Since the Supreme Court’s revitalization of the Sixth Amendment’s Confrontation Clause in Crawford v. Washington, the legal world has waited for each installment of the doctrine with bated breath. In the years that the Confrontation Clause laid dormant, however, expert testimonies unsubstantiated by personal knowledge were blessed by Rule 703 of the Federal Rules of Evidence. This exemption appears irreconcilable with Crawford’s stance on the Confrontation Clause, which has generated a series of disputes arising from experts testifying without personal knowledge in various postures of criminal trials. Last term, in Smith v. Arizona, the Supreme Court held that the Confrontation Clause is violated when statements from laboratory analysts are admitted without an opportunity for cross-examination. Here, Petitioner Jason Smith was charged with five drug-related offenses for which an Arizona Department of Public Safety (DPS) forensic scientist, Elizabeth Rast, performed the analysis. By the time Smith went to trial, Rast was no longer a DPS employee and was unavailable for testimony. Another DPS scientist, Greggory Longoni, testified using Rast’s notes. Smith was convicted, and the Arizona Court of Appeals affirmed his conviction. The Supreme Court vacated and remanded. While the decision has provided much-needed clarity regarding the admission of statements from forensic experts, the cadence at which analysis techniques are developing will demand the elevation of another admissibility threshold.
The series of cases that emerged from the tension between the Confrontation Clause and Rule 703, notably Melendez-Diaz v. Massachusetts and Bullcoming v. New Mexico, have affirmed that the Confrontation Clause does not relent in the face of expert testimonies. However, these decisions did not provide a coherent bright line for when a criminal defendant’s Confrontation Clause rights attach in the admission of a forensic report. The most recent case in the series before Smith, Williams v. Illinois, was a 5-4 decision in the government’s favor, but not a single opinion’s line of reasoning earned the votes of five justices. The 9-0 decision that came of Smith, then, could not have been more timely.
For the current state of drug analysis, the Smith Court was correct that cross-examination is vital for the protection of defendants’ constitutional rights. Rast had used an array of “scientific methods” to analyze the substances found on and around Smith including a microscopic examination, a chemical color test, and a gas chromatograph/mass spectrometer test. These analyses are performed nonblind in forensic laboratories across the country, which necessarily places analysts as accusatory witnesses. Besides the testimonial nature of such analyses, the reliability of the analysts themselves may be questionable — a defect in the prosecution that may be cured or assuaged if defendants are given an opportunity for confrontation.
The government, however, raised a legitimate concern that the weight of the Confrontation Clause may be too heavy for the burdened forensic laboratories around the country. Forensic laboratories have already been facing an unprecedented growth in expenditures due to the opioid crisis: “In 2019, the average drug chemistry laboratory in the United States had a backlog of 1,862 cases and it took 60 days for a case to be analyzed and results reported.” Aside from the operational challenges from analysts being routinely taken out of an overwhelmed operation, there is also a logistical matter of identifying the analyst(s) that oversaw specific samples. Most tests are performed by one analyst, but forensic analyses are typically carried out stepwise, which may “[f]requently” lead to involvement of different individuals along a chain. For the government to identify each person responsible and bring them to trial is not as straightforward as one might think — “Longoni traveled approximately six hours round-trip to testify in [Smith’s] case.” When fluctuations in laboratory personnel are added to the equation, the weight of Smith’s ruling will be felt soon enough: to admit a forensic report for the truth of the matter, the government must either present the original analyst or order retesting by an available analyst.
Even before the Confrontation Clause entered the picture, forensic laboratories had been looking for ways to streamline the process for testing samples. Advances in laboratory automation and the push for carrying out more tests have prompted a variety of manufacturers to produce fully automated chemistry analyzers. These automated analyzers require laboratory staff only to prepare the samples themselves and will display either “Positive” or “Negative” results next to each substance. For routine drug screening, some laboratories only use automated chemistry analyzers, and streamlined workflows assisted by automation “could partially account for the significant decrease in backlogs in DNA database casework (e.g., convicted offender and arrestee samples), which decreased by 85% from 2009 to 2014.” Such developments beg the question: what happens when techniques become sophisticated enough to bypass the need for interpretation?
This set of facts almost presented itself in Bullcoming, where the lower court stated that Bullcoming’s “true accuser” was the machine that measured and reported his blood alcohol content, “while testing analyst [Curtis] Caylor’s role was that of ‘mere scrivener.’” However, the Supreme Court disagreed:
Caylor’s certification, however, reported more than a machine-generated number. Caylor certified that he received Bullcoming’s blood sample intact with the seal unbroken, that he checked to make sure that the forensic report number and the sample number “correspond[ed],” and that he performed on Bullcoming’s sample a particular test, adhering to a precise protocol. He further represented, by leaving the “[r]emarks” section of the report blank, that no “circumstance or condition . . . affect[ed] the integrity of the sample or . . . the validity of the analysis.” These representations, relating to past events and human actions not revealed in raw, machine-produced data, are meet for cross-examination.
Justice Sotomayor was prescient when she concurred in part. She clarified that the Court did not rule on whether “a State could introduce (assuming an adequate chain of custody foundation) raw data generated by a machine in conjunction with the testimony of an expert witness.” To reach this conclusion, she also cited Caylor’s statements about how the blood sample was handled, which distinguished Bullcoming’s case from the scenario presented above.
The Court’s reasoning in Bullcoming appeared to extend beyond that of Crawford’s original concern about testimonial statements and into the realm of authentication. Under Rule 901(a) of the Federal Rules of Evidence, “[t]o satisfy the requirement of authenticating or identifying an item of evidence, the proponent must produce evidence sufficient to support a finding that the item is what the proponent claims it is.” For authenticating forensic reports, establishing a robust chain of custody is key. In Bullcoming, the disputed report itself contained the absent analyst’s statements about how the blood sample was handled. In Williams, the chain of custody of the evidence samples in question was established at trial, which was enough for the plurality to set aside the confrontation issues of the original analyst’s absence: “The foundational facts . . . were established not by expert testimony but by ordinary chain-of-custody evidence.”
Taken together, “Williams suggests that if produced under certain stringent conditions — namely, those that comply with sound scientific practice — forensic reports do not violate the Confrontation Clause, and the analyst need not testify.” In the background of the Supreme Court’s Confrontation Clause jurisprudence, forensic analysis has been moving rapidly toward sound scientific practice. In addition to using automated chemistry analyzers for substance-related cases like Melendez-Diaz, Bullcoming, and Smith, there has been a push for forensic laboratories to adopt “modern scientific processes and procedures” such as “blind[ing] incoming samples, us[ing] standardized processes, and routinely insert[ing] known test samples into [their] workstream for quality control.”
In an amicus brief supporting the government’s case in Smith, the American Board of Forensic Toxicologists (ABFT) argued that Smith’s concerns about fraud and incompetence being free flowing in forensic science absent a witness with personal knowledge are not “true Confrontation Clause concerns.” It went on to provide clarity as to how forensic science laboratories and organizations operate, including the measures that are taken to ensure the validity and reliability of results (such as personnel training and instrument calibration). The ABFT brief also detailed how software-based laboratory information management systems (LIMS) document the “chain of custody from the time an item is entered into the system, through analysis, and finally until transfer of evidence is made back to the agency or court.” Such information is “discoverable and is typically provided as part of a standard disclosure.” Even with Smith’s cautioned approach to forensic reports, Williams would suggest that introducing automated readouts reporting a “Positive” result neatly packaged with authenticating disclosures from the LIMS would position analysts as non-accusatory witnesses and sidestep the Confrontation Clause’s reach. With the backlogs in forensic laboratories around the country and thousands of defendants awaiting trial, a future with a fast, impersonal, and automated system for testing samples and introducing results at trial without the need for analysts to testify has its attractive qualities. The utility of testimonies introduced to satisfy the Confrontation Clause will also diminish as each analyst oversees more samples without having to interpret their results.
The concern, however, is that valuable authenticating evidence may decline alongside the Confrontation Clause. In the rigorous application of the Confrontation Clause over forensic reports, courts may have taken for granted the authenticating function of having analysts testify. The Smith Court focused on the issues with Longoni’s surrogate testimony but still recognized the inherent value of having any analyst from the same laboratory testify. In addition to analyzing the actual readings, Longoni had to confirm that Rast carried out the tests properly: “[A]s to each [technique], he stated that the testing had adhered to ‘general principles of chemistry,’ as well as to the lab’s ‘policies and practices’ . . . . [H]e noted, for example, that Rast had run a ‘blank’ to confirm that testing equipment was not contaminated.” The Smith majority pointed out how “[b]ecause Longoni worked in the same lab as Rast, he could testify from personal knowledge about how that lab typically functioned — the standards, practices, and procedures it used to test seized substances, as well as the way it maintained chains of custody.” This reasoning is consistent with how most testimonies involve a mixture of subjective judgments and standardized processes; even as more laboratories transition into automated testing, human intervention is still required. This is most obvious in equipment calibration and maintenance, but the Court’s tone would suggest that these checkpoints are secondary to chain of custody. If the Confrontation Clause required neither Rast nor Longoni’s testimonies, say, if Rast had merely transcribed a “Positive” reading, the authenticating evidence may not have been introduced at all.
For drug evidence like the reports in Smith, courts apply different standards of admissibility. Even when forensic scientists commit misconduct, some courts do not presume falsity in every case on which they worked. When it comes to authenticating such evidence, the Melendez-Diaz Court made clear that its ruling did not mean “that anyone whose testimony may be relevant in establishing the chain of custody, authenticity of the sample, or accuracy of the testing device, must appear in person as part of the prosecution’s case.” Any gaps in authentication go to the weight of the evidence as opposed to the admissibility because “[i]t is up to the prosecution to decide what steps in the chain of custody are so crucial as to require evidence; but what testimony is introduced must (if the defendant objects) be introduced live.” However, since “most reports in criminal cases are prepared by government laboratories, they often qualify as self-authenticating documents, and thus may be admitted without extrinsic evidence.”
Only time will tell whether courts will still apply the Confrontation Clause to testimonies that resemble more machine than man. In a future where these procedural safeguards may no longer apply to forensic analyses, what is an indigent criminal defendant to do when the evidence introduced against them is an automated machine readout that was produced as one among thousands? Can they rely on the prosecution to introduce authenticating evidence alongside conclusive forensic reports? The ease with which chain of custody information can be obtained is promising, but without an affirmative burden imposed by the Confrontation Clause, it is unclear whether criminal defendants across the country will receive the same treatment. Today, Smith provides clarity about where the boundaries of the Confrontation Clause are drawn when laboratory results require interpretation. But with the increasing availability of conclusive non-testimonial forensic reports, courts cannot continue taking authentication for granted and enabling “the principal evil at which the Confrontation Clause was directed”: “ex parte examinations as evidence against the accused.”
The post Machines Need Not Testify: Forensic Reports Under the Confrontation Clause appeared first on Harvard Law Review.