This project has been a success in its display of how considerations for digital humanities projects should play out in the practical. The process of collecting and digitizing data is just the first step in then preparing data to be fed into sophisticated systems such as Tableau Public, Gephi, Flourish, and others. Finally, the results of those analyses will likely produce results that furnish additional questions.
Questions of data validity have plagued this project ranging from the acknowledgment of missing data due to the lack of physical records, or silences in the archives and data due to the illegal slave trade that will make it impossible to ever know how many enslaved persons were bought and sold in the United States.
While some of these issues, such as the reality of missing records, regardless of the reason, can be recognized in a project, some issues with this dataset are harder to overlook. For example, the crowdsourced way in which this data became digitized has left noticeable errors in it, whether that be duplicate records, departure dates coded as arrival dates, or the same person being classified as two people due to inconsistent naming conventions, all call into question whether the data being analyzed is accurate in the first place.
While I am confident in the acknowledgements of silences in my report and the methodology I have used to analyze the data and uncover these errors, I cannot help but consider that the data itself is incorrect and therefore any academic conclusions about the New Orleans Slave Ship Manifests that come from this data would be at best incomplete and at worst, incorrect.
If I were to continue this project beyond the scope of this course and include it in academic work, I would be compelled to abandon my current data set and transcribe the records directly from the manifest microfilm myself. The methodology I learned and used here would remain the same, but the data itself would need to be more accurate.
This data and methodology demonstrate opportunities for future historians and digital humanists to discover the lives of enslaved persons, having been named in these manifests. Applying Saidiya Hartman's "critical fabulation" framework, it is possible to reconstruct the lives of the enslaved as part intentional fiction and part archival research. (Venus in Two Acts)
A larger project that explores more than four years of data may also be able to understand the frequency with which ships do or do not return to the port, and even further research on individual vessels may help understand if those same vessels return to the Port of New Orleans for other reasons and therefore appear in port more often than to deliver enslaved persons.
A larger project should also include departure data to determine the number of enslaved persons entering the deep south from other parts of the United States, which would help to understand how enslaved labor was moving out of the northern southern colonies and into the deep south following the bust of the tobacco market as the cash crop reliant on enslaved labor.
Ideally, I would like to produce a network analysis of some kind that connects ships and shippers to see if shippers hired the same ships over time or with any consistency. I tried loading a version of my spreadsheet into Gephi that contains just the ship and shipper's name per column but continued to get errors. At this time I am attempting an analysis using the online tool Flourish, which appears to be building those connections, though not in the easiest to view and digest way.
NOLA Slave Manifests Data 1818-1821, Ships and Shippers via Flourish
NOLA Slave Manifests Data 1818-1821, Ships and Shippers via Flourish
Similar analysis could also be completed in larger project by using departure ports as an additional metric.
Next Page:
Back to Home:
CDH 501 Project - Inbound Slave Manifests of New Orleans, 1818-1821