dissenting.
The Sixth Amendment Confrontation Clause binds the States and the National Government. Pointer v. Texas, 380 U. S. 400, 403 (1965). Two Terms ago, in a case arising from a state criminal prosecution, the Court interpreted the Clause to mandate exclusion of a laboratory report sought to be introduced based on the authority of that report’s own sworn statement that a test had been performed yielding the results as shown. Melendez-Diaz v. Massachusetts, 557 U. S. 305 (2009). The Court’s opinion in that case held the report inadmissible because no one was present at trial to testify to its contents.
Whether or not one agrees with the reasoning and the result in Melendez-Diaz, the Court today takes the new and serious misstep of extending that holding to instances like this one. Here a knowledgeable representative of the labo*675ratory was present to testify and to explain the lab’s processes and the details of the report; but because he was not the analyst who filled out part of the form and transcribed onto it the test result from a machine printout, the Court finds a confrontation violation. Some of the principal objections to the Court’s underlying theory have been set out earlier and need not be repeated here. See id., at 330-3B2 (Kennedy, J., dissenting). Additional reasons, applicable to the extension of that doctrine and to the new ruling in this case, are now explained in support of this respectful dissent.
I
Before today, the Court had not held that the Confrontation Clause bars admission of scientific findings when an employee of the testing laboratory authenticates the findings, testifies to the laboratory’s methods and practices, and is cross-examined at trial. Far from replacing live testimony with “systematic” and “extrajudicial” examinations, Davis v. Washington, 547 U. S. 813, 835, 836 (2006) (Thomas, J., concurring in judgment in part and dissenting in part) (emphasis deleted; internal quotation marks omitted), these procedures are fully consistent with the Confrontation Clause and with well-established principles for ensuring that criminal trials are conducted in full accord with requirements of fairness and reliability and with the confrontation guarantee. They do not “resemble Marian proceedings.” Id., at 837.
The procedures followed here, but now invalidated by the Court, make live testimony rather than the “solemnity” of a document the primary reason to credit the laboratory’s scientific results. Id., at 838. Unlike Melendez-Diaz, where the jury was asked to credit a laboratory’s findings based solely on documents that were “quite plainly affidavits,” 557 U. S., at 330 (Thomas, J., concurring) (internal quotation marks omitted), here the signature, heading, or legend on the document were routine authentication elements for a report that would be assessed and explained by in-court testimony sub*676ject to fall cross-examination. The only sworn statement at issue was that of the witness who was present and who testified.
The record reveals that the certifying analyst’s role here was no greater than that of anyone else in the chain of custody. App. 56 (laboratory employee’s testimony agreeing that “once the material is prepared and placed in the machine, you don’t need any particular expertise to record the results”). The information contained in the report was the result of a scientific process comprising multiple participants’ acts, each with its own evidentiary significance. These acts included receipt of the sample at the laboratory; recording its receipt; storing it; placing the sample into the testing device; transposing the printout of the -results of the test onto the report; and review of the results. See id., at 48-56; see also Brief for New Mexico Department of Health, Scientific Laboratory Division, as Amicus Curiae 4 (hereinafter New Mexico Scientific Laboratory Brief) (“Each blood sample has original testing work by... as many as seve[n] analysts ... ”); App. 62 (indicating that this case involved three laboratory analysts who, respectively, received, analyzed, and reviewed analysis of the sample); cf. Brief for State of Indiana et al. as Amici Curiae in Briscoe v. Virginia, O. T. 2009, No. 07-11191, p. 10 (explaining that DNA analysis can involve the combined efforts of up to 40 analysts).
In the New Mexico scientific laboratory where the blood sample was processed, analyses are run in batches involving 40-60 samples. Each sample is identified by a computer-generated number that is not linked back to the file containing the name of the person from whom the sample came until after all testing is completed. See New Mexico Scientific Laboratory Brief 26. The analysis is mechanically performed by the gas chromatograph, which may operate — as in this case — after all the laboratory employees leave for the day. See id., at 17. And whatever the result, it is reported to both law enforcement and the defense. See id., at 36.
*677The representative of the testing laboratory whom the prosecution called was a scientific analyst named Mr. Razatos. He testified that he “help[ed] in overseeing the administration of these programs throughout the State,” and he was qualified to answer questions concerning each of these steps. App. 49. The Court has held that the government need not produce at trial “everyone who laid hands on the evidence,” Melendez-Diaz, supra, at 311, n. 1. Here, the defense used the opportunity in cross-examination to highlight the absence at trial of certain laboratory employees. Under questioning by Bullcoming’s attorney, Razatos acknowledged that his name did not appear on the report; that he did not receive the sample, perform the analysis, or complete the review; and that he did not know the reason for some personnel decisions. App. 58. After weighing arguments from defense counsel concerning these admissions, and after considering the testimony of Razatos, who knew the laboratory’s protocols and processes, the jury found no reasonable doubt as to the defendant’s guilt.
In these circumstances, requiring the State to call the technician who filled out a form and recorded the results of a test is a hollow formality. The defense remains free to challenge any and all forensic evidence. It may call and examine the technician who performed a test. And it may call other expert witnesses to explain that tests are not always reliable or that the technician might have made a mistake. The jury can then decide whether to credit the test, as it did here. The States, furthermore, can assess the progress of scientific testing and enact or adopt statutes and rules to ensure that only reliable evidence is admitted. Rejecting these commonsense arguments and the concept that reliability is a legitimate concern, the Court today takes a different course. It once more assumes for itself a central role in mandating detailed evidentiary rules, thereby extending and confirming Melendez-Diaz’s “vast potential to disrupt criminal procedures.” 557 U. S., at 331 (Kennedy, J., dissenting).
*678II
The protections in the Confrontation Clause, and indeed the Sixth Amendment in general, are designed to ensure a fair trial with reliable evidence. But the Crawford v. Washington, 541 U. S. 36 (2004), line of cases has treated the reliability of evidence as a reason to exclude it. Id., at 61-62. Today, for example, the Court bars admission of a lab report because it “is formalized in a signed document.” Ante, at 665 (internal quotation marks omitted). The Court’s unconventional and unstated premise is that the State — by acting to ensure a statement’s reliability — makes the statement more formal and therefore less likely to be admitted. Park, Is Confrontation the Bottom Line? 19 Regent U. L. Rev. 459, 461 (2007). That is so, the Court insists, because reliability does not animate the Confrontation Clause. Ante, at 661; Melendez-Diaz, supra, at 317-318; Crawford, supra, at 61-62. Yet just this Term the Court ruled that, in another confrontation context, reliability was an essential part of the constitutional inquiry. See Michigan v. Bryant, 562 U. S. 344, 358-359, 361-362 (2011).
Like reliability, other principles have weaved in and out of the Crawford jurisprudence. Solemnity has sometimes been dispositive, see Melendez-Diaz, 557 U. S., at 310-311; id., at 329-330 (Thomas, J., concurring), and sometimes not, see Davis, 547 U. S., at 834-837, 841 (Thomas, J., concurring in judgment in part and dissenting in part). So, too, with the elusive distinction between utterances aimed at proving past events, and those calculated to help police keep the peace. Compare Davis, supra, and Bryant, 562 U. S., at 371-376, with id., at 384-387 (Scalia, J., dissenting).
It is not even clear which witnesses’ testimony could render a scientific report admissible under the Court’s approach. Melendez-Diaz stated an inflexible rule: Where “analysts’ affidavits” included “testimonial statements,” defendants were “entitled to be confronted with the analysts” themselves. *679557 U. S., at 311 (internal quotation marks omitted). Now, the Court reveals, this rule is either less clear than it first appeared or too strict to be followed. A report is admissible, today’s opinion states, if a “live witness competent to testify to the truth of the statements made in the report” appears. Ante, at 651. Such witnesses include not just the certifying analyst, but also any “scientist who . . . performed] or observe[d] the test reported in the certification.” Ante, at 652.
Today’s majority is not committed in equal shares to a common set of principles in applying the holding of Crawford. Compare Davis, supra (opinion for the Court by Scalia, J.), with id., at 834 (Thomas, J., concurring in judgment in part and dissenting in part); and Bryant, supra (opinion for the Court by Sotomayor, J.), with id., at 378 (Thomas, J., concurring in judgment), id., at 379 (Scalia, J., dissenting), and id., at 395 (Ginsburg, J., dissenting); and ante, p. 647 (opinion of the Court), with ante, p. 668 (SOTO-MAYOR, J., concurring). That the Court in the wake of Crawford has had such trouble fashioning a clear vision of that case’s meaning is unsettling; for Crawford binds every.judge in every criminal trial in every local, state, and federal court in the Nation. This Court’s prior decisions leave trial judges to “guess what future rules this Court will distill from the sparse constitutional text,” Melendez-Diaz, supra, at 331 (Kennedy, J., dissenting), or to struggle to apply an “amorphous, if not entirely subjective,” “highly context-dependent inquiry” involving “open-ended balancing,” Bryant, supra, at 393 (Scalia, J., dissenting) (internal quotation marks omitted) (listing 11 factors relevant under the majority’s approach).
The persistent ambiguities in the Court’s approach are symptomatic of a rule not amenable to sensible applications. Procedures involving multiple participants illustrate the problem. In Melendez-Diaz the Court insisted that its opin*680ion did not require everyone in the chain of custody to testify but then qualified that “what testimony is introduced must ... be introduced live.” 557 U. S., at 311, n. 1; ante, at 656, n. 2. This could mean that a statement that evidence remained in law-enforcement custody is admissible if the statement’s maker appears in court. If so, an intern at police headquarters could review the evidence log, declare that chain of custody was retained, and so testify. The rule could also be that the intern’s statement — which draws on statements in the evidence log — is inadmissible unless every officer who signed the log appears at trial. That rule, if applied to this case, would have conditioned admissibility of the report on the testimony of three or more identified witnesses. See App. 62. In other instances, 7 or even 40 witnesses could be required. See supra, at 676. The court has thus— in its fidelity to Melendez-Diaz — boxed itself into a choice of evils: render the Confrontation Clause proforma or construe it so that its dictates are unworkable.
III
Crawford itself does not compel today’s conclusion. It is true, as Crawford confirmed, that the Confrontation Clause seeks in part to bar the government from replicating trial procedures outside of public view. See 541 U. S., at 50; Bryant, supra, at 358-359. Crawford explained that the basic purpose of the Clause was to address the sort of abuses exemplified at the notorious treason trial of Sir Walter Raleigh. 541 U. S., at 51. On this view the Clause operates to bar admission of out-of-court statements obtained through formal interrogation in preparation for trial. The danger is that innocent defendants may be convicted on the basis of unreliable, untested statements by those who observed — or claimed to have observed — preparation for or commission of the crime. And, of course, those statements might not have been uttered at all or — even if spoken — might not have been true.
*681A rule that bars testimony of that sort, however, provides neither cause nor necessity to impose a constitutional bar on the admission of impartial lab reports like the instant one, reports prepared by experienced technicians in laboratories that follow professional norms , and scientific protocols. In addition to the constitutional right to call witnesses in his own defense, the defendant in this case was already protected by checks on potential prosecutorial abuse such as free retesting for defendants; result-blind issuance of reports; testing by an independent agency; routine processes performed en masse, which reduce opportunities for targeted bias; and labs operating pursuant to scientific and professional norms and oversight. See Brief for Respondent 5, 14-15, 41, 54; New Mexico Scientific Laboratory Brief 2, 26.
In addition to preventing the State from conducting ex parte trials, Crawford’s rejection of the regime of Ohio v. Roberts, 448 U. S. 56 (1980), seemed to have two underlying jurisprudential objectives. One was to delink the intricacies of hearsay law from a constitutional mandate; and the other was to allow the States, in their own courts and legislatures and without this Court’s supervision, to explore and develop sensible, specific evidentiary rules pertaining to the admissibility of certain statements. These results were to be welcomed, for this Court lacks the experience and day-to-day familiarity with the trial process to suit it well to assume the role of national tribunal for rules of evidence. Yet far from pursuing these objectives, the Court rejects them in favor of their opposites.
Instead of freeing the Clause from reliance on hearsay doctrines, the Court has now linked the Clause with hearsay rules in their earliest, most rigid, and least refined formulations. See, e. g., Mosteller, Remaking Confrontation Clause and Hearsay Doctrine Under the Challenge of Child Sexual Abuse Prosecutions, 1993 U. Ill. L. Rev. 691, 739-740, 742, 744-746; Gallanis, The Rise of Modern Evidence Law, 84 Iowa L. Rev. 499, 502-503, 514-515, 533-537 (1999). In *682cases like Melendez-Diaz and this one, the Court has tied the Confrontation Clause to 18th-century hearsay rules unleavened by principles tending to make those rules more sensible. Sklansky, Hearsay’s Last Hurrah, 2009 S. Ct. Rev. 1, 5-6, 36. As a result, the Court has taken the Clause far beyond its most important application, which is to forbid sworn, ex parte, out-of-court statements by unconfronted and available witnesses who observed the crime and do not appear at trial.
Second, the States are not just at risk of having some of their hearsay rules reviewed by this Court. They often are foreclosed now from contributing to the formulation and enactment of rules that make trials fairer and more reliable. For instance, recent state laws allowing admission of well-documented and supported reports of abuse by women whose abusers later murdered them must give way, unless that abuser murdered with the specific purpose of foreclosing the testimony. Giles v. California, 554 U. S. 353 (2008); Sklansky, supra, at 14-15. Whether those statutes could provide sufficient indicia of reliability and other safeguards to comply with the Confrontation Clause as it should be understood is, to be sure, an open question. The point is that the States cannot now participate in the development of this difficult part of the law.
In short, there is an ongoing, continued, and systemic displacement of the States and dislocation of the federal structure. Cf. Melendez-Diaz, 557 U. S., at 307-309, 327-329. If this Court persists in applying wooden formalism in order to bar reliable testimony offered by the prosecution — testimony thought proper for many decades in state and federal courts committed to devising fair trial processes — then the States might find it necessary and appropriate to enact statutes to accommodate this new, intrusive federal regime. If they do, those rules could remain on state statute books for decades, even if subsequent decisions of this Court were to better implement the objectives of Crawford. This underscores *683the disruptive, long-term structural consequences of decisions like the one the Court announces today.
States also may decide it is proper and appropriate to enact statutes that require defense counsel to give advance notice if they are going to object to introduction of a report without the presence in court of the technician who prepared it. Indeed, today's opinion relies upon laws of that sort as a palliative to the disruption it is causing. Ante, at 666-667 (plurality opinion). It is quite unrealistic, however, to think that this will take away from the defense the incentives to insist on having the certifying analyst present. There is in the ordinary case that proceeds to trial no good reason for defense counsel to waive the right of confrontation as the Court now interprets it.
Today’s opinion repeats an assertion from Melendez-Diaz that its decision will not “impose an undue burden on the prosecution. ” Ante, at 665 (plurality opinion). But evidence to the contrary already has begun to mount. See, e. g., Brief for State of California et al. as Amici Curiae 7 (explaining that the 10 toxicologists for the Los Angeles Police Department spent 782 hours at 261 court appearances during a 1-year period); Brief for National District Attorneys Association et al. as Amici Curiae 23 (observing that each blood-alcohol analyst in California, processes 3,220 cases per year on average). New and more rigorous empirical studies further detailing the unfortunate effects of Melendez-Diaz are sure to be forthcoming.
In the meantime, New Mexico’s experience exemplifies the problems ahead. Prom 2008 to 2010, subpoenas requiring New Mexico analysts to testify in impaired-driving cases rose 71%, to 1,600 — or 8 or 9 every workday. New Mexico Scientific Laboratory Brief 2. In a State that is the Nation’s fifth largest by area and that employs just 10 total analysts, id., at 3, each analyst in blood-alcohol cases recently received 200 subpoenas per year, id., at 33. The analysts now must travel great distances on most working days. The result *684has been, in the laboratory’s words, “chaotic.” Id., at 5. And if the defense raises an objection and the analyst is tied up in another court proceeding; or on leave; or absent; or delayed in transit; or no longer employed; or ill; or no longer living, the defense gets a windfall. As a result, good defense attorneys will object in ever-greater numbers to a prosecution failure or inability to produce laboratory analysts at trial. The concomitant increases in subpoenas will further impede the state laboratory’s ability to keep pace with its obligations. Scarce state resources could be committed to other urgent needs in the criminal justice system.
* * #
Seven years after its initiation, it bears remembering that the Crawford approach was not preordained. This Court’s missteps have produced an interpretation of the word “witness” at odds with its meaning elsewhere in the Constitution, including elsewhere in the Sixth Amendment, see Amar, Sixth Amendment First Principles, 84 Geo. L. J. 641, 647, 691-696 (1996), and at odds with the sound administration of justice. It is time to return to solid ground. A proper place to begin that return is to decline to extend Melendez-Diaz to bar the reliable, commonsense evidentiary framework the State sought to follow in this case.