My mom, still right about everything. In a previous post, I explained her immediate skepticism about the Brandon Mayfield arrest. From the NYT today:
…the F.B.I. at one point told federal prosecutors that Spanish officials were “satisfied” with their conclusion. But in interviews this week, Spanish officials vehemently denied ever backing up that assessment, saying they had told American law enforcement officials from the start, after their own tests, that the match was negative. The Spanish officials said their American counterparts relentlessly pressed their case anyway, explaining away stark proof of a flawed link — including what the Spanish described as tell-tale forensic signs — and seemingly refusing to accept the notion that they were mistaken….
Carlos Corrales, a commissioner of the Spanish National Police’s science division, said he was also struck by the F.B.I.’s intense focus on Mr. Mayfield. “It seemed as though they had something against him,” Mr. Corrales said, “and they wanted to involve us.”
The FBI continues to maintain it was just a random mistake by an examiner who didn’t even know Mayfield’s name, much less his religion, that initially led them to focus on Mayfield. I continue to maintain that’s total BS. Finally, does this fingerprint examiner have a family? Because I bet they would really, really like more time to be spent with them.
Statistical Update: This 2001 Washington Post article lists some widely varying estimates as to how many Muslims live in the US. The highest number was produced by a group of Muslim organizations and has been the subject of some doubt (numbers in millions).
Mosque Study Project: 6 to 7
2001 Britannica Book of the Year: 4.1
National Opinion Research Center: 1.5 to 3.4
CUNY Religious Identification Survey: 2.8
Reading the article, the methodology of the Mosque Study Project was obviously pretty bad. The total U.S population, according to the census bureau, is 293, 425, 566. So it seems as if probably more than 1% but substantially fewer than 2% of Americans are Muslims. I think that in the original article the FBI said the computer provided them with 50 close matches, from which Mayfield’s print was chosen as the best by an examiner (again, allegedly, without reference to his personal details).
Although motives are difficult, some statistical analysis is helpful sometimes. How many people does the FBI have fingerprints? 1 million? How many of those people are Muslim? 50,000? How many Muslim converts: 5,000? How many high profile Muslim converts: 500? If it was a genuine mistake, the chances are 1 in 20 the person picked up would be not Muslim: 95%. If 1 person in 2,000 is a high profile Muslim convert, then the chances of a mistake selecting them are slight. Or were they only checking Muslim fingerprints? What is strange about the case is that if the fingerprint doesn’t match now, why did it match then? What has changed? Does anyone have a photo of the two fingerprints so we can compare?
I don’t know of any pix, but if you read the article the Spanish investigators detail what a poor match it was: wrong number of loops, ridges turning up instead of down, that sort of thing. They gave it 7 points of similarity, I think, while the FBI claimed 15, but actually there is no one accepted standard for the number of points which need to match. This is especially true of latent partial prints. They claimed to have run it through a database of 44 million prints IIRC; they had Mayfield’s because of Army service. Given how few Muslims there are of any stripe in the states I think the odds that a random match from those 44 million would be Muslim are so slim as to defy credulity. I guess stranger things have happened.
Mr. Mayfield’s prints were in the F.B.I.’s central database of more than 44 million prints because they had been taken when he joined the military, where he served for eight years before being honorably discharged as a second lieutenant.I don’t think this is actually crucial. The vital question is the second one posed by q. How likely it is that a mis-identification would pick up a Muslim with some sort of dubious connection (convert status isn’t critical, but is probably one of the factors that would count against someone)?
I’d guess about 2 per cent of the US population is Muslim and of those maybe 1 in 10 would have some sort of link that might give rise to suspicion, giving a 1 in 500 chance of this thing happening entirely by accident.
My guess is that the examination process spat out quite a few partial matches (say 50 to 100) and that Mayfield was the only one who produced any red flags when they cross-checked.
As the story points out, this episode is going to be very damaging for the FBI’s use of fingerprint evidence, and will almost certainly force them to tighten up criteria and rely less on “expert” judgement.
The 9/11 commission is preparing to hand the FBI its ass for having failed to connect various dots prior to Sept. 11. Under that circumstance, if a list of 50-100 pops up a Muslim convert with at least some small ties to one of the Portland Six, it’s no surprise that the FBI is going up jump on that person. The only way of assuring no false positives is to assume, like Belle’s mother, that any identification is wrong, and the FBI’s incentives don’t lean that way these days.
Think of it this way, Belle: if you’re looking for a person with ties to an al-Qaeda bombing, what are the odds that that person is not going to be a Muslim? Certainly, the person that Spain is now naming as belonging to that fingerprint is a Muslim.
The only way of assuring no false positives is to assume, like Belle’s mother, that any identification is wrong
Although, in this case, apparently double-checking Mayfield’s fingerprints against the one found in Spain would have sufficed.
Lets not forget that Portland is a hotbed of muslim militancy. Also, this guy is a muslim convert (likely a bit of a cipher/psychotic).
I’m not comfortable with what the government did, but I’m also not comfortable with this idiot living in america.
Raymond sez: The guy is a convert to Islam — and hence is likely psychotic or tending towards. I am not comfortable with this idiot living in America.
I say: Raymond — how do you feel about converts to Lutheranism? Judaism? Agnosticism? I am not comfortable with an idiot of your sort living in America.
tom t: if a list of 50-100 pops up a Muslim convert with at least some small ties to one of the Portland Six, it’s no surprise that the FBI is going up jump on that person. Quite agree with you - just seems a bit amazing that out of 44 million checks, assuming 100 hits, with 500 high profile Muslims on the books (1 in 88,000) that one of them appeared at all. Using my bad maths I calculate the chance that a selection of 100 random people from a database of 44 million wont include someone from a subset of 500 people is 99.886% which is approximately 1 in 1000. If they have done 500 similar fingerprint checks this would reduce the odds dramatically, and it suddenly would not be so incredible.
In order to know any more, we need to understand the frequency and methodology of the fingerprinting operation.
Muslim converts are creepy and often zealots. There’s way too many better religions out there. We have ufo religions and such; surely no sane person converts to the religion of hatred, violence and oppression.
Muslim converts are creepy and often zealots. There’s way too many better religions out there. We have ufo religions and such; surely no sane person converts to the religion of hatred, violence and oppression.
Just for the record, my comment above should not be read to imply that I agree with Raymond’s views, which I expressly disavow.
À Gauche
Jeremy Alder
Amaravati
Anggarrgoon
Audhumlan Conspiracy
H.E. Baber
Philip Blosser
Paul Broderick
Matt Brown
Diana Buccafurni
Brandon Butler
Keith Burgess-Jackson
Certain Doubts
David Chalmers
Noam Chomsky
The Conservative Philosopher
Desert Landscapes
Denis Dutton
David Efird
Karl Elliott
David Estlund
Experimental Philosophy
Fake Barn County
Kai von Fintel
Russell Arben Fox
Garden of Forking Paths
Roger Gathman
Michael Green
Scott Hagaman
Helen Habermann
David Hildebrand
John Holbo
Christopher Grau
Jonathan Ichikawa
Tom Irish
Michelle Jenkins
Adam Kotsko
Barry Lam
Language Hat
Language Log
Christian Lee
Brian Leiter
Stephen Lenhart
Clayton Littlejohn
Roderick T. Long
Joshua Macy
Mad Grad
Jonathan Martin
Matthew McGrattan
Marc Moffett
Geoffrey Nunberg
Orange Philosophy
Philosophy Carnival
Philosophy, et cetera
Philosophy of Art
Douglas Portmore
Philosophy from the 617 (moribund)
Jeremy Pierce
Punishment Theory
Geoff Pynn
Timothy Quigley (moribund?)
Conor Roddy
Sappho's Breathing
Anders Schoubye
Wolfgang Schwartz
Scribo
Michael Sevel
Tom Stoneham (moribund)
Adam Swenson
Peter Suber
Eddie Thomas
Joe Ulatowski
Bruce Umbaugh
What is the name ...
Matt Weiner
Will Wilkinson
Jessica Wilson
Young Hegelian
Richard Zach
Psychology
Donyell Coleman
Deborah Frisch
Milt Rosenberg
Tom Stafford
Law
Ann Althouse
Stephen Bainbridge
Jack Balkin
Douglass A. Berman
Francesca Bignami
BlunkettWatch
Jack Bogdanski
Paul L. Caron
Conglomerate
Jeff Cooper
Disability Law
Displacement of Concepts
Wayne Eastman
Eric Fink
Victor Fleischer (on hiatus)
Peter Friedman
Michael Froomkin
Bernard Hibbitts
Walter Hutchens
InstaPundit
Andis Kaulins
Lawmeme
Edward Lee
Karl-Friedrich Lenz
Larry Lessig
Mirror of Justice
Eric Muller
Nathan Oman
Opinio Juris
John Palfrey
Ken Parish
Punishment Theory
Larry Ribstein
The Right Coast
D. Gordon Smith
Lawrence Solum
Peter Tillers
Transatlantic Assembly
Lawrence Velvel
David Wagner
Kim Weatherall
Yale Constitution Society
Tun Yin
History
Blogenspiel
Timothy Burke
Rebunk
Naomi Chana
Chapati Mystery
Cliopatria
Juan Cole
Cranky Professor
Greg Daly
James Davila
Sherman Dorn
Michael Drout
Frog in a Well
Frogs and Ravens
Early Modern Notes
Evan Garcia
George Mason History bloggers
Ghost in the Machine
Rebecca Goetz
Invisible Adjunct (inactive)
Jason Kuznicki
Konrad Mitchell Lawson
Danny Loss
Liberty and Power
Danny Loss
Ether MacAllum Stewart
Pam Mack
Heather Mathews
James Meadway
Medieval Studies
H.D. Miller
Caleb McDaniel
Marc Mulholland
Received Ideas
Renaissance Weblog
Nathaniel Robinson
Jacob Remes (moribund?)
Christopher Sheil
Red Ted
Time Travelling Is Easy
Brian Ulrich
Shana Worthen
Computers/media/communication
Lauren Andreacchi (moribund)
Eric Behrens
Joseph Bosco
Danah Boyd
David Brake
Collin Brooke
Maximilian Dornseif (moribund)
Jeff Erickson
Ed Felten
Lance Fortnow
Louise Ferguson
Anne Galloway
Jason Gallo
Josh Greenberg
Alex Halavais
Sariel Har-Peled
Tracy Kennedy
Tim Lambert
Liz Lawley
Michael O'Foghlu
Jose Luis Orihuela (moribund)
Alex Pang
Sebastian Paquet
Fernando Pereira
Pink Bunny of Battle
Ranting Professors
Jay Rosen
Ken Rufo
Douglas Rushkoff
Vika Safrin
Rob Schaap (Blogorrhoea)
Frank Schaap
Robert A. Stewart
Suresh Venkatasubramanian
Ray Trygstad
Jill Walker
Phil Windley
Siva Vaidahyanathan
Anthropology
Kerim Friedman
Alex Golub
Martijn de Koning
Nicholas Packwood
Geography
Stentor Danielson
Benjamin Heumann
Scott Whitlock
Education
Edward Bilodeau
Jenny D.
Richard Kahn
Progressive Teachers
Kelvin Thompson (defunct?)
Mark Byron
Business administration
Michael Watkins (moribund)
Literature, language, culture
Mike Arnzen
Brandon Barr
Michael Berube
The Blogora
Colin Brayton
John Bruce
Miriam Burstein
Chris Cagle
Jean Chu
Hans Coppens
Tyler Curtain
Cultural Revolution
Terry Dean
Joseph Duemer
Flaschenpost
Kathleen Fitzpatrick
Jonathan Goodwin
Rachael Groner
Alison Hale
Household Opera
Dennis Jerz
Jason Jones
Miriam Jones
Matthew Kirschenbaum
Steven Krause
Lilliputian Lilith
Catherine Liu
John Lovas
Gerald Lucas
Making Contact
Barry Mauer
Erin O'Connor
Print Culture
Clancy Ratcliff
Matthias Rip
A.G. Rud
Amardeep Singh
Steve Shaviro
Thanks ... Zombie
Vera Tobin
Chuck Tryon
University Diaries
Classics
Michael Hendry
David Meadows
Religion
AKM Adam
Ryan Overbey
Telford Work (moribund)
Library Science
Norma Bruce
Music
Kyle Gann
ionarts
Tim Rutherford-Johnson
Greg Sandow
Scott Spiegelberg
Biology/Medicine
Pradeep Atluri
Bloviator
Anthony Cox
Susan Ferrari (moribund)
Amy Greenwood
La Di Da
John M. Lynch
Charles Murtaugh (moribund)
Paul Z. Myers
Respectful of Otters
Josh Rosenau
Universal Acid
Amity Wilczek (moribund)
Theodore Wong (moribund)
Physics/Applied Physics
Trish Amuntrud
Sean Carroll
Jacques Distler
Stephen Hsu
Irascible Professor
Andrew Jaffe
Michael Nielsen
Chad Orzel
String Coffee Table
Math/Statistics
Dead Parrots
Andrew Gelman
Christopher Genovese
Moment, Linger on
Jason Rosenhouse
Vlorbik
Peter Woit
Complex Systems
Petter Holme
Luis Rocha
Cosma Shalizi
Bill Tozier
Chemistry
"Keneth Miles"
Engineering
Zack Amjal
Chris Hall
University Administration
Frank Admissions (moribund?)
Architecture/Urban development
City Comforts (urban planning)
Unfolio
Panchromatica
Earth Sciences
Our Take
Who Knows?
Bitch Ph.D.
Just Tenured
Playing School
Professor Goose
This Academic Life
Other sources of information
Arts and Letters Daily
Boston Review
Imprints
Political Theory Daily Review
Science and Technology Daily Review