Search for: All records

Many researchers insist that computational methods will transform the historical profession, while an equally large number reject these claims as unwarranted hype. In this study, we attempt to place the debate in historical and social context. The article is divided into three parts. The first part offers a brief review of computational history. We ask whether the “computational turn” bears any resemblance to quantitative history, and how it fits within the digital humanities. The second part describes our recent attempts to apply computational methods to an existing project in the history of science using a standardized workflow. We demonstrate that each step of the workflow adds another layer of subjectivity. The third part reflects on what computational methods mean for the historical profession. We systematically review the positive aspects of computational history (open access, interdisciplinary collaboration, and new perspectives) as well as the negative aspects (inequality, fragility, and the threat of automation), and we offer prescriptions based on our experiences. 

Every project in digital and computational history of science starts with the collection of data. Depending on the research project, subject of study, and other factors, data can be comprised of a variety of different types such as full-texts, images, audio, video, or bibliographic metadata. Often publications and project reports describe a project’s results and the employed methods and algorithms, but few publications discuss the challenges of the initial data collection process or how it fits into the overall research data life cycle. In this paper, we discuss a concrete research data life cycle and take a look at the difficulties it involves. Furthermore, we explore the strategies and challenges of data collection, and the question of comparability of datasets. 

Digital technologies have transformed both the historical record and the historical profession. This Focus section examines how computational methods have influenced, and will influence, the history of science. The essays discuss the new types of questions and narratives that computational methods enable, and the need for better data management in the HPS community. They showcase various methodological approaches, including textual and network analyses, and they place the computational turn in historiographical and societal context. Rather than surrender to technophilia or technophobia, the essays articulate both the benefits and the drawbacks of computational HPS. They agree that the future of the field depends on the successful integration of technological developments, social practices, and infrastructural support, and that historians of science must learn to embrace collaboration both within and beyond disciplinary boundaries. 

This essay describes essential considerations and select methods in computational text analysis for use in the study of history. We explore specific approaches that can be used for understanding conceptual change over time in a large corpus of documents. By way of example, using a corpus of 27,977 articles collected on the microbiome, this paper studies: 1) the general microbiome discourse for 2001 to 2010; 2) the usage and sense of the word “human” from 2001 to 2010; and 3) highlights shifts in the microbiome discourse from 2001 to 2010.