Read February-March 2006
This book is a good argument for visiting English-language books when traveling. It isn't available in the US; I found it on the shelf of the excellent British Bookshop in Frankfurt.
On to the topic. We have always ``known'' that fingerprints are unique identifiers. This begs a few obvious questions: How do we actually know that? When did we know it? And when did we realize we could use them exclusively? Like a scientist unraveling a problem, Sengoopta, a professor of the history of science and technology, dissects these questions to reveal far more of interest and curiosity than we might have imagined.
The pressing need for identification arose in the West with the rise of the city, which provided anonymity that was especially dangerous in the hands of criminals. Sengoopta describes the somewhat painfully weak and expensive strategies employed by police forces simply to identify recidivists. Enter Alphonse Bertillon's strategy of biometrics, which were adopted enthusiastically. Bertillon's strategy had two benefits: it could identify repeat criminals and, equally important, it lent itself to efficient indexing and retrieval (a subtle but extraordinarily important point that Sengoopta thoroughly understands).
Bertillon's system, however, suffered from several major flaws: The difficulty of making accurate measurements with instruments of insufficient sensitivity and officers with limited training. The changing of these measurements over time. The duration it took to make accurate measurements (up to an hour, apparently). Who knew whether all the variables were even independent? Would the same metrics work for all races? What to do with children? And very few crime scenes contain many (or any) impressions that can lead to tracing by Bertillon's scheme. So the police were ready for something at least complementary.
Meanwhile, the British had, they imagined, a different problem in India. In the demonology of the Raj, the Natives were easy with the truth, prone to offering whatever testimony would please their superiors, and physically indistinguishable. Many pensions were, the administrators suspected, being collected for those long dead. Legal contracts were hard to enforce when illiterates could simply place a generic mark, and later repudiate the signing. What to do?
The value of fingerprints as a means of identification had been noticed several times and in several places. But it was in India that a British administrator began to systematically collect prints of just about everyone he encountered, thus assembling a huge database. His experiment had two critical features. First, he had lots of latitudinal data: numerous people of all ages, races, castes, and so on. Equally importantly, he also had lots of longitudinal data: the fingerprints of the same people recorded over and over across time, starting with his own. It was this dataset that was eventually to cement the role of fingerprinting as being not only unique but also invariant over time. The administrator's name was William James Herschel: yes, the scion of those Herschel's, told by his father to pursue a career in something other than astronomy.
This book is the story of three main people. Herschel, we have seen, gathered the data. Francis Galton studied them extensively and wrote the seminal book about it, though he was later disappointed for his real reason had been a hope that fingerprints could advance his pet theory of eugenics. And Edward Henry was credited with the elegant indexing scheme that made it possible to discard the use of Bertillonage entirely.
Several other important characters gracefully flow through, too. Henry Faulds fought a lifelong battle against Herschel for credit; Sengoopta gives this controversy just as much as space as it deserves, highlighting an important experiment Faulds conducted: to scar the finger and notice that the patterns grew right back. John Garson, the Bertillonage expert of the Metropolitan police force, fought a rearguard battle, though he later proved to be dipping his toes on both sides. And the many Indians who contributed to the process, and indeed may even have been partially responsible for Henry's discovery. (Again, Sengoopta has a wonderful touch, presenting the controversy but probing each side's weakness enough to leave us with a sense that it will never be resolved.)
Sengoopta could have written more about the meaning and notion of identity, but he chooses to forego the obvious. He does frequently raise the British value in liberty, which proves to be taken seriously enough (when applied to the white subjects) that schemes that should have had no trouble passing grind to a halt. He also takes to task the received wisdom of knowledge and discovery flowing in one direction only, from Britain to its colonies. In turn, his feel for colonial history is strong, and a lengthy digression into indigo farming in India is prefaced by begging the reader for forebearance, and justifies its inclusion in multiple ways. This is the kind of smart book that includes an appendix on the evolution of fingerprint classification, and is enlivened by several excellent illustrations, not least the cover of Galton's book Finger Prints (p. 100) that features Galton's own prints as the cover art. All books should be this educated, informative and yet such an enjoyable read.