Next week I will be visiting Rome to join the Associazione per l’Informatica Umanistica e le Culture Digitali (AIUCD) conference which will be held from 26-28 January at Sapienza University. See the entire programme here. The topic of the conference is “Il telescopio inverso: big data e distant reading nelle discipline umanistiche”, and as a result Mark Hill and I have formed a panel on big data, distant reading, concept drift, and digital history. In this blogpost I’ll post the abstract of the panel, and my own abstract; if the full proceedings including the abstracts of the other panel members are online I’ll add it to the presentations page. We are excited to have brought together scholars working on concept detection, ambiguity, and methodology of history, so we hope we will get a very nice discussion going.

Panel: Reflections on reading history from a distance

Max Kemman* (University of Luxembourg), Mark Hill* (London School of Economics), John Regan*, Paul Nulty*, Peter de Bolla (University of Cambridge), Pim Huijnen* (Utrecht University), Tom Kenter (University of Amsterdam), Daniele Guido* (University of Luxembourg). *Presenting authors

This panel aims to discuss how digital tools fit in to historical practices, and reflect on the interaction between digital and historical methods. While methods of computational textual analysis have been used in many other disciplines,^[1]Laver, Michael, Kenneth Beniot, and John Garry. 2003. “Extracting Policy Positions from Political Texts Using Words as Data.” American Political Science Review 97 (02). Cambridge University Press (CUP). doi:10.1017/s0003055403000698.^[2]Grimmer, J., and B. M. Stewart. 2013. “Text as Data: The Promise and Pitfalls of Automatic Content Analysis Methods for Political Texts.” Political Analysis 21 (3). Cambridge University Press (CUP): 267–97. doi:10.1093/pan/mps028.^[3]Schonhardt-Bailey, Cheryl. 2006. From the Corn Laws to Free Trade: Interests, Ideas, and Institutions in Historical Perspective. Mit Press.^[4]Moretti, Franco. 2013. Distant Reading. Verso Books. historians (in particular, intellectual historians) have been suspicious of ‘distant reading.’ As Skinner argued, historical texts are not positive facts from which we can reconstruct empirical meaning.^[5]Skinner, Quentin. 2002. Visions of Politics. Cambridge University Press (CUP). doi:10.1017/cbo9780511613784. Nonetheless, there is an interest in these ideas and their application to historical studies. Thus, as a case study of the interaction between digital and historical methods, this panel will focus on how tools, and specifically distant reading techniques, may be used to investigate “concepts” in historical documents.

This panel follows recent investigations by the likes of Guldi and Armitage, who argue in The History Manifesto that, with the emergence of new digital materials, methods, and techniques, historians may be able to return to longue durée historical investigations.^[6]Guldi, Jo, and David Armitage. 2014. The History Manifesto. Cambridge University Press (CUP). doi:10.1017/9781139923880. However, how this is exactly to be done is not a triviality, and while most reflections have foregrounded the technical at the expense of the methodological, this panel hopes to investigate this relationship in more detail. That is, there is a shared concern that some digital investigations may introduce unexpected methodological problems and thus may ultimately have an impact on the accuracy of analysis and claimed conclusion. In essence, distant reading depends on counting occurrences of terms, but what these terms mean conceptually may change over time and context. When treating historical sources as big data and interpreting the outcome of distant reading methods, historians risk remaining ignorant to such concept drift. These issues have not been ignored,^[7]de Bolla, Peter. 2013. The Architecture of Concepts. Fordham University Press. doi:10.5422/fordham/9780823254385.001.0001.^[8]Edelstein, Dan. 2015. “Intellectual History And Digital Humanities.” Modern Intellectual History 13 (01). Cambridge University Press (CUP): 237–46. doi:10.1017/s1479244314000833.^[9]London, Jennifer A. 2016. “Re-Imagining the Cambridge School in the Age of Digital Humanities.” Annual Review of Political Science 19 (1). Annual Reviews: 351–73. doi:10.1146/annurev-polisci-061513-115924.^[10]Wang, Shenghui, Stefan Schlobach, and Michel Klein. 2011. “Concept Drift and How to Identify It.” Web Semantics: Science, Services and Agents on the World Wide Web 9 (3): 247–65. doi:10.1016/j.websem.2011.05.003. but there remains a need for further serious methodological reflection, in addition to technological solutions, if we are to practice this sort of digital history.

To address these issues this panel brings together a number of researchers who are engaging with these problems from differing perspectives. This includes researchers who are at the cutting edge of the use of quantitative text analysis on historical documents, with John Regan and Peter de Bolla aiming to show how historically-sensitive distributional text analysis can facilitate the recognition of conceptual architectures, and Pim Huijnen and Tom Kenter focusing on continuities and changes in conceptual structures over time. Mark Hill looks at ways one may sidestep the traditional methodological problems quantitative analysis may be introducing by turning to niche historical investigations. Daniele Guido will reflect on his experience with both developing tools to be used by historians, and the potential pitfalls which may lurk. Finally, Max Kemman will reflect on how historians collaborate in digital history projects. The goal, then, is to bring together these practitioners, contrast their differing technological and historiographical approaches, and ultimately offer some further reflections on how digital history facilitates an interaction between digital technology and historical practices.

Paper: Digital History Projects as Boundary Objects

Digital history as a subfield of the digital humanities constitutes a form of methodological interdisciplinarity; using methods, concepts, or tools from other disciplines to try to improve historical research.^[11]Klein, Julie Thompson. 2014. Interdisciplining Digital Humanities: Boundary Work in an Emerging Field. Online. University of Michigan Press. doi:10.3998/dh.12869322.0001.001.. However, as this panel demonstrates, this is not a straightforward process of taking something from another discipline and implementing it in historical research. Instead, what we see is a negotiation of practices to align the new methods with the scholarly values of the discipline.^[12]Kaltenbrunner, Wolfgang. 2015. “Reflexive Inertia: Reinventing Scholarship through Digital Practices.” PhD thesis, Leiden University. This negotiation regularly takes place in the context of a research project, where participants with different backgrounds work together on a shared problem. Yet despite working on a shared problem, the individual participants may still have different research goals and incentives to enter the collaboration. Although the research project defines a common research problem, how this research problem is or should be approached differs between the different collaborators dependent of, among other factors, their disciplinary background. This paper will therefore analyze the research project as boundary object, i.e., as an object that maintains a common identity among the different participants, yet is shaped individually according to disciplinary needs.^[13]Star, S.L., and J. R. Griesemer. 1989. “Institutional Ecology, ‘Translations’ and Boundary Objects: Amateurs and Professionals in Berkeley’s Museum of Vertebrate Zoology, 1907-39.” Social Studies of Science 19 (3): 387–420. doi:10.1177/030631289019003001.^[14]Star, S.L. 2010. “This Is Not a Boundary Object: Reflections on the Origin of a Concept.” Science, Technology & Human Values 35 (5): 601–17. doi:10.1177/0162243910377624. In order to investigate how participants shape the research project and align the project with their scholarly values, we will look at the individual incentives for collaboration, following research by Weedman on incentives for collaborations between earth scientists and computer scientists.^[15]Weedman, Judith. 1998. “The Structure of Incentive: Design and Client Roles in Application-Oriented Research.” Science, Technology & Human Values 23 (3): 315–45. doi:10.1177/016224399802300303. For several digital history projects, we will discuss collaborators’ reasons for joining the project, their individual goals with the project, and the expected effects of the participation after the project has ended.

This research is part of a PhD research on how the interdisciplinary interactions in digital history have methodological and epistemological consequences for the practice of historians.^[16]Kemman, Max. 2016. “Dimensions of Digital History Collaborations.” In DHBenelux. Belval, Luxembourg. By untangling the individual interests in digital history projects, we aim to gain better insight into how digital history functions as a coordination of practices between historians and collaborators from other disciplinary backgrounds.

Slides

You can find the slides for my presentation on SlideShare here.