Guldi, Jo. 2019. "Parliament's Debates About Infrastructure: An Exercise in Using Dynamic Topic Models to Sythesize Historical Change." Edited by Suzanne Moon. Technology and Cuture: The International Quarterly of the Society for the History of Technology (Johns Hokins University Press) 60 (1): 1-33.
Reviewed by Michael Beach
This article looks at data generated through text mining of debates held in British Parliament regarding social infrastructure from 1800 through 1890 with the specific intent to “test the usefulness of text-mining methods” (p.3) for a specific microhistory and for synthesized macrohistorical trends. The study links infrastructure words with specific project types, and also shows how evolving infrastructure terms show government priority shifts over time. Guldi intends to show how the data can “reveal new tensions and turning points that characterized the uptake of infrastructure” (Ibid.) over time.
Historians have long debated the usefulness of looking at a microhistory and extrapolating trends in macrohistory, or how such trends can be read into individual microhistorical events. One example shared based on the word association data was the amount of infrastructure debate investment Parliament put toward piers and dredging in the River Shannon. As expected there was more attention to major needs on River Thames, but what is surprising was how such a lesser-known body of water like River Shannon attracted so much attention. For example it is mentioned considerably more than work destined for the Nile, Clyde, Trent and Severn rivers, all of which the author notes as having more significant economic value.
The data tables offered are persuasive in terms of how computer-based datamining can quickly find these historical shifts not easily captured otherwise. Guldi ultimately concludes that micro- and macrohistories complement each other. This larger conclusion seems more opinion than can be synthesized from the arguments made through the word-association data. Historians of technology are the specific intended audience, but historians in general may find such a computer-based tool helpful in other areas of research.
Reviewed by Michael Beach
This article looks at data generated through text mining of debates held in British Parliament regarding social infrastructure from 1800 through 1890 with the specific intent to “test the usefulness of text-mining methods” (p.3) for a specific microhistory and for synthesized macrohistorical trends. The study links infrastructure words with specific project types, and also shows how evolving infrastructure terms show government priority shifts over time. Guldi intends to show how the data can “reveal new tensions and turning points that characterized the uptake of infrastructure” (Ibid.) over time.
Historians have long debated the usefulness of looking at a microhistory and extrapolating trends in macrohistory, or how such trends can be read into individual microhistorical events. One example shared based on the word association data was the amount of infrastructure debate investment Parliament put toward piers and dredging in the River Shannon. As expected there was more attention to major needs on River Thames, but what is surprising was how such a lesser-known body of water like River Shannon attracted so much attention. For example it is mentioned considerably more than work destined for the Nile, Clyde, Trent and Severn rivers, all of which the author notes as having more significant economic value.
The data tables offered are persuasive in terms of how computer-based datamining can quickly find these historical shifts not easily captured otherwise. Guldi ultimately concludes that micro- and macrohistories complement each other. This larger conclusion seems more opinion than can be synthesized from the arguments made through the word-association data. Historians of technology are the specific intended audience, but historians in general may find such a computer-based tool helpful in other areas of research.
| project_muse_719944.pdf |
RSS Feed