The Act of Identification: Bertillon and Chinese Exclusion


The Act of Identification: Bertillon and Chinese Exclusion

Last week’s break-through led us to begin researching the use of the Bertillon system to enforce the Chinese Exclusion Act.  The Chinese Exclusion Act prohibited the immigration of Chinese laborers from 1882 to 1943.  Whereas prisons and police departments used the Bertillon system to identify criminals, the United States government used the Bertillon system from approximately 1903 to 1906 to identify those Chinese immigrants who were allowed in the United States because they met certain sets of requirements, while excluding all others.

The use of the Bertillon system of identification to enforce the Chinese Exclusion Act was short-lived, but was the result of twenty years of increasingly stricter immigration polices. After 1882, the any Chinese laborer who was already in the United States was banned from obtaining citizenship and needed to obtain a certificate to leave and re-enter the country.  Furthermore, Chinese members of the groups still permitted entry, such as students and wealthy travelers, were required to obtain certificates verifying their status and their access to enter the United States.  After 1902, the regulations were stiffened, and all Chinese residents in the United States were required to obtain a certificate of residency or risk deportation.  Shortly thereafter, around 1903, the Bertillon system was briefly implemented.  These regulations also burgeoned an extensive human smuggling and document forgery industry.

Therefore, for potential immigrants, having, creating, or purchasing the "correct" identity was key; identification was the avenue to immigration.  Paperwork was the basis of entry and exclusion.  As the certificates of residency attest, "laborer" did not refer to an occupation but rather was a statement of fact that enabled thinly veiled racial exclusion.  In one such certificate from 1900, a baby-laborer's occupation is listed simply as "infant."  Thus, the language of exclusion was bound-up in the identification requirements, which evolved as the twentieth century approached.

In 1882, when the law was first enacted, it was required that the Collector of Customs board all departing vessels carrying Chinese laborers to foreign ports, and "on such vessel make a list of all such Chinese laborers, which shall by entered into the registry books to be kept for that purpose, in which shall be stated the name, age, occupation, last place of residence, physical marks or peculiarities, and all facts necessary for the identification of each such Chinese laborers, which books shall be safely kept in the customs house."  Likewise, every every Chinese person who was not a laborer and who was therefore allowed to enter the United States need to be "identified" by the Chinese government in an official certificate, translated into English.  The certificate stated "the right to come, and which certificate shall state the name, title, official rank if any, the age, height, and all physical peculiarities, former and present occupation or profession, and place of residence in China" and required the applicant’s “proper signature.”  In 1884, “individual, family, and tribal name in full” were added to both methods of identification.  By 1888, the identification certificates for the eligible classes, “Chinese officials, teachers, students, merchants or travelers for pleasure or curiosity” were required to include “a full description of the person, of his age, height, and general physical features.” 

Notably, between 1884 and 1888, fifteen years before the implementation of the Bertillon system for Chinese Exclusion, identification requirements changed from “peculiarities” to “general features.”  What does this shift mean for the nature of identification? The initial focus on “peculiarities” marks the person by his or her difference.  In this case, that difference is what permits the person to be identified as one with special rights and priveledges.  The later turn towards “general features” suggests that instead of individual marks of uniqueness, the entire body much be subjected to the systematization of the identification process.  Eventually, this will lead to the implementation of the Bertillon system.

Furthermore, the shared use of the Bertillon system on criminals as well as immigrants created problematic parallels.  Those who received identification cards using the Bertillon system were permitted into American society, while the prisoner identification cards were meant to keep repeat offenders out.  In the case the prisoners, the act of measuring renders the prisoner into a permanent piece of metadata.  The goal, in an extreme interpretation, is to identify the person as a criminal and remove that criminal from society, made invisible behind a prison cell.  For the immigrants, the metadata is instead a tenuous guarantee to remain visible in the United States, yet also a reminder of the invisibility of the uncountable, unwanted potential Chinese immigrants banned from admission to the country.

Thus, the act of subjugation to the system and judgment under the schematization of numbers was a humiliation.  "Lan Qiche…noted how his country people were measured 'as if they were criminals.' Like many others he found this 'an insult to our nation's dignity.'" (Parenti, The Soft Cage).  The question here lies in the nature of the insult of the Bertillon system.  What was commonly known of the Bertillon system and how was it received?  The shame of the Bertillon system was folded into a deeper institutional insult, revealed in language of exclusion of the law and its appeals.  "The exclusion of paupers, criminals, and persons afflicted with incurable diseases, for which statutes have been passed, is only an application of the same power to particular classes of persons, whose presence is deemed injurious or a source of danger to the country.  As applied to them, there has never been any question as to the power to exclude them.  The power is constantly exercised; its existence is in solved in the right of self-preservation…"

  • Agency
  • Identity
  • Decomposing Bodies
  • Graduate Work
  • VMW


Jen, My guess is that the

Jen, My guess is that the military pioneered general identification systems.  In enlistment records in the Civil War, the cards recorded three or four types of bodily data: height, eye color, hair color, and complexion.  The problem was that so many men shared the same name.  But if you had a name and four pieces of metadata, chances are you had enough data to narrow the field to a single individual.

There's a historian at Cornell working on a DH project to digitize all the slave runaway notices ever published in the U.S.  These runaway ads sometimes included general characteristics but more often (I think) tried to identify peculiarities (often particular scars or brands).

In both these cases photography was not part of the system at all.

And, interestingly enough, as

And, interestingly enough, as our colleague Josh Ellenbogen has demonstrated...Bertillon himself didn't even include the photograph as "proof." The photography used in identification systems not only didn't revolutionize the concept of "identity," it even put more doubt into the system in some circles. What you see is not necessarily what you get!

I'd like to see Josh himself

I'd like to see Josh himself weigh in on this question of personal identity.  Did Bertillon claim that no two people would have the same set of measurements, or that the possibility was so unlikely that it wasn't worth worrying about?  If there was a real statistical chance that two different people could clock in with the same measurements, then you would need the photograph (or some other set of physical markers) to make the determination.  But if the system of measurement was really robust enough on its own to establish unique identities, then what was the role of the photograph in the system?  Why not just record a name instead?  What "knowledge" did the photograph offer if any?

In Bertillonnage, the

In Bertillonnage, the anthropometrical measurements primarily served to narrow the field of possible individuals. With a set of eleven anthropometrical measurements, even if you had an archive of hundreds of thousands of records, Bertillon held that you could eliminate so many possible matches that you would finally end up looking at only a dozen or so candidates. Here, the photograph played the role of the final identity test. Sometimes, however, Bertillon gets so taken with anthropometry that he claims the measurements alone will suffice to make a match, and that the photograph serves only to confirm the match you've already made (I think he gets carried away like this for the simple fact that the measurements are numerical, and therefore more rigorously scientific and definite than the qualitative image). On the other hand, Bertillon knows full well that the measurements are only accurate within particular margins of error, and so it is not actually possible to just make the match based on numbers. In actual practice then, the photograph does play a meaningful role in making identifications. This leaves aside the problem of what to do with measurements that are so close to the boundaries between two groupings (say "average height," and "above average height") that the margin of error sees to it that an individual could go into either case. Say a guy is just barely "above average" in height. If that guy gets measured the first time as "average," which is quite possible within the margin of error, and then the next time gets measured as "above average," that second search will ultimately end up taking you to a totally different part of the archive from where his record is stored. If you get to that part of the archive and don't find him, you're supposed to go back and redo the search, this time treating his height as "average" (you can imagine how rough this becomes when you've got a guy with two or three or four separate measurements, each one of which happens to be on a border). Well, what's the standard that allows you to say, when you get to a certain drawer of the archive and don't find the guy (and thus have to go back and redo the search) that "hey, I am not finding this guy"? It has to be the photograph. So, the photograph always play a role, although the lion's share of the work is done by anthropometry.  

Thanks Josh.  Yes, I have had

Thanks Josh.  Yes, I have had the experience of going to the right drawer and not finding the person there many times in many contexts!

I'm still pondering the ramifications of the photograph as the final arbiter of identity in this system.  The system seems despite itself to privilege photography as the most reliable "match" for the person, yet at the same time to throw the status of photography into doubt.  One could imagine cases where the measurements of the person lined up pretty well with the measurements on the card but the photograph itself was ambiguous, not clearly unlike enough to disqualify the identification or like enough to make a definitive match.  Presumably these were rare, or rare enough that the combination of visual and numerical data seemed to guarantee identification in all but the most exceptional cases?