Handwriting, fingerprints, and the admissibility of evidence

[minty_green]

Re: Is handwriting analysis legit science?

Just some short follow-ups and additional info on the legal points Cecil mentions in the column.

First of all, the case in which the federal court ruled against the admissibility of the government’s handwriting analyst’s testimony is United States v. Saelee, 162 F. Supp. 2d 1097 (D. Alaska 2001). It’s a federal district court case that was apparently not appealed to the 9th Circuit. Its value as precedent is, like handwriting analysis, kinda shaky. It’s been cited a few other times by other federal district court cases, but those cases go both ways. Here’s a neat little summary I found out of U.S. v. Crisp, an opinion out of the Fourth Circuit issued just a couple of weeks ago:

While the admissibility of handwriting evidence in the post Daubert world appears to be a matter of first impression for our Court, every circuit to have addressed the issue has concluded, as on the fingerprint issue, that such evidence is properly admissible. See United States v. Jolivet, 224 F.3d 902, 906 (8th Cir. 2000) (citing Eleventh Circuit’s Paul decision and upholding admission of expert handwriting testimony); United States v. Paul, 175 F.3d 906, 911 (11th Cir. 1999) (emphasizing “flexible” nature of district court’s gatekeeping function, and noting that “the ability of the jury to perform the same visual comparisons as the experts cuts against the danger of undue prejudice from the mystique attached to experts” (internal quotation omitted)); United States v. Jones, 107 F.3d 1147, 1161 (6th Cir. 1997) (upholding admission of expert handwriting testimony and observing that “just because the threshold for admissibility [of expert testimony] under Rule 702 has been crossed, a party is not prevented from challenging the reliability of the admitted evidence”); United States v. Velasquez, 33 V.I. 265, 64 F.3d 844 (3rd Cir. 1995) (discussing standard methodology applied by handwriting analysts, and upholding admission of expert handwriting testimony). n5

• • • • • • • • • • • • • • • • • -Footnotes- - - - - - - - - - - - - - - - - -

n5 Certain district courts, however, have recently determined that handwriting analysis does not meet the Daubert standards. See, e.g., United States v. Lewis, 220 F. Supp. 2d 548, 554 (S.D. W. Va. 2002) (finding proficiency tests and peer review meaningless where the evidence showed that handwriting experts “always passed their proficiency tests, . . . [and that] peers always agreed with each others’ results” (emphasis in original)); United States v. Brewer, 2002 U.S. Dist. LEXIS 6689, 2002 WL 596365 (N.D. Ill. 2002); United States v. Saelee, 162 F. Supp. 2d 1097 (D. Alaska 2001); United States v. Hines, 55 F. Supp. 2d 62 (D. Mass. 1999).

Crisp, by the way, affirmed a district court’s decision to admit expert testimony on both fingerprints and handwriting.

I also found a case from the 10th Circuit last year called *U.S. v. Hernandez*, which sort of splits the difference. In that case, the district court had allowed the handwriting expert to testify about the similarities between the defendant’s known handwriting and the fraudulent tax documents at issue in the case. However, the trial court refused to allow the expert to offer his opinion about whether the handwriting was from the same person. The defendant was convicted, and the conviction was affirmed on appeal.

The case Cecil mentions regarding fingerprint analysis is United States v. Llera Plaza, 179 F. Supp. 2d 492 (E.D. Pa. 2002). What Cecil doesn’t mention is that a couple months later, the court changed its ruling and decided to admit the fingerprint analysis testimony after all. That opinion is United States v. Llera Plaza, 188 F. Supp. 2d 549 (E.D. Pa. 2002).

The consensus of the courts so far–and this has really only been an issue for the last couple of years–seems to be that fingerprint analysis is going to be admissible if the testimony is from an expert qualified in the field, but they might start to reconsider that conclusion if new data comes in indicating that it is not a reliable means of identifying a person. It’s a bit odd, actually. Under the Daubert standards, it is the proponent of the expert testimony that has to prove the testimony is reliable; with fingerprints, they’re sort of saying it’s been around so long that they’re going to make the opponent prove it’s not reliable.

Anyway, that ends today’s review of Federal Rule of Evidence 702.

[qdman]

Minty,

Thanks for saving me a lot of typing. You did a great job of explaining. I hope Cecil read it.

For everyone else, links to some of these and other cases involving forensic document examination can be found at:

www.abfde.org

and

www.forensic-documents.com

[ElJeffe]

Regarding this bit of Cecil’s response:

Prosecutors scrambling to find scientific validation for handwriting analysis last year touted a study by Sargur Srihari, a professor of computer science at the State University of New York at Buffalo. Srihari subjected 1,500 writing samples to computer analysis. Conclusion: In 96 percent of cases, the writer of a sample could be positively identified based on quantitative features of his handwriting such as letter dimensions and pen pressure. Skeptics objected that lab results using a computer prove nothing about what a human can do in the real world, and who can argue?

It would seem that if the computer analysis had a 96% success rate, that would imply - at the least - that handwriting analysis was a valid science, even if something more meticulous than the human eye is required to pick out the differences. Is it possible that courts may begin to only admit computer analyses of handwriting, rather than human analyses, or is the process of having a computer to the work prohibitively time-consuming thus far?
Jeff

[C_K_Dexter_Haven]

Well, a 96% success rate is not usually very good for expert testimony. “Ladies and Gentlemen of the jury, we are 96% certain that the changes to the will were made by the deceased.”
“Ah, but that means that, out of a hundred people, there are four others who might have made those changes? So, in the Metropolitan NY area, that’s about 400,000 people who might have made those changes to the will, whose handwriting would be indistinguishable from the deceased?”

[Sparteye]

That an item of evidence is not conclusive does not mean it is inadmissible. Rarely is any evidence conclusive, including scientific evidence.

“Relevant evidence” means evidence having any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence."

– FRE 401; MRE 401*
Certainly, evidence with a 96 percent probability makes the existence of the pertinent fact more or less probable than it would be without the evidence.

Relevant evidence can nonetheless be excluded:

“Although relevant, evidence may be excluded if its probative value is substantially outweighed by the danger of unfair prejudice, confusion of the issues, or misleading the jury, or by considerations of undue delay, waste of time, or needless presentation of cumulative evidence.”

– FRE 403; MRE 403

Because scientific evidence tends to be allocated significant weight by juries, the possibility of unfair prejudice or misleading the jury perhaps is stronger than it is regarding other types of evidence, and so special considerations have developed in ascertaining the admissibility of scientific evidence. Underlying it all, however, is the basic rule that relevant evidence is admissible, even if not conclusive.

*FRE = Federal Rule of Evidence; MRE = Michigan Rule of Evidence

[minty_green]

Sparteye nailed it, Dex. It’s kinda like bloodtype evidence. If the perpetrator and the defendant both have type A+ blood, that fact is going to be admissible, even though something like 25% of the population has type A+ blood. The jury may give those facts due consideration and, in conjunction with the rest of the evidence and testimony, determine whether the defendant is the perpetrator or whether the deceased made the handwritten changes to the will.

[Duck_Duck_Goose]

minty’s in comments, minty’s in comments!

We are not worthy…

[minty_green]

What, you don’t remember my all-time classic, Copyright map trap identified in Dallas?

[C_K_Dexter_Haven]

Just for clarification, minty: I didn’t say that the evidence wasn’t admissible, I only said that it wasn’t very good – the other side can demolish it. Just as with your example of blood type – it can be indicative, but it wouldn’t be hard for the other side to arouse “reasonable doubt” if there’s only a 25% (or, in the handwriting case, 96%) chance of being a correct match.