Decomposing Bodies

Primary Contacts:
Dr. Alison Langmead
Department of History of Art & Architecture and School of Computing & Information

Dr. Josh Ellenbogen
Department of History of Art & Architecture

Project Manager, 2014-2015
Research Assistant, 2013-2014

Aisling Quigley
School of Computing & Information

University of Pittsburgh


For a project overview, please refer to this tidy write-up on the DHRX website. I became involved in this project in September of 2013.


I’ve been invested in this project from the start because I’m interested and concerned about surveillance, American penal institutions, and the way human beings are so routinely reduced to numbers. I was fortunate to become a part of this project through my work at the Visual Media Workshop (VMW) with Dr. Langmead.


PHASE 1: preparation and data seeking
September-December 2013
In the first stage of this project, Dr. Alex Oliver (1983-2017) and I endeavored to locate archival records related to the implementation of the Bertillon system of criminal identification in the United States. Alex and I emailed and telephoned various archives around the nation. In those early days, we posited that we were “primarily interested in statistical data related to the cards.” We were trying to determine the total extent of the federal holdings of Bertillon cards.

After reaching out to the Chicago Crime Commission, the National Archives at Atlanta, and the National Archives at Kansas City, among many other institutions, I received a very promising email back from the Ohio History Center (now the Ohio History Connection, or OHC). We had faced considerable obstacles: a government shut-down, the unexpected reverberations of Hurricane Sandy, the perils of the academic calendar…but had finally made significant progress, and at an archive that was a mere three-hour drive from Pittsburgh.

The OHC finding aids suggested that their archive was home to several thousand Bertillon cards with photographs, and hundreds of related ledger books (to see the relevant collections, enter the search term “bertillon” into this online catalog). These photographs and books originated at the Ohio Penitentiary (1884-1984) and the Ohio State Reformatory (1896-1990), and span decades.

December 2013
Dr. Ellenbogen, Dr. Langmead, and I made our first trip to the OHC to look at the archival holdings and meet with their manager of digital services.

February 2014
Alex, Dr. Ellenbogen, Dr. Langmead and I provided a progress report at the Visual Media Workshop Team Colloquium,”Producing Collaborative Work in the Humanities: The Case of Decomposing Bodies.”

Having made a preliminary trip to Columbus, Ohio to view the archival holdings, we now aimed to photograph some of this material for the purposes of data transcription and further analysis.

PHASE 2: data collection & transcription
Edited from the original blogpost on Constellations, originally published on 11 June 2014:

On Monday, June 9th, Alison Langmead, Alexandra Oliver, Isabelle Chartier, and I embarked on a road trip to Columbus, Ohio. Towing various digital cameras, tripods, copy stands, laptops, and coffee mugs, we headed for the Ohio History Connection.

The purpose of our two-day trip was to photograph as many Bertillon identification cards and corresponding ledger recordings as possible, while retaining a sufficiently sharp image to permit human readability for transcription purposes. We focused on records created between 1887-1919, carefully removing folders from boxes and photographing the contents of each folder. Approximately sixteen working hours, a few coffees, and a night in Columbus later, we had collectively photographed more than 2,700 cards and countless book pages.

More updates will undoubtedly follow, but for now we are processing the bounty of information gleaned over the past two days, and attempting to fulfill our current data storage needs…

September 2014
We began to transcribe the Bertillon cards using Omeka, as I describe in the following excerpt from a September 2014 Constellations blogpost:

Doctoral, masters, and undergraduate students are collaborating in the Visual Media Workshop, transcribing the ever-inconsistent Bertillon Cards as a key component of the Decomposing Bodies project.  We are utilizing the Agile Project Management model to coordinate our work and track our progress through “sprints,” or focused, highly-structured project periods. Our team is utilizing Omeka as our platform for uploading and transcribing images, as this particular tool accommodates image collections, robust metadata, tagging, and data exportation.

Although we are in the early stages of the transcription process, interesting (and baffling) trends are already emerging. Through the transcription of the first 250 cards, alone, I encountered three distinct types of cards bearing the Bertillon stamp and containing Bertillon measurements (with slight alterations evident in each card type).

We will, no doubt, encounter other types of cards as we continue this process, and there will most likely be complete anomalies that carry no apparent explanation. All of these observations inform our impression of the Bertillon system as it was employed at the Ohio State Reformatory and Ohio Penitentiary, and how the cards were used subsequent to their initial creation.

January 2015
I wrote this post (also copied below) after our December 2014 trip to Columbus:

On the eve of winter break, Alison Langmead, Josh Ellenbogen and I once again emerged from our cozy domiciles at a cold and dark hour and found our way to I-70, onward to Columbus. Columbus: the 15th largest city in the United States, the namesake of Christopher Columbus, and home to over 40,000 Bertillon identification cards.

In the high-ceilinged reading room at the OHC, we have explored and photographed thousands of inmate cards. On this most recent trip, however, we devoted our time to the documentation that occurred on the periphery of these cards: Bertillon ledgers, Warden’s reports, scrapbooks, inventories, blue prints, postcards, newspaper clippings, etc.

The Ohio Penitentiary’s robust Registries of Anthropometric Descriptions provided documentation of the first recorded fingerprint classification of Ohio Felons in June 1910, and helped us to concoct various hypotheses about when and how Bertillon measurements were taken—and when this data was transferred from the cards to the ledgers (simultaneously? retroactively? why weren’t the measurements of pardoned or transferred inmates included in the ledger, while escapees’ measurements were?). Although we didn’t answer these questions, the process of investigating them provoked thoughtful conversations…

Indeed, the trip provided more evidence of redundant or inconsistent record-keeping than anything else, but also helped contextualize the cards in a way they hadn’t been previously and certainly substantiated further research.

Transcription of the cards continues on the home front, but the related records will certainly be incorporated into our ongoing work. A new configuration of the research team will reconvene in Columbus next week, and we will undoubtedly return with new theories and questions that will contribute to this rapidly unfurling research project.

Update posted here, and described below:

On Monday, January 29th, Jen Donnelly, Alison Langmead and I braved a wintry mix of snow and slush, arriving at the OHC in Columbus by mid-morning. We came for another marathon digitization session, equipped with tripods and digital cameras, laptops, and sufficiently dexterous hands. Our task for the 36-hour visit was to photograph as many Bertillon identification cards as possible, while retaining a sufficiently high-quality image for transcription, and ensuring the safety of the already-brittle cards (some of which are almost 120-years-old).

Alison tackled boxes of cards from the Ohio State Reformatory, beginning in 1907, while Jen and I convened around a less-familiar set of Ohio Penitentiary cards dating from 1896. The Ohio State Reformatory, to clarify between the two, operated from 1896-1990, and was located in Mansfield, Ohio (about 67 miles Northeast of Columbus). The OHC has Bertillon cards from the Reformatory dating back to 1901.

The Ohio State Penitentiary was a prison operated in Columbus, Ohio between 1834 and 1984, and was infamous for its corruption and inhumanity (for example, newspaper clippings from a 1908 scrapbook allude to various kinds of torture: including instances of paddling, being “hung up,” and enduring the “water cure”- modern day waterboarding).

In the midst of all this, or at least until 1919, the Bertillon system was alive and well at the Pen. Marvin E. Fornshell’s The Historical and Illustrated Ohio Penitentiary (1903) provides insight into how the Bertillon system was implemented at the Penitentiary, although his account is extremely skewed (as demonstrated by his effusive subtitle: “How the Wonderful System Works in Picking Out Any One Particular Individual,” p. 49). The Pen adopted the Bertillon system in 1887, taking its first measurement in October of that year (so almost 15 years before the Reformatory adopted the system).

On this January trip, Jen and I lingered mostly in 1896 and 1897, making several subjective observations (as well as more concrete discoveries) along the way. For example, we noted that these early Bertillon cards featured remarkably gaunt and emaciated-looking prisoners, and these inmates tended to be older than their Reformatory counterparts. This may be attributable to the fact that when the system was implemented, some of these prisoners had already been in prison for a while, whereas the Reformatory seemed to mostly track incoming prisoners.

Our more concrete findings included locating and photographing the identification card of William Haas, the first prisoner to be executed by the electric chair at the Penitentiary. We discovered- much to our horror- that Haas, convicted of murder, was only seventeen at the time of his execution (shockingly and disturbingly, the policy on juvenile executions was only formally changed in March 2005. The United States Supreme Court ruled that “the death penalty for those who had committed their crimes at under 18 years of age was cruel and unusual punishment and hence barred by the Constitution” (DPIC)).

Mugshot 1: Fraternal Twins at the OP, 1896

We also located not just one, but TWO sets of twins in our selection of cards. Twins are particularly interesting in the context of the Bertillon system because of the famous Will West case, a mixed-identity debacle that apparently revealed cracks in Bertillon’s system. If twins had identical measurements and similar photographs, the system would obviously be rendered fairly useless, although twins by no means shared equivalent measurements. During this trip, we found one set of identical twins and one set of fraternal twins (the latter is photographed here).

Mugshot 2: Fraternal Twins at OP, 1896

It was, overall, a fruitful and thought-provoking trip, yielding digital documentation of over 3,500 cards that will now enter the transcription phase.

From February 11, 2015:

A particularly confusing component of our Decomposing Bodies research has concerned the inconsistent and seemingly illogical tangle of records that surround any particular inmate within the grim walls of the Ohio State Reformatory. In an attempt to get a grip on the record-keeping system of this institution in the late 18th and early 19th century, I took a few hours today mining our own unofficial image archive and trying to connect the various record-keeping nodes extant at the Ohio History Connection’s archive. I made a little timeline (see image attached) following, in particular, prisoner #1087, a laborer arrested for burglary and larceny, for whom we have both a Bertillon card and a photo of his corresponding page in the Bertillon Examination Record ledger. It seems that, as he was received in 1901, it is likely that the OSR had a version of this prisoner’s record in three different locations: the Bertillon Examination Record, the Ohio Reformatory Historical Conduct Record, and his Bertillon Card. Based on the information in the Examination Record, it is evident that inmate #1087 was a recidivist, so he was within the prison system until 1919 (although he was released from the OSR in 1904).

So, to review, prisoner #1087 was documented in three places: two of which included Bertillon descriptions, and the third of which (the Conduct Record) included information about the inmate’s ancestry, upbringing, and “condition on admittance,” none of which adhered to the Bertillon system of anthropometric measurement. Had this inmate stayed at the OSR until 1910, he would likely have acquired at least a fourth record: in the “Register of Identification,” which tracked the movement of inmates among institutions, but also later incorporated the categorization of prisoners by race. Finally, if this inmate had been admitted to the OSR after 1913, he would have had at least 5 extant records: in the Bertillon Examination Record ledger, the Ohio State Reformatory Historical Conduct Record ledger, the Bertillon card system, the Register of Identification, and the Bertillon Photo Book, a book comprisedof mugshots without any additional metadata.

Why was each inmate recorded in so many disparate ledgers and drawers? When were these different recordings made, in relation to one another? These are all things that we’re still trying to figure out, but I thought I’d get this initial timeline out while it’s still hot off the press.

May 2015
Update from May 18, 2015: 

In the past two years, we’ve digitized approximately 3,500 Bertillon identification cards and transcribed about 43,200 discrete data points (1,800 cards with 24 data points per card).

Preliminary analysis of a small sample of the data (cards #412-948 from the Ohio State Reformatory) already supports some of our nascent theories. For example, our initial encounters with the cards led us to believe that prisoners were not measured in numerical order, although the cards are organized thusly. For example, Prisoner #412, the earliest prisoner documented in the cards at the OHS Archives, was measured on January 18, 1902. The first prisoner with recorded Bertillon measurements is actually prisoner #738 (measured on September 14, 1901). Why would this be the case? Were the Bertillon Officers measuring the long-term inmates inconsistently, on a case-by-case basis, while the incoming prisoners were measured in a more predictable manner?

Please refer to the drawing of the “average inmate” attached to this post to see some average measurements from this cohort. As you can see, the average height is around 5 ft 6 inches or 169.4 cm. Although this seems somewhat short compared to today’s averages, it actually adhered to height averages reported in men born in the 1880s (which was around 169.5 cm).