Monthly Archives: March 2010

California Supreme Court Exempts Negligent Inclusion of Individuals in the DNA Database from the Exclusionary Rule

In United States v. Leon, 468 U.S. 897 (1984), the Supreme Court adopted a narrow, “good-faith exception to the Fourth Amendment exclusionary rule” for “reliable physical evidence seized by officers reasonably relying on a warrant issued by a detached and neutral magistrate.” Later cases have applied this exception to a police officer’s “reasonable reliance” in making arrests or executing searches on the basis of information or authorization from courts or other government officials. 1/ But if the exception is not to swallow the rule, the Leon exception should not apply when the police department conducting the unreasonable search or seizure is itself the source of the mistake that appears to justify the search. In this situation, it can hardly be said that “there is no police illegality and thus nothing to deter.” 2/

In People v. Robinson, 47 Cal. 4th 1104 (2009), however, the Supreme Court of California unanimously adopted this radical expansion of Leon. And it did so in a case in which it did not need to address the issue. In 1999, when Paul Eugene Robinson “was in custody at [a detention center] for two misdemeanor convictions and awaiting transfer to state prison based on a parole revocation [for a] burglary,” California’s DNA database law went into effect. Although Robinson’s crimes did not qualify him for inclusion in the new database, “an unknown person in the Center‟s records department . . . mistakenly identified [him] as a prisoner with a qualifying offense . . . . As a result of that mistake, a [blood] sample . . . was drawn . . . .” The California Department of Justice Laboratory analyzed the sample, uploaded the DNA profile, and got a cold hit to the man wanted on a “John Doe” DNA warrant for “five felony sexual offenses, all perpetrated against Deborah L. on August 25, 1994.” A jury convicted Robinson of these offenses, the California Court of Appeal affirmed, and the state supreme court granted review.

The California Supreme Court held that an arrest warrant for “John Doe, unknown male” with a particular 13-locus STR profile was valid and thus tolled the statute of limitations on rape prosecutions. It held that the erroneous extraction of blood violated state law, but not the Fourth Amendment. For good measure, it added that even if there had been a constitutional violation, it would not have mattered under the Leon good-faith exception to the exclusionary rule. Although there is reason to doubt the Court’s conclusion that the statutorily unauthorized seizure of blood was consistent with the Fourth Amendment, my target here is the pointed rejection of the exclusionary rule. The United States Supreme Court has never approved the admission of evidence seized in violation of the Fourth Amendment by a police agency relying on Its own, mistaken information, and no court should do so.

The Robinson Court relied largely on the recent–and controversial–decision in Herring v. United States, 555 U. S. __ (2009). 3/ In Herring, police officers arrested and searched a suspect and his vehicle. The search uncovered contraband and an illegally possessed weapon. The officers lacked probable cause to detain or search the suspect; however, a police clerk in a neighboring county had advised them that a current warrant called for the suspect’s arrest. In fact, the court had recalled the warrant. The police database did not reflect this fact. Regarding the false report as a single act of “nonrecurring and attenuated negligence,” a bare majority of the Court held that the application of the exclusionary rule was not warranted. Four Justices objected that the extension of Leon to acts of distinct but cooperating police agencies was “no occasion to further erode the exclusionary rule.” 4/

Despite the division within the Court over Herring, one thing is clear. In every Supreme Court case that has treated an officer’s reliance of erroneous information as grounds for suspending the exclusionary rule, the information has come from an unrelated and apparently reliable governmental source. 5/ In these narrow circumstances, courts may balance “the culpability of the police [against] the potential of exclusion to deter wrongful police conduct.” 6/ Within these boundaries, ordinary and “isolated negligence” normally is not enough to warrant exclusion. 7/

To apply this balancing test outside these confines, however, would open every case of a Fourth Amendment violation arising from inaccurate information supplied by fellow police officials to complicated litigation over how the balance should be struck in light of the facts of the case. Courts would need to draw a difficult line between simple negligence and “deliberate, reckless, or grossly negligent conduct,” or between “isolated” negligence and “recurring or systematic negligence.” 8/ Being unable to predict the outcome of a Fourth Amendment violation, police would have sharply reduced incentives to comply with the amendment and to seek judicial warrants. They could be tempted to avoid the dictates of the amendment by dividing up investigations so that each officer can rely on a report from a colleague rather than pursue the investigation in a more direct fashion. The resulting regime would benefit neither the public, the police, nor the courts.

The novel theory that a police agency may rely on its own negligence to avoid the exclusionary rule deviates from Herring‘s express requirement that the negligence be not merely “isolated” (itself a sharply contested proposition in Robinson) but “attenuated.” 9/ As we have seen, in Herring and in every other case applying Leon to admit evidence, the negligent misstatement was attenuated in the sense that a police officer reasonably relied on plausible information from an independent government agency. In Robinson, however, the correctional facility misinformed itself. To allow such nonattenuated misconduct to escape the exclusionary rule would open the courthouse door to widespread, negligent police misconduct in violation of the Fourth Amendment. 10/

Robinson is not the only star in the expanding university of Herring. See, e.g., United States v. Song Ja Cha, No. 09-10147, 2010 WL 775238 (9th Cir. Mar. 9, 2010) (assuming, without analysis, that Herring applies to an unreasonably long, warrantless seizure of a residence to allow officers to obtain a warrant). The Supreme Court needs to arrest this inflation by explicitly confining the Leon exception to reasonable reliance by the police on generally accurate information or judgments from unrelated government officials. A police agency should not be permitted to rely on its own, negligent mistakes to escape the century-old rule of Weeks v. United States, 232 U. S. 383, 398 (1914), that “forbids the use of improperly obtained evidence at trial.” 11/


1. Arizona v. Evans, 514 U.S. 1 (1995) (reliance on court clerk); Illinois v. Krull, 480 U. S. 340 (1987) (reliance on state statute), Herring v. United States, 555 U. S. __ (2009) (reliance on police clerk).

2. Leon, 468 U.S. at __.

3. See generally Albert W. Alschuler, Herring v. United States: A Minnow or a Shark?, Ohio St. J. Crim. L. 463 (2009); Wayne R. Lafave, The Smell of Herring, 99 J. Crim. L. & Criminology 757 (2009); The Supreme Court 2008 Term, 123 Harv. L. Rev. 153 (2009) (criticizing Herring).

4. 555 U.S. at __, __ (concurring opinion of Ginsburg, Stevens, Souter, & Breyer, JJ.).

5. See Leon (a judicial officer); Krull (a legislature whose enactments enjoy a presumption of constitutionality); Evans (judicial staff); Herring (a records clerk at another police department).

6. Herring, 555 U. S. at __.

7. Id. at __.

8. Id. at __.

9. Id. at __.

10. See Albert W. Alschuler, Herring v. United States: A Minnow or a Shark?, 7 Ohio St. J. Crim. L. 463, 463 (2009) (discarding the attenuation requirement “would leave most violations of the Fourth Amendment without a remedy [and] would create a regime in which courts would make most of their Fourth Amendment rulings in dictum if they decided Fourth Amendment questions at all.”).

11. Herring, 555. U.S. at __,

Up in Smoke: 5 Million Neonatal Blood Samples Incinerated

An effort to avoid the pointless destruction of the millions of Guthrie cards maintained by the Texas Department of Health Services has come to naught. Plaintiffs who sued the department as well as privacy advocates initially were open to the idea of preserving all the cards with neonatal bloodspots for future research while seeking consent for their storage from millions of parents. [1]

However, this was not to be. After the state quickly settled the dubious lawsuit, an enterprising but inadequately informed journalist published Internet stories alleging that the department had turned “over hundreds of dried blood samples to the federal government to help build a vast DNA database–a forensics tool designed to identify missing persons and crack cold cases” and that the samples “were forwarded along to the federal government to create a vast DNA database, one that could help crack cold cases and identify missing persons.” [2] In her latest installment of this tall tale she continues to write that the samples will “help identify missing persons and crack cold cases.” [1]

The suggestion that the U.S. military is using the samples to build a database to “crack cold cases” or to identify “missing persons” in this country is preposterous. To summarize a previous posting [2]: First, the research project is limited to mitochondrial DNA, which rarely is used in forensic investigations because it is not capable of providing specific identification. Second, AFDIL does not maintain any databases of DNA profiles to crack cold cases. Third, even if AFDIL were authorized to maintain a database of civilian DNA profiles for criminal investigations, a collection of nameless mtDNA sequences from de-identified samples would be pretty useless. Finally, the true purpose of the research is clear from “Federal MtDNA Paper” posted on the Tribune‘s website. The AFDIL paper explains that the research database, which cannot be used to identify individuals, simply allows geneticists to put estimates of random-match probabilities for mtDNA on a sounder footing. These estimates are necessary to understand the probative value of an mtDNA match in any criminal investigation or trial. They have nothing in particular to do with cold hits or missing persons.

In sum, the research database has virtually no meaningful privacy implications. Some parents might not want their children’s blood samples used to improve the criminal justice system, but that alone is not much of a reason to destroy what the article calls a medical “treasure trove.” The children’s DNA is not going into any military or law-enforcement database for tracking down missing persons or cracking cold cases.

Yet, this fear apparently was the monkey wrench that jammed the effort to preserve the samples while seeking consent. Here are some excerpts from the latest news as described by the same journalist:

[T]he Department of State Health Services . . . agreed in December to destroy the blood spots, after a civil rights attorney and several Texas parents sued the state for storing them for research purposes without permission. But after the court settlement was signed, privacy advocates lobbied the agency for an alternate solution: a research database that would keep the blood spots intact while seeking electronic consent from parents. They got the go-ahead from some key lawmakers and from the lawsuit’s plaintiffs, who pledged to void the settlement, but not from DSHS.

When The Texas Tribune discovered last month that state health officials had turned hundreds of baby blood spots over to a federal Armed Forces lab between 2003 and 2007 to build a mitochondrial DNA database . . . any chance for saving the blood spots fizzled out. All 5 million blood spots were sent to a Houston-area incinerator last week.

“If there was any way the blood spots were going to be saved, the whole thing fell apart at that point,” said state Sen. Bob Deuell, R-Greenville, . . . “When this came out about these specimens going to the military, I said, ‘We’ve lost this one.'”

. . . State health officials say Austin-based national patient privacy advocate Deborah Peel and Deuell, a physician, approached DSHS Commissioner David Lakey early this year about using electronic consents to save the 5 million existing blood spots from destruction. The agency reviewed the idea but never pursued it. . . .

Critics say . . . DSHS conveniently settled the lawsuit before the trial went to the discovery phase, meaning the documents on the federal DNA study were never disclosed to the plaintiffs. (The Tribune obtained the documents on the federal project — designed to build a forensics tool to help identify missing persons and crack cold cases — through Texas open-records laws.) “Unfortunately, that of course confirmed the plaintiffs’ worst fears,” said Peel, founder of the nonprofit advocacy group Patient Privacy Rights.

Peel said the state’s decision not to seek a non-destructive solution is a shame. . . . “We were going to … reach out to those 5 million families and let them know they had an alternative to having their blood spots destroyed,” Peel said. . . .

Deuell said the impression he got from state health officials was that they feared they would be subject to litigation from other parents if they negotiated with the plaintiffs not to destroy the blood spots. . . . “They said, ‘The plaintiffs are just three people out of 5 million. Who’s to say somebody else wouldn’t come back and file a new suit?'”

Harrington [plaintiffs’ attorney] said that worry is “utter nonsense.” He said both sides could have gone back to the judge to have a new settlement drafted — one that would’ve protected the agency. “What’s the harm in that?” Harrington asked. “We would have supplemented or amended the settlement. It would have been totally possible.”

But once news broke that some of the blood spots had been turned over to the federal lab — and that the state had no intention of destroying those samples — the plaintiffs’ offer was off the table. Instead, they have demanded that the state get the blood spots back from the federal government, or they’ll file another lawsuit. . . . [1]

Maybe another lawsuit would be a good thing. With competent lawyering and journalism, the people of Texas finally might realize that none of their children’s DNA has found its way into any DNA database for identifying anyone.


1. Emily Ramshaw, DNA Destruction, Tex. Tribune,

2. David H. Kaye, A Texas Tall Tale of “DNA Deception,” Double Helix Law, Mar. 4, 2010

A Texas Tall Tale of “DNA Deception”

A “non-profit, nonpartisan public media organization,” the Texas Tribune [1] broke a story that is bound to attract national outrage. The story goes like this. Texas, like every other state, pricks the heels of new born children for a blood sample. It screens these samples for rare, metabolic genetic diseases and stores spots of blood on a card for each child. As the March of Dimes explains, “[w]hen test results show that the baby has a birth defect, early diagnosis and treatment can make the difference between lifelong disabilities and healthy development.” [2]

As these “Guthrie cards” began to accumulate, it became clear that they might be useful for medical research. In 1994, law professor Jean McEwen and doctor-lawyer Phil Reilly called them “inchoate databases” and found that many laboratories were open to the idea of sharing them — in anonymized form — for research that would benefit the public. [3]

The Texas State Department of Health Services did exactly this. It provided medical researchers with de-identified Guthrie cards to study “the gene involved in club foot, to inspect the DNA of infants who develop childhood cancer, [and] to examine prenatal lead exposure.” [4] For its efforts, the department was sued. It had treated the cards as free for the taking, without going back to every pair of parents to obtain explicit permission to release their (nameless) child’s blood spots. Although it is a huge jump from any case law, and even though the legally cognizable damages suffered by any parent whose unknown child’s blood spot made its way to a laboratory are obscure, five plaintiffs alleged violations of the protection of the Fourth Amendment, the Texas Constitution, and the common law. On their behalf and seeking to represent a much larger class of plaintiffs, the Texas Civil Rights Project sought declaratory and injunctive relief. [5]

The case promptly settled. The state agreed to destroy millions of cards, to give parents clearer procedures to opt out of the storage of the cards, and to pay $26,000 in attorneys fees and costs.

There things might have stayed — but for a journalist’s “review of nine years’ worth of e-mails and internal documents on the Department of State Health Services’ newborn blood screening program.” [4] She found that the state had concealed its involvement in a nefarious and far-reaching military or law-enforcement project. The Texas doctors had turned “over hundreds of dried blood samples to the federal government to help build a vast DNA database — a forensics tool designed to identify missing persons and crack cold cases.” [4] The samples, she repeated, “were forwarded along to the federal government to create a vast DNA database, one that could help crack cold cases and identify missing persons.” [6] The database would be shared worldwide, “for international law enforcement and investigation in the context of homeland security and anti-terrorism efforts.” [4]

Incensed, the lawyer for the five plaintiffs fired off a letter to the governor and the attorney-general. He accused the “TDSHS [of] supplying those blood samples taken from newborn babies to the military, not just for research, but so that the military can build a mitochondria DNA data base, which can be used in part for law enforcement purposes.” [6] He complained that “[t]his … alarming development … raises the specter of the federal government building an international DNA data base,” and he demanded that “within ten (10) days of this letter, you retrieve from the federal government all the blood samples that Texas has sent to the U.S. military and retrieve and destroy all information taken from those samples … .” [6] Indeed, he suddenly realized that this military project was why the state was so willing to settle the case: “‘Sometimes there are slam-dunk cases, but I’d never seen this kind of case settle without discovery,’ says [Jim] Harrington, director of the Texas Civil Rights Project. ‘This explains the mystery of why they gave up so fast.'” [4]

The trouble is that it’s all smoke and no fire. The reporter and the lawyer apparently have misread the report of the Armed Forces DNA Identification Laboratory (AFDIL) detailing its efforts to collect and study mitochondrial DNA (mtDNA) from varied people and places. As explained in Chapters 11 of The Double Helix and the Law of Evidence, AFDIL is a world leader in mitochondrial DNA sequencing because the technique is exceedingly valuable in identifying the remains of soldiers missing in action. [7] But mtDNA is not used to “crack cold cases,” at least not by generating cold hits in any law-enforcement database of DNA profiles from possible offenders. The national database (NDIS) maintained by the FBI — the one that actually helps in cracking cold cases — is limited to STR profiles in the DNA from the cell nucleus. These DNA sequences are wonderful for discriminating among individuals. When a 13-locus match from a crime-scene to one of the more than seven million profiles in NDIS pops up, it can constitute a practically conclusive identification to a known individual. [7] And, the bigger NDIS is, the more likely it is that the culprit will be in it. This kind of database is “only as valuable as its … size.” [4]

Not so with mtDNA. Everyone in the same maternal line shares the same sequence, and other essentially unrelated maternal lineages might have the same sequences. [7] Moreover, it would be inane to put anonymous sequences — nuclear or mitochondrial — into the database used in searching for cold hits. A hit from a crime-scene sample to a profile from a Guthrie card with no name attached to it would have little or no investigative value. The (nameless) Texas children need not fear being swept up in criminal or terrorist investigations because AFDIL sequenced their anonymous DNA.

But if the federal government does not want the samples for a database that will be used to catch criminals or terrorists, what nefarious international database are these profiles going into? Prosaically, they are part of a scientific, population-genetics database that will be helpful in understanding the significance of a match in an ordinary criminal case. Consider State v. Ware, the very first case with mtDNA evidence. Hairs were found in the bed where a young girl was attacked. [7, chap. 12] The hairs looked similar to the defendant’s under a microscope, but there have been false convictions with hairs that happen to look similar. (Just check with the Innocence Project.) Nuclear DNA, which could yield well-nigh conclusive results, were absent in the hair shafts, but there were enough mitochondria to get a useful sequence, and this sequence matched the defendant’s. [7]

Because mtDNA just does not have the power of nuclear DNA to differentiate among individuals, however, defense counsel in such cases can object (appropriately) that the evidence is confusing or misleading without statistics on how rare the mitotype in question would be in the general population. How many people would be falsely incriminated by the mtDNA sequence in the case?

By understanding the variations in the mtDNA sequences in different places and populations, scientists can estimate how rare or how common a mitotype that incriminates a suspect might be. Such estimates require reference databases, but the existing forensic-statistical-reference databases, defense counsel and a number of scientists have argued, are too small  and full of gaps in the population groups represented. [7] Indeed, the federal government has received considerable flak from the media and a vocal group of scientists, lawyers, and sundry others for its refusal to supply de-identified nuclear-DNA profiles from law-enforcement databases for new studies to supplement the existing statistical-reference databases long used to estimate the probability of random STR-profile matches in criminal cases. [9, 10]

In sum, the AFDIL study is a response to a legitimate scientific and legal concern. The federal government (as it should) wants to improve the infrastructure for using mtDNA evidence in court by enlarging the statistical-reference databases. Thus, the AFDIL report — the supposed smoking gun posted on the Tribune‘s website — is entitled “Development and Expansion of High-quality Control Region Databases to Improve Forensic mtDNA Evidence Interpretation.” As the title indicates, these scientific databases do not generate DNA evidence. They “improve” the “intepretation” of mtDNA evidence from other sources. The very first sentence of the report makes it plain that the databases are for statistical purposes only:

Mitochondrial DNA testing in the forensic context requires appropriate, high-quality population databases for estimating the rarity of questioned haplotypes. However, large forensic mtDNA databases, which adhere to strict guidelines in terms of their generation and maintenance, are not widely available for many regional populations of the United States or most global populations outside of the United States and Western Europe.

After elaborating, the report continues:

In order to address this issue, the Armed Forces DNA Identification Lab (AFDIL) has undertaken a high-throughput control region databasing effort. … Global populations that are currently underrepresented in available forensic mtDNA databases will comprise approximately 25% of the total number of samples. The remaining individuals will represent regional samples of various U.S. populations and global populations that contribute to the overall mtDNA diversity of the U.S. The high-quality mtDNA data generated from these efforts will be publicly available to permit examination of regional mtDNA substructure and admixture, and ultimately to improve our ability to interpret mtDNA evidence.

This population-genetics study is entirely different from building a huge database of mitotypes to generate cold hits. MtDNA does not work well for this purpose, and even if the FBI wanted to do it, anonymous data from AFDIL would be useless. All that those data can do is help investigators, judges and juries better assess the results of a match to a known suspect or defendant. Suggestions that neonatal samples are being put into databases that could result in the unknowing “donors” being swept up in future investigations of crime or terrorism are troubling — but not because they are true.


[1] Texas Tribune, About the Texas Tribune,

[2] March of Dimes, Newborn Screening Tests, Mar. 2008,

[3] J. E. McEwen & P. R. Reilly, Stored Guthrie Cards as DNA “Banks,” 55 Am. J. Human Genetics 196-200 (1994), available at

[4] Emily Ramshaw, DNA Deception, Texas Tribune, Feb. 22, 2010, available at, last viewed, March 2, 2010

[5] Beleno v. State Dep’t of Health Serv., Civ. No. SA09CA1088 (W.D. Tex. Mar. 12, 2009) (complaint)

[6] Emily Ramshaw, TribBlog: AG’s Office Fires Back at Blood Spot Attorney, Feb. 22, 2010, available at, last viewed, March 2, 2010

[7] David H. Kaye, The Double Helix and the Law of Evidence (2010)

[8] Jodi A. Irwin et al., Development and Expansion of High-quality Control Region Databases to Improve Forensic mtDNA Evidence Interpretation, 1 Forensic Sci. Int’l: Genetics 154-157 (2007)

[9] David H. Kaye, Trawling DNA Databases for Partial Matches: What Is the FBI Afraid Of?, 19 Cornell J. L. & Public Pol’y 145-171 (2009)

[10] D. E. Krane et al., Time for DNA Disclosure, 326 Science 1631-1632 (2009), DOI: 10.1126/science.326.5960.1631

� 2010 David H. Kaye

Another expert succumbs to the transposition fallacy

A book that attempts to inform defense lawyers on how to handle DNA cases is Dealing with DNA Evidence: A Legal Guide (London: Routledge-Cavendish 2007). In this short primer, Andrei Semikhodskii, Director of Medical Genomics, Ltd., explains that “[u]nderstanding how DNA evidence is obtained and evaluated helps lawyers to find pitfalls in evidence and in data interpretation … .” (P. xi).

Fair enough, but the burden on a book whose purpose is to provide accurate explanations is a heavy one. A common mistake in DNA and other statistical testimony is transposition — mistaking the probability of the evidence given a hypothesis, P(E|H), for the probability of the hypothesis given the evidence, P(H|E). (See the blog of January 18, 2010, on McDaniel v. Brown.) A variation on the transposition fallacy occurs in parentage tests. Dr. Semikhodskii’s laboratory advertises the “world’s most accurate  paternity testing,” but Dealing with DNA Evidence is less than pellucid when it explains that

DNA testing does not give a 100 per cent probability of confirming parentage. When biological parentage is possible, its likelihood is estimated by the CPI [Combined Parentage Index] The value of the CPI indicates how many more times the alleged parent is likely to be the true biological parent of the child than in comparison to an untested unrelated individual from the same population.  (P. 45).

Apparently, the book is referring to a likelihood ratio for the hypothesis that the tested man is the father as against the hypothesis that an unknown man (with no close genetic relationship to the accused) is. But a likelihood ratio that takes on some value x does not mean that the tested man is x times more likely to be the father than is the untested man. It means that the genetic data are x times more likely to arise if he is the father.

Not clear? Well, suppose a ridiculously limited genetic test indicated that a child is 10 times more likely to inherit a genotype from his mother and the putative father than from his mother and a randomly selected man (of equal fertility). Does this mean that the putative father is ten times more likely than Mr. Random to be the biological father? It cannot mean this (in general). After all, if the putative father were up in the International Space Station (and the mother was not) during any plausible date of conception, the likelihood ratio would still be 10. Geneticists can compute the chances of a child’s inheriting various alleles if and when a given man is the father. Even with the best paternity test in the universe, the laboratory cannot compute the chance that the man is the father just by knowing the alleles the child inherited from his father.

Therefore — and contrary to this expensive guide for lawyers — the likelihood ratio does not “show how many times more plausible the prosecution hypothesis is given the DNA evidence.” (P. 76). The ubiquitous transposition fallacy is at work here, as it is in the case law. (I discuss some cases involving such transposition in the likelihood ratio in The Modern Wigmore on Evidence: Expert Evidence.)

This confusion between a “likelihood” P(E|H) (the probability of data given a hypothesis) and a “posterior probability” (that the hypothesis is true given evidence in support of that hypothesis) infects a later discussion of the rule that “[t]he expert should not be asked his opinion on the likelihood that it was the defendant who left the crime stain … .” R. v. Doheny [1997] 1 Cr. App. R. 369. Dr. Semikhodskii thinks that “in contravention of this ruling, almost every DNA report submitted to courts does contain the verbal expression of how much support is to be given to the prosecution hypothesis and in most cases this is allowed to be admitted and aired in front of the jury.” (P. 60). But if “what is admitted and aired” is merely a likelihood ratio and a characterization of its magnitude in English, the expert is not giving “an opinion on the likelihood that it was the defendant who left the crime stain.” An expert who states that it is, say, 100,000 times more likely for certain evidence to arise when the defendant really is the source than otherwise and that this means that the evidence gives “very strong support” to this hypothesis is avoiding rather than offering a statement about the source probability.

Somehow or other, the expert must explain the strength of the evidence to the jury, and classifying it as weak or strong is one way to do it. Indeed, a committee of the U.S. National Academy of Science recently recommended that forensic scientists use such standardized terminology to characterize evidence. The problem with this recommendation is not that it invades the province of the jury by directly expressing an opinion on an ultimate issue, but that the verbal predicate is superfluous. If the expert can state the numerical value of the likelihood ratio — the quantity that measures the strength of the evidence rather than the probability of the hypothesis — then what does adding an arbitrary but standard adjective accomplish?

Let’s hope there is a better guide for lawyers.