the defendant’s blood; the conclusion that the patient has di-

abetes is closer to the conclusion that the defendant violated

the law. The underlying chemical composition of cocaine, which

the Seventh Circuit considers to be nontestimonial, is insepar-

able from the fact that the underlying chemical composition of

cocaine is in fact simultaneously cocaine itself, which the Se-

venth Circuit considers to be testimonial.

In part, the Fourth Circuit’s decision in Washington is

merely another example of the Roberts-type analysis. That Rule

801 defines hearsay as a statement of a person (and not of a

machine)

179

is be irrelevant for constitutional purposes—

because “hearsay” for Confrontation Clause purposes is not

necessarily identical to “hearsay” as defined by subject-to-

173. Id. at 230.

174. United States v. Moon, 512 F.3d 359, 362 (7th Cir. 2008).

175. Id.

176. Id.

177. Id.

178. See id.

179. F

ED. R. EVID. 801(a), (c).

1082 MINNESOTA LAW REVIEW [93:1058

change evidence rules.

180

Instead, the relevant question should

be whether the statement is offered to prove the truth of the

matter asserted.

181

The Fourth Circuit’s theory that the tests

were a mere contemporaneous recordation of the state of the

suspect’s blood is based on the misreading of Davis explained

above.

182

And the court’s reliance on the idea that the defen-

dant would not have gleaned anything of value from an oppor-

tunity for cross-examination is completely out of line with cur-

rent Confrontation Clause doctrine.

183

The Fourth Circuit’s twist on the problem—categorizing

the laboratory reports as statements of the machines only—

functions as an attempt to cheat Crawford and Davis.

184

Most

importantly, this classification is conceptually bizarre. As the

dissent explained, “[t]he test results, although computer-

generated, were produced with the assistance and input of the

technicians and must therefore be attributed to the techni-

cians.”

185

Traditionally, only if “the assertion is produced with-

out any human assistance or input” have federal courts consi-

dered computer-generated assertions not to be statements of

persons.

186

Here, the technician played a “significant role” in

running the tests and creating and documenting the results.

187

One implication of the Washington majority’s classification

is that the defendant has no right to confront the laboratory

technician—there is no statement of the technician to cross-

examine.

188

Because the court found that the report was not

testimonial hearsay, it did not address the question of how the

accused might enforce his right to confront a machine. Technol-

180. See Crawford v. Washington, 541 U.S. 36, 61 (2004).

181. Cf. id. at 59 n.9.

182. See Davis v. Washington, 547 U.S. 813, 827–28 (2006) (discussing con-

temporaneousness as only one of many factors to be considered in determining

whether evidence is testimonial).

183. See Crawford, 541 U.S. at 67–69 (rejecting the Roberts balancing test

in favor of a bright line approach to Confrontation Clause analysis).

184. This approach has also gained traction in the Eleventh Circuit. See

United States v. Lamons, 532 F.3d 1251, 1263 (11th Cir. 2008) (“In light of the

constitutional text and the historical focus of the Confrontation Clause, we are

persuaded that the witnesses with whom the Confrontation Clause is con-

cerned are human witnesses, and that the evidence challenged in this appeal

does not contain the statements of human witnesses.” (citing United States v.

Moon, 512 F.3d 359, 362 (7th Cir. 2008) and United States v. Washington, 498

F.3d 225, 231 (4th Cir. 2007))).

185. Washington, 498 F.3d at 231–32 (Michael, J., dissenting).

186. Id. at 233 (emphasis added).

187. Id.

188. See id. at 230 (majority opinion).

2009] FORENSIC SCIENCE AT TRIAL 1083

ogy is not yet at the point where a computer is capable of tak-

ing the stand and answering questions in response to counsel’s

cross-examination. Yet while Crawford and Davis would accor-

dingly render the machine’s statement inadmissible, the

Fourth Circuit explained that “reliability concerns” could be

addressed “through authentication” of the underlying technolo-

gical processes, thereby suggesting that machine-generated tes-

timonial hearsay statements are admissible even though the

witness is not subject to cross-examination.

189

Ultimately, machine-generated laboratory reports are pro-

duced jointly by machines and the persons operating the ma-

chines. The reports must be considered statements of those

persons. Although the analysis in this section indicates that the

laboratory reports at issue in Washington and Moon should

have been held to be testimonial, more important is the under-

standing that such reports must be deemed either testimonial

or nontestimonial on the same grounds as any other laboratory

report. The machine-generated-statement theory is not persua-

sive.

III. AN ANALYSIS FOR COURTS TO FOLLOW

In light of the split among lower courts, this Note proposes

that courts should, in order to apply Crawford and Davis faith-

fully, start analyzing Confrontation Clause cases pertaining to

scientific evidence in a consistent manner, set forth below. Two

generalized groups of cases need to be treated differently. For

dealing with laboratory reports, in most instances a bright-line

rule declaring the reports testimonial statements of the persons

who generated them will be appropriate. Autopsy reports

should be considered on a case-by-case basis, while adhering to

the principles outlined below.

A. L

ABORATORY REPORTS: A (NEARLY) BRIGHT-LINE RULE

Most laboratory reports used by prosecutors against the

accused are prepared for criminal prosecution. Frequently oc-

curring examples include blood tests of persons suspected of

driving while intoxicated and chemical analyses of seized sub-

stances suspected of being illegal narcotics. In such cases,

courts should hold that the reports are per se testimonial.

However, there will be rare cases in which such a ruling is in-

189. See id. at 231. The court’s consideration of reliability is, of course,

misplaced. See Crawford v. Washington, 541 U.S. 36, 67–69 (2004).

1084 MINNESOTA LAW REVIEW [93:1058

appropriate—this “Magyari exception” should be approached in

an ad hoc manner.

1. Many Laboratory Reports Are Testimonial Per Se

Caulfield is, in all relevant factual aspects, typical of drug

seizure cases. Police observed Scott Caulfield behaving suspi-

ciously in and outside of a bar in Rochester, Minnesota.

190

short, he was behaving like a drug dealer.

191

The police lawful-

192

took six plastic bags from Caulfield, which contained what

the suspect identified as cocaine.

193

The police recorded a posi-

tive field test and sent the drugs to a crime laboratory, where

an analyst produced a report identifying the substance as co-

caine.

194

The prosecution introduced the analyst’s report with-

out putting the analyst on the stand.

195

For Confrontation Clause purposes, the facts of Kent are

analogous to those of Caulfield. Late at night, Adam J. Kent

crashed his car; he hit the curb and his car wound up upside-

down on somebody’s front yard.

196

A police officer arrived and

noticed that Kent smelled like alcohol.

197

Kent’s speech was

slurred and, when questioned, he admitted that he had con-

sumed five beers that evening.

198

The officer concluded that

Kent was intoxicated and arrested him, taking him to the hos-

pital because of the crash.

199

At the request of the police, a hos-

pital worker drew a sample of Kent’s blood and signed a certifi-

cate indicating that he had followed protocol.

200

The police sent

the blood sample to their forensic laboratory, which issued a

report concluding that Kent’s blood alcohol content was above

the legal limit.

201

State v. Crager

202

was a murder case in which the DNA

analysis conducted also fits the pattern of Caulfield and

190. State v. Caulfield, 722 N.W.2d 304, 307 (Minn. 2006) (en banc).

191. Id.

192. Had the police behaved unlawfully, that would be a Fourth Amend-

ment question outside of the scope of this Note.

193. Caulfield, 722 N.W.2d at 307.

194. Id.

195. See id. at 306.

196. State v. Kent, 918 A.2d 626, 628 (N.J. Super. Ct. App. Div. 2007).

197. Id. at 629.

198. Id.

199. Id.

200. Id. at 629–30.

201. Id. at 630–31.

202. State v. Crager, 116 Ohio St.3d 369, 2007-Ohio-6840, 879 N.E.2d 745.

2009] FORENSIC SCIENCE AT TRIAL 1085

Kent.

203

The police found a woman lying murdered in the bed-

room of her home.

204

They identified Lee Crager as a suspect,

and upon finding his clothes covered in blood, submitted it for

DNA testing; it was a match for the victim’s DNA.

205

Crager’s

DNA also matched that found on cigarettes at the crime

scene.

206

The prosecution offered the DNA report into evidence

without having the DNA analyst who performed the tests testi-

fy at trial.

207

In each case—Caulfield, Kent, and Crager—the police sus-

pected the defendant of having committed a crime, and in order

to prove that crime, had scientific tests run at a laboratory. In

each case, a court can determine under a straightforward Davis

analysis that the laboratory report is a testimonial statement of

the person who authored it. It is difficult, perhaps impossible,

to imagine any circumstances under which these tests would

have been run in which law enforcement was not attempting to

establish the facts relevant to the potential prosecution of a

crime. It is similarly difficult or impossible to imagine any cir-

cumstances under which ordering a laboratory analysis of a

seized substance believed to be illegal narcotics, a blood screen-

ing of a suspected drunk driver, or a DNA or other analysis

matching a suspected criminal to evidence found at a crime

scene, could possibly be for the purpose of enabling the police to

meet an ongoing emergency.

In short, where the police request a laboratory analysis of

what seems to be evidence, that analysis will always be testi-

monial. In order to promote adherence to the Supreme Court’s

presentation of the Confrontation Clause in Crawford and Da-

vis, courts should therefore articulate a bright-line rule an-

nouncing that such evidence is per se testimonial.

203. Courts in Kent and Caulfield and the appellate court in State v. Crag-

er, 164 Ohio App. 3d 816, 2005-Ohio-6868, 844 N.E.2d 390, found that labora-

tory reports are testimonial under Crawford. The Supreme Court of Ohio,

however, reversed the appellate court’s decision in Crager, finding that under

Crawford, the reports were “non-testimonial.” See 116 Ohio St.3d 369, 2007-

Ohio-6840, 879 N.E.2d 745, at ¶ 78.

204. 116 Ohio St.3d 369, 2007-Ohio-6840, 879 N.E.2d 745, at ¶ 2.

205. Id. ¶ 2–4.

206. Id. ¶ 4.

207. Id. ¶ 8 (the DNA analyst who prepared the two DNA reports was on

maternity leave at the time of trial).

1086 MINNESOTA LAW REVIEW [93:1058

2. The Magyari Exception

In Magyari, the Court of Appeals for the Armed Forces en-

countered the issue of an employer’s regularly conducted, ran-

domized urine testing which had the potential to trigger crimi-

nal prosecution.

208

The laboratory technicians had no way of

knowing whose sample they were working with; it was identical

to the other 199 samples in the batch.

209

A civilian analogue to

the facts of Magyari may not even presently exist.

210

The facts of this case are extremely important, and a totali-

ty of the circumstances test was appropriately applied. The

technicians knew or should have known that a positive test

could trigger prosecution. The tests were serving a law en-

forcement function.

211

But no person whose sample was tested

should be considered a suspect, and in the case of each sample,

it was far more likely than not that no crime had been commit-

ted.

212

When sufficiently similar cases arise, courts ought to

look at the facts of each case, apply the rules articulated in

Crawford and Davis where possible, and consider the aims of

the Confrontation Clause and the spirit in which the two cases

were decided. Courts should keep in mind that the Sixth

Amendment expresses a preference for face-to-face confronta-

tion at trial. In asking whether the circumstances surrounding

the request for the scientific analysis indicate that the purpose

is to prove facts potentially relevant to a subsequent prosecu-

tion, the particularity of the request is relevant—the existence

of a suspect or criminal investigation makes it more likely that

the scientific analysis will be testimonial. And civilian law en-

forcement should be given less leeway than the military be-

cause the military deals only with the members of its own or-

208. See United States v. Magyari, 63 M.J. 123, 127 (C.A.A.F. 2006) (apply-

ing the Confrontation Clause analysis framework of Crawford and Roberts,

civilian criminal proceedings, to court-martial proceedings).

209. Id. at 126.

210. The closest analogous situation might be employer drug testing. A

nonmilitary employer’s sanctions for a failed drug test, however, do not in-

clude criminal or equivalent proceedings. Absent an employer working in co-

operation with law enforcement, there is not even the potential for criminal

proceedings—an obvious requirement for application of the Confrontation

Clause.

211. But see Magyari, 63 M.J. at 126 (“In this context, the better view is

that these lab technicians were not engaged in a law enforcement function, a

search for evidence in anticipation of prosecution or trial.”).

212. See id.

2009] FORENSIC SCIENCE AT TRIAL 1087

ganization, in what the Supreme Court has long recognized as

“the unique nature of . . . the military society.”

213

B. A

UTOPSY REPORTS

An autopsy report is not “prepared for use at trial,” but

“any medical examiner preparing such a report must expect

that it may later be available for use at trial.”

214

An autopsy re-

port records the medical examiner’s observations and an-

nounces a cause of death.

215

Rather than being prepared at the

request of law enforcement officers, it is frequently ordered by

statute and performed by a nonadversarial medical examin-

er.

216

However, the fact that the medical examiner is not sup-

posed to be in an adversarial relationship with the suspect is

not a valid constitutional consideration.

217

Courts will have to weigh the relevant constitutional con-

siderations under the surrounding circumstances of each case.

To what extent does using an autopsy report as evidence

against a criminal defendant resemble the use of ex parte evi-

dence against the accused? How much potential is there for

prosecutorial abuse, especially in highly political cases? To

what extent should the medical examiner anticipate that the

autopsy will prove past events relevant to use at a criminal tri-

al?

On a general level, one commentator has suggested that

“the Confrontation Clause should allow a defendant to confront

the crucial piece of hearsay that directly establishes an element

of the crime.”

218

More particularly, it is clear that an autopsy

report is at least functionally equivalent to one type of forma-

lized statement—an affidavit—about which the Court ex-

pressed concern.

219

Whether the police have identified a suspect

213. See United States v. Rankin, 63 M.J. 552, 555 (N-M. Ct. Crim. App.

2006) (citing Goldman v. Weinberger, 475 U.S. 503, 507 (1986)).

214. See United States v. Feliz, 467 F.3d 227, 235 (2d Cir. 2006).

215. See id. at 236 n.6; B

LACK’S LAW DICTIONARY 145 (8th ed. 2004).

216. Bradley Morin, Note, Science, Crawford, and Testimonial Hearsay:

Applying the Confrontation Clause to Laboratory Reports, 85 B.U.

L. REV.

1243, 1264 (2005).

217. Cf. Crawford v. Washington, 541 U.S. 36, 66 (2004) (explaining this in

the context of “neutral” government officers). Contra Carolyn Zabrycki, Com-

ment, Toward a Definition of “Testimonial”: How Autopsy Reports Do Not Em-

body the Qualities of a Testimonial Statement, 96 C

AL. L. REV. 1093, 1111

(2008) (arguing that autopsy reports should be deemed nontestimonial be-

cause medical examiners are nonadversarial and are not government officers).

218. Morin, supra note 216, at 1264.

219. See Crawford, 541 U.S. at 51–52 (setting forth the three potential de-

1088 MINNESOTA LAW REVIEW [93:1058

should be a significant factor in a court’s decision. If there is a

suspect, there is an ongoing criminal investigation and a medi-

cal examiner has a reason to reach a certain result. At the mo-

ment it becomes clear that foul play was likely involved in the

subject’s death, whether because there is an ongoing criminal

investigation or because the body itself so indicates, an objec-

tive observer would think that the report is reasonably likely to

be used at trial.

220

While these factors indicate that there is a

good chance an autopsy report used by the prosecution in a

criminal trial will be testimonial, there are imaginable circums-

tances in which it will be nontestimonial—such as where there

are no identified suspects and there is low potential for

abuse.

221

CONCLUSION

Due to a sharp split among the lower courts in the wake of

Crawford and Davis, criminal defendants in a significant num-

ber of jurisdictions are being denied their constitutional right to

confront the witnesses against them. In cases involving forensic

science, many courts are applying federal constitutional law

correctly. Many others, however, are still analyzing the evi-

dence under a now-defunct Roberts test and reaching the wrong

result, thereby preventing a large number of defendants from

cross-examining the forensic scientists who function as power-

ful witnesses against them. Courts should announce that where

the police request a forensic laboratory analysis of what seems

to be evidence, that analysis is a per se testimonial statement

of the person who played a significant role in conducting the

scientific testing. Courts should undertake fact-intensive case-

finitions of “testimonial,” one of which deals with formalized statements such

as affidavits).

220. The “what an objective person would reasonably think” standard has

been used frequently in constitutional law, particularly in constitutional crim-

inal procedure. See Jenny Roberts, The Mythical Divide Between Collateral

and Direct Consequences of Criminal Convictions: Involuntary Commitment of

“Sexually Violent Predators,” 93 M

INN. L. REV. 670, 713–19 (2008). With re-

spect to the Confrontation Clause, the text of the Court’s jurisprudence creates

the foundation for such a test. See Crawford, 541 U.S. at 51–52 (discussing the

“objective witness” formulation of a “testimonial” statement).

221. Somewhat paradoxically, there might be as much potential for prose-

cutorial abuse where the circumstances indicate that foul play might—but

might not—have been involved, because in this situation the medical examin-

er’s statement has the ability to propel or stall prosecution of any eventual

suspect. This further supports the idea that autopsy reports will require fact-

intensive analysis under the Confrontation Clause.

2009] FORENSIC SCIENCE AT TRIAL 1089

by-case analyses where autopsy reports are at issue, applying a

totality of the circumstances test in which the identification of

a suspect in a criminal investigation is an especially weighty

factor.