Despite contemporary beliefs that tattoos were unrespectable our evidence shows that tattoos designs evoked a range of positive emotions, particularly love. – Candace Sutton, Secret Codes & Hidden Symbols of Australian Convict Tattoos
Never have I ever had a tattoo or considered data and text mining as a part of digital humanities. Now consider my surprise to see the combination of the two with the database of over 60,000 Australian and British criminals from 1788-1925 with research by Zoe Alker. Her use of data mining measured the cultural trends of the time and offered another perspective on the meaning of tattoos for a person of the time. Alker wrote, “Here in this research I shifted attention from branding as a tool for state control and official surveillance to examine the convict body as a site upon which gender, ethnicity and class were symbolically marked by the convicts themselves.and offered a perspective of how working class people embraced the medium giving tattoos respectability in popular culture.” The use of visualizations and data mining demonstrated chronological patterns, explain differences in tattooing practices depending on convicts’ gender, age, place of origin, occupation, religion, offending and punishment history, and types of sentiments expressed, including love, faith, identity, and personal history. From Alker’s example and a better understanding of how data/text mining, students like us can implement these practices on our quest to become productive members of digital humanities.
Histories of the tattoos worn by transported convicts have shown the importance of ‘giving access to the convict voice’ while also noting the rich variety in sentiments expressed through a wide range of designs (Kent 1997; Bradley and Maxwell-Stewart, 1997; Maxwell-Stewart and Bradley, 1998, Rogers, 2004).
What struck me about Alker’s presentation was her declaration on how tattoos are a form of expression by whom no other records are available to recognize that group. For data mining this offered optical character recognition. With her very specific topic she bypassed a warning about data/text mining mentioned in Carly Minsky’s article, How AI Helps Historians Solve Ancient Puzzles, “Even with high accuracy, letters and words are frequently misidentified or missing entirely. At a more fundamental level, historical data may mislead historians when algorithms are used to select what gets archived and what gets deleted.” In history entire groups of people are missing and Dr. Alker’s research offers a record of Victorian London from the perspective of those considered working class and below and the database with the tattooed convicts filled in a historical gap. Granted, there was a question of accurate record keeping and that is whether the convicts had all their tattoos accounted for on every intake with a fully naked examination every time. With over 75,688 descriptions of tattoos recorded, I believe the early twentieth century British judicial authorities demonstrated their commitment to detail. The system of record-keeping was developed in response to fears over re-offending and the apparent existence of a clearly delineated criminal class (Shoemaker and Ward, 2017). Dr. Alker’s data mining of convict tattoos, physical characteristics (eye colour, weight, etc.), bodily infirmities (scars, etc.), and personal details (age, gender, ethnicity, etc.) provided information to create graphs on occupational class, body placement, and the very interesting Changing Mix of Tattoo Subjects. For example, in the 1900s the most popular tattoos documented were justicepunishment and sex. In Dr. Alker’s conclusion she wrote, The diverse range of designs and subjects suggests that tattoos were very much embedded in the wider culture of England in the ‘long’ nineteenth century, while the demographic spread of the tattooed and the public location of most tattoos suggests that there was not much stigma associated with having a tattoo.
Long-term goal for data/text mining and academic life in general: Make research data findable so that it can continue to have impact, even after you have finished with it. – Examples of Text Mining & Text Analysis, University of Queensland.
The Dataset for the Convict Tattoos involved more than research, photo scanning and countless hours of processing the information but also creating a user-friendly database. Data and text mining often involves working with and storing large data sets. It should be a consideration when starting research to have secure storage. (Examples of Text Mining & Text Analysis) Also be considerate when accessing research data made available by other organizations since it is important that your mining activities do not inadvertently disclose confidential information or breach the privacy of research subjects. It was interesting to discover Australia’s Copyright Act of 1968 makes no specific exemption for text or data mining while American copyright law views it a culture clash between publishers and technologists on individualism vs. data is meant to be extracted for meaning and information. (Copyright and the Progress of Science: Why Text and Data Mining is Lawful, Michael Carroll.) Never have I ever seen topics as debated as Digital Humanities and why face tattoos work for Post Malone.
A “Convict Connection” for the San Juan Island Pig War fans: One drawing in Barnard’s book depicts 31-year-old convict Isaac Comer undergoing inspection at Hobart Town on arrival in 1845. “Comer, who had a record in his home town of Bath for stealing a pig and assault, had been transported for 14 years for receiving stolen goods.” Leave the pigs alone people!