The Japan News by The Yomiuri Shimbun

Home > Opinion > Culture and Education


Culture and Education

From Classifying to Searching—A New System of Knowledge Starting from “Fingerprints”?

Kazumichi Hashimoto
Associate Professor, Faculty of Letters, Arts and Sciences, Waseda University

How long has it been since “searching” for information has become an essential academic tool? In addition, of course, to searching for desired books or papers in library databases, if the entire contents of academic papers or literary works have been put into digital form, searches can allow one to find even specific keywords or phrases within their text. There are more and more cases of people conducting research using “search” investigations to find their way to descriptive information that would otherwise be near-impossible to locate, and one could argue that research methods themselves are on their way to becoming something fundamentally different than they once were.

This topic is not limited only to the academic world. When one encounters an unfamiliar word, the act of first typing it into a browser search window has become firmly established as a behavioral pattern of many people, including myself. In another example, there are numerous cases of comments posted to Twitter or other SNS, originally intended only to be shared between friends, that are found through searches and appended to repeatedly by third parties, ultimately resulting in incidents of “flaming”. As though out of fear of such occurrences, “search evasion” methods using abbreviations or code words are becoming general practice, with “searching” becoming deeply rooted for better or for worse in our daily lives, and affecting our behavior to varying degrees.

Before “searching” held the position of prominence it has today, “classification” was the method used by those looking for desired items within huge quantities of materials or information. The classification system used by libraries is almost archetypal, categorizing books into layered subdivisions based on their genres and other criteria, with users able to find specific books on shelves or in catalogs according to those classifications. Of course, “classifying” and “searching” do not necessarily conflict with each other, and “classifying” can actually make “searching” easier by placing initial limits on the quantity of core content to be the target of a search. In recent years, advances in information processing technology have already made it possible for “searching” to be done without a need for the “classification” process. Although I apologize for giving a personal example, at some point in the past, I decided to stop classifying data into various folders when saving it on my own computer. By instead naming files with easily-searchable names and then placing them all in a single folder, I am later able to just search within that folder and have no need for more detailed classifications.

“Classification” in the past was not only confined to organization of books and other materials. Just as was the case with the Linnaean biological classification system, classification serves as a way for mankind to divide the world into parts and obtain clues to its understanding. It would not even be an exaggeration to say that it is a representation of human thought itself. If such “classification” were to be surpassed instead by “searching”, it could be considered a momentous turning point in the history of human consciousness. The beginning of such a major turning point can be found in none other than the emergence of identity confirmation by fingerprints.

The method of fingerprinting was first introduced as a way for the police to organize the enormous quantities of criminal records in its possession. Before fingerprints were used, this function was performed by an anthropometric method invented by Alphonse Bertillon. With this method, the sizes of a criminal’s hands, feet, head, ears, etc. were measured and the results were separated into “large”, “medium” and “small” divisions. For example, a category reading “hands: large, feet: large, head: medium” would be used to classify the records of people fitting those descriptions. It was a mechanism that allowed an applicable record to be found by repeating those measurements, if a person was arrested again. At the time this method started to be employed in individual countries, the method of fingerprinting was already known, but the anthropometric method was preferred simply because its classifications were simpler. In the early stages of fingerprinting, classifications were made using independent criteria, and identification by searching without classification was made possible only after the establishment of computerized search systems. Still, the overtaking of the anthropometric method, which was excellent as an example of a classification system, by the fingerprinting method in the early 20th century, can be seen as an event symbolic of the shift from “classifying” to “searching”.

When I published my book Fingerprinting Theory (Seidosha) in 2010, I intended to cover to a certain extent the problems faced by fingerprinting as a method of identity confirmation. After its publication, however, as I come to recognize the deep connection between the problems of fingerprinting and “searching”, I am feeling more and more strongly that there is a need to return once more to fingerprinting and trace back the history of thought related to “searching” that began with it. But is “searching” actually something that can give rise to some form of thought the way “classification” can? For example, if books or papers screened out and selected from a database by a search have no more than a few keywords in common and are otherwise completely unrelated, it seems that nothing could be found but disorder and chaos.

The information presented by Gaston Bachelard in his Essay on Approximate Knowledge [Kinjiteki Ninshiki Shiron] (Kokubunsha, 1982) is intriguing with regards to this point. In this book, he mentions that apart from the truth that can be arrived at by systemization through classification, another truth exists as well (pg. 309), and he suggests that this other truth is linked to the “details”. As is often the case with fingerprinting, this is because searches in many cases are matters closely concerned with finding out details. It is obviously pointless, though, to expect the discussions of this book, which was initially published in France in 1927, to encompass topics such as the appearance of the Internet and the growing range of “searching” within their scope. Is “searching” able to awaken any and all knowledge in mankind? The preparations needed to answer this question have only just begun.

Kazumichi Hashimoto
Associate Professor, Faculty of Letters, Arts and Sciences, Waseda University

Born 1974. Graduated from Department of Philosophy and Religion, Faculty of Letters, The University of Tokyo. Completed DEA program at Faculty of Science and Technology, University of Nantes, France. Completed doctoral program at Graduate School of Arts and Sciences, The University of Tokyo. After obtaining position as lecturer at Aichi University of Technology, has held current position from 2012. Area of specialization is Culture and Representation. Literary works include: Fingerprinting Theory – From Spiritualism to Biometrics [Shimonron – Shinrei Shugi kara Seitai Ninshou made] (Seidosha, 2010). Translated works include: Didi-Huberman, Georges, Images malgré tout [Imeeji, sore demo nao] (Heibonsha, 2006); Vigarello, Georges, editor, Histoire du corps, Tome 1 [Shintai no Rekishi 1] (co-translator, Fujiwara-Shoten, 2010); and Legendre, Pierre, La balafre [Douitsusei no Nazo – Shiru koto to Shutai no Yami] (Ibunsha, 2012).