The Concept Lab studies the architectures of conceptual forms. It is committed to the view that concepts are not equivalent to the meanings of the words which express them. The Lab considers conceptual architectures as generating structured environments for sensing that one has arrived at understanding. Through computational methodologies it seeks to describe and analyse these structured environments which have both internal and external manifestations. With respect to the internal structure of concepts we aim to characterise not only form and function but also modalities and other attributes of conceptual entities. In the case of external features we seek to uncover the conceptual networks within which concepts circulate. In both cases the Lab develops data driven accounts of these structures and networks derived from large scale corpora.
The Concept Lab aims to generate a substantial repository of data that will allow us to understand better how concepts work. Although our primary focus is on the cultural expression and manifestation of conceptual forms, deriving our data from the written record of linguistic usage, we intend our work to contribute to the understanding of concepts in the widest sense. Thus, although the Lab does not engage in experimental methodologies for exposing cognitive behaviours and does not, correspondingly, seek to explain how human agents possess concepts it does regard its work as complementary to such studies.
A second strand of work concerns the historical extension of conceptual forms. This aspect of our project seeks to identify where, how and why concepts change or emerge over time. To this end it is engaged in the production of a number of case studies that explore both the computational methodologies that are continuously revised by the Lab and the histories of words, concepts and usages that heretofore have provided the common currency of the history of ideas.
The Lab has a North American counterpart working in tandem with the Cambridge group. The four researchers in Cambridge – Peter de Bolla, Paul Nulty, Gabriel Recchia and John Regan – regularly meet remotely with the four researchers in North America – Clifford Siskin, Mark Algee-Hewitt, Bill Blake and Yohei Igarashi. The entire group has a yearly workshop that meets either in Cambridge or North America.
The work of the Lab is intended to contribute to two larger projects. One, the Cambridge Centre for Digital Knowledge based in CRASSH, seeks to further our understanding of how new knowledge is produced in the digital domain. The second, the Re: Enlightement Project, seeks to promote innovative means for creating new knowledge.
The Concept Lab Cambridge
Professor Peter de Bolla (English, University of Cambridge)
The Concept Lab North America
Clifford Siskin (English, NYU)
Publications and Working Papers
Nulty, Paul. (2017). “Network Visualisations for Exploring Political Concepts”. Proceedings of the 12th International Conference on Computational Semantics (IWCS).
Recchia, G. & Nulty, P. Improving a fundamental measure of lexical association. To appear in Proceedings of the 39th Annual Conference of the Cognitive Science Society.
Recchia, G., Jones, E., Nulty, P., Regan, J., & de Bolla, P. (2016). Tracing shifting conceptual vocabularies through time. In Ciancarini, P. et al. (Eds.): Knowledge Engineering and Knowledge Management: EKAW 2016 Satellite Events, EKM and Drift-an-LOD, Bologna, Italy, November 19–23, 2016, Revised Selected Papers (pp. 19-28). Springer International AG: Cham, Switzerland. doi: 10.1007/978-3-319-58694-6
Recchia, G. (2016). The utility of count-based models for the digital humanities. Abstract in proceedings of the Digital Humanities Congress 2016. Sheffield: HRI Online Publications, 2016. https://www.hrionline.ac.uk/dhc/2016/paper/98
Recchia, G. (2017). Fall and rise of AI: Computational methods for investigating cultural narratives. Invited presentation to AI Narratives: Workshop 1, Leverhulme Centre for the Future of Intelligence and the Royal Society, 16 May 2017, Hughes Hall, Cambridge.
de Bolla, P., Jones, E., Nulty, P., Recchia, G., & Regan, J. (2016). The Concept Lab. Invited presentation to The Stanford Literary Lab and Alan Liu’s research group at UC Santa Barbara.
Recchia, G. (2016). Tracing concepts through time. Invited talk at Natural Language and Information Processing Seminar Series, University of Cambridge, UK.
Recchia, G. (2016). Big data in the social sciences and the humanities. Invited talk/workshop at Cambridge AHRC Doctoral Training Partnership: Workshop on Big and Small Data, University of Cambridge, UK.
Recchia, G. (2015). Making sense of language: It’s okay to count. Invited talk at Microsoft Research Cambridge, UK.
Recchia, G. (2015). The unreasonable effectiveness of co-occurrence based models. Big Data Methods for Social Sciences and Policy, University of Cambridge, UK.
Working Papers and Presentations
Dr John Regan, Blog post: The Concept of System in David Hume’s The History of England
Dr Paul Nulty, Blog post: Titles in Digital Book Collections, 1700–1900
Concept Drift Workshop at International Conference on Knowledge Engineering and Knowledge Management, Bologna, Italy, 2016
Gabriel Recchia, Ewan Jones, Paul Nulty, John Regan, and Peter de Bolla presented: Tracing Shifting Conceptual Vocabularies Through Time
Gabriel Recchia, Blog post: Numberless Degrees of Similitude: A Response to Ryan Heuser’s ‘Word Vectors in the Eighteenth Century, Part 1’
Nulty, P. Detecting and Visualizing the Linguistic Structure of Political Concepts. presented at: Applied Quantitative Text Analysis Conference, LSE. 2017
Nulty, P. and Regan, J. Methods for Distributional Concept Analysis. presented at: The Digital Archive Today Colloquium, Faculty of English, University of Cambridge. 2017
The Cambridge Centre for Digital Knowledge (CCDK) constitutes an ambitious response to the current state of digital knowledge, in a form that enables swift, scalable and dynamic response to rapidly changing intellectual, cultural and technological conditions.
The premise of CCDK is that we are now entering the third phase of Digital Humanities. The first phase prioritized the digitization of analogue materials. The second phase involved the growth of a digital humanities discipline, which has promoted new working practices in the humanities and social sciences. One result of these two phases has been the facilitation and increased speed of access to data. The third phase now urgently requires new forms of understanding that will use new technologies to transcend rather than perpetuate well-worn approaches in the humanities and social sciences. The CCDK is structured around two strands of research which represent the two most pressing concerns of digital humanities: Digital Epistemology and Digital Society.
The Cambridge Centre for Digital Knowledge comprises of the Technology and Democracy project, the Concept Lab and the Digital Society project.