Actor-Network Theory, Knowledge Work, and Digital Libraries
 
Nancy Van House
School of Information Management and Systems
vanhouse@sims.berkeley.edu
 

This is adapted from a proposal funded by the UC Committee on Research under their Research Bridging Grant Program, 1999-2001.
 

Abstract
 

Information technology is advancing much faster than our understanding of its role in social, behavioral, and organizational dimensions of knowledge creation and use. To design effective information systems, we need to better understand the relationships among the social and material bases of knowledge work and communities, and information practices, artifacts, systems, and institutions. This project will study the processes of knowledge creation, with particular attention to issues of trust and credibility. It will focus on a specific knowledge community and an existing or emerging digital library. It will address in particular whether and how Actor Network Theory can aid in understanding the processes of knowledge creation and use and of digital library creation and use.
 

Social Informatics of Digital Libraries and Knowledge Work

I am part of an emerging interdisciplinary group of researchers concerned with social aspects of information technology, with an emphasis on information activity. We are calling this emerging field Social Informatics. It includes researchers from anthropology, sociology, information systems, management, library and information studies, computer science, and other fields who share a focus on the interrelationships among information technology, information, and society.

My particular concern is with the relationships among the social and material bases of knowledge work and information systems, institutions, and artifacts. I have become involved in the developing area of Digital Libraries (DLs). A DL provides some combination of digital information content and services. Most DL research is concerned with technology and functionality (e.g., Digital Libraries '98, 1998). This is important research, but is not the whole story. The technology of DLs is advancing much faster than our understanding of its role in human, social, behavioral, organizational, and work dimensions of knowledge creation and use.

DLs support cognitive or knowledge work. People use information systems for a purpose, to support their work (broadly defined). The user's concern is not with the DL, but with the doing and the learning. The user comes to the DL or other information system it with not only a task but a history and a context which affects what her work is, how she does it, and how she interprets and uses information. The interaction with the DL is wrapped in a larger learning and task domain.

DLs and other developments in information technology are changing the social and material matrix of knowledge work. Effective DL design is not simply a matter of converting existing information practices and artifacts to a digital world, but understanding how people do their work, use information, and create understanding, and then how DLs both shape and are shaped by those processes. Work and tools evolve together: the DL affects the work that is to be done, and how, and the work affects how the DL is used and ultimately how it is designed.

If we wish to design effective digital libraries, then, we must first consider the work, the human activity, that DLs support. We must be concerned with how people learn, individually and collectively, and how they use information artifacts (documents, images, maps, databases, and so on) to do their work, to extend their work over time, and to share their work with others, across space and time.

The importance of DLs in potentially transforming processes of knowledge creation and use is indicated by the importance placed upon it by NSF and other research agencies, including the original Digital Libraries Initiative funded by NSF, NASA, and DARPA, and two more recent major initiatives, Knowledge and Distributed Intelligence in the Information Age from NSF, and the Digital Libraries Initiative Phase 2 sponsored by NSF, DARPA, NASA, the National Library of Medicine, the Library of Congress, and the National Endowment for the Humanities. Both these programs include a strong social/behavioral component, noting the need for research not only on the technology but also it social implications.

A closely-related area (but with virtually no overlap in researchers or professional communities) is knowledge management, which is currently of great interest in management (e.g., Cohen, 1998). Knowledge management (KM) is concerned with how organizations create and capitalize on their members' knowledge. Although KM is enabled by information technology, most proponents describe it as being primarily concerned with organizations and their members, organizational culture and processes, and learning. KM is currently of great interest in the business world, indicating, I believe, a perceived need to better understand knowledge creation and use and the role of technology and information artifacts.
 

Intellectual Bases

A major weakness of research in Library and Information Studies has been its lack of a strong conceptual base. So where do we find an intellectual basis for understanding knowledge work and the creation and use of information artifacts? And what are the methods of this kind of research? To answer these questions, I have turned to research in a number of areas, all new to me.

Usability Assessment and User-Centered Design

One area I investigated was the growing body of work on usability assessment for computer-based systems and on user-centered design. Usability assessment and user-centered design take the user's point of view in designing and evaluating systems according to how well they support the users' work. This focus on the user is critical, and both usability assessment and user-centered design have developed useful methods and heuristics. They are useful for professional practice, and I am now teaching courses in these areas (e.g., IS214, Needs and Usability Assessment of Information Systems, Spring, 1999: http://www.sims.berkeley.edu/courses/is214. As bases for research, however, these areas share four major limitations. Both essentially pick up the user at the point of interaction with the system, ignoring the larger context of the user's work. Both are generally concerned with system functionality and interfaces but not the content and information functions of systems. Neither is concerned with the large impact of the information system on individuals, organizations, and society. And both have been criticized for a disconnect from their supposed conceptual bases.
 

Situated Learning, Work Practices and Technology

To understand information creation and use and information artifacts in knowledge work, I have turned to several areas of social science. For understanding of learning and knowledge creation, I have looked at the research on situated cognition (e.g., Hutchins, 1995) and situated learning (e.g., Chaikin and Lave, 1993; Lave, 1988), coming out of social theory, cognitive science, anthropology, and education. For understanding work practice and technology, I have looked at work coming primarily out of a part of anthropology ( e.g., Blomberg et al, 1994, 1995; Suchman, 1987, 1994; Suchman and Trigg, 1993). There is also highly relevant work in organizational studies (Rochlin, 1997).

To summarize briefly, learning and knowledge work are done in and by communities of practice (including work groups, disciplines, and the like). Communities of practice (Lave and Wenger, 1991; Wenger, 1998) share the concrete, daily practices by which work, including knowledge work, is done. They share the tools and that both shape and carry their work, the material bases of their work. And they decide what is known, what counts as evidence, how an argument is constructed, what is considered proof; as well as who decides, who speaks, who is a competent observer. Communities grow and are transformed by educating new members. Learning is not simply a matter of knowledge transfer but of its construction and reconstruction in action, in interaction with a setting and with other people.

Artifacts of all kinds, including documents, images, and other information artifacts, both carry and shape the work. The genre of Environmental Impact Report, for example, determines, not only how the work is reported, but what work is done. The document itself then takes on a life of its own and carries the work across space and time, to be used by others.
 

Science Studies

Another area of research that investigates the processes, practices, and artifacts of knowledge work is science studies (Shapin, 1995; Hess, 1997). In our society, science is the prototype of rationale modes of knowing and of knowledge construction and validation. Yet ethnographic studies of science have concluded that what scientists do is not fundamentally different from ordinary activity (e.g., Latour and Woolgar,1991). Much of the work of science studies, I contend, is directly applicable to knowledge work generally and to DLs and other information services and systems, not only in science, but in other areas as well.

Science studies, also called, among other things, Science and Technology Studies (STS) and Social Studies of Scientific Knowledge Production (SSKP)(1), is a term for a collection of empirical studies and analytical perspectives concerned with scientific knowing, and with the practices and principles by which scientists, decide, collectively and individually, what they know. It is concerned with how the participants in science come to an agreement about what counts as fact or discovery, what inferences are made from facts, what is regarded as rational or proper conduct, how objectivity is recognized, and how the credibility of claims is assessed.

These approaches share, in general, two characteristics. One is a contention that scientific and technical knowledge is determined neither entirely by the nature of reason, on the one hand, nor by social influences, on the other (e.g., Law, 1986). The implication is that careful attention needs to be paid to the community of practice and its methods for determining whom and what to believe. They also share a methodological orientation based on careful attention to the actual daily work of science (as in Latour's (1987) memorable much-quoted principle, "follow the actor") and its artifacts, including the inscriptions (documents, images, graphics, and so on) that scientists use and create.

A scientist relies largely on others' reports of their research, and on methods and equipment developed by others; these are the foundations on which a scientist's own, original work is based. The recognition of trustworthy sources is a necessary component of all systems of knowledge. Each individual and community needs practical solutions to problems of trust and authority (Star, 1995).
 

Actor-Network Theory

One analytical stance that deals quite directly with the establishment of trust and credibility in science, and, particularly important for our purposes, the role of documents and other kinds of representations, is Actor Network Theory (Callon, 1986a, 1986b; Callon, Law, and Rip, 1986l Latour, 1986., 1987, 1991; Law 1986, 1990, 1992; Law and Hassard, 1999). Actor-network theory (ANT) is concerned with the processes by which scientific disputes become closed, ideas accepted, tools and methods adopted - that is, with how decisions are made about what is known. These decisions are often - usually - temporary, but closing the black box, in Latour's terms, of disputes allows people to take the work of others as a resource and move on, rather than continually reproducing and questioning it. According to their model, the work of science consists of the enrollment and juxtaposition of heterogeneous elements - rats, test tubes, colleagues, journal articles, funders, grants, papers at scientific conferences, and so on - which need continual management. They conclude that scientists' work is "the simultaneous reconstruction of social contexts of which they form a part - labs simultaneously rebuild and link the social and natural contexts upon which they act."

Without delving too far into some of the more the subtle (and controversial) arguments of ANT, it is nevertheless possible to use ANT as an analytical framework that helps us to understand the role of DLs in the production of scientific and other kinds of knowledge. My goal is not to "do" ANT (which would require that we address those subtleties) but a much more modest one: to see whether the orientation and some of the major concepts of ANT helps us to make sense of some of empirical findings about people's uses of information, information artifacts, and digital libraries.

Methodogically, ANT has two major approaches. One is to "follow the actor," via interviews and ethnographic research. The other is to examine inscriptions.

Inscriptions - including texts, but also images of many sorts, databases, and the like -- are central to knowledge work. Some (e.g., Latour and Woolgar, 1991; Callon, Law, and Rip, 1986) say that texts (including journal articles, conference papers and presentations, grant proposals, and patents) are among the major, if not the major, products of scientific work. Inscriptions make action at a distance possible by stabilizing work in such a way that it can travel across space and time and be combined with other work.

Texts are also central to the process of gaining credibility. They carry work to other people and institutions. They attempt to present work in such a way that its meaning and significance are irrefutable. And texts are where authors establish equivalences among problems, which Callon et al. (1986) identify as a major strategy of enrolling others. An important part of the standard journal article or grant application, for example, is to say, in essence, "If you are interested in X (major issue) you must be interested in Y, which is the topic of the work reported/proposed here."
 

The Proposed Research

I propose to select a community doing knowledge work that is using and possibly creating or contributing to a Digital Library and study their processes of knowledge creation and, in particular, of assessing and establishing credibility. My underlying question is whether and how Actor Network Theory can be used or adapted to understand processes of knowledge creation and use, and Digital Library creation and use.

In the research that I have been doing on Digital Libraries, I have found that Digital Libraries often threaten existing practices of trust within and across knowledge communities (Van House and others, 1998). Existing practices are tightly coupled with existing artifacts and genres. For example, DLs make possible the distribution of massive data sets, which previously were shared only in the form of published summaries and reports, analyzed, synthesized and interpreted. Electronic distribution of raw data raises questions both of the credibility of the data and its source, and the capability of users to appropriately re-use and re-analyze the data. It also changes the nature of people's work and responsibilities, as researchers, for example, become information providers - what responsibility do they have to ensure that their datasets are clean and documented for use by others? Are kept up to date?

Selecting an appropriate community to work with will take careful consideration. It needs to be a group that is located nearby, is actively engaged in knowledge work and the creation and use of inscriptions and digital libraries, and is willing to cooperate. One possibility is people engaged in environmental planning (with whom I have done some previous work) but I have found it difficult to find people who are actually doing environmental planning - much of the actual work is done by consultants, who bill by the hour and are not amenable to being subjects of research. Another possibility is people in a scholarly field that is served by an existing or emerging Digital Library.
 
 

REFERENCES

Blomberg, Jeannette, Lucy Suchman, and Randall Trigg. (1994). "Reflections on a work-oriented design project." Proceedings of the Participatory Design Conference (PDC '94) Chapel Hill, NC, Oct 27-28.

Blomberg, Jeannette, Lucy Suchman, and Randall Trigg (1995). "Back to Work: renewing old agendas fo cooperative design." Proceedings of the 3rd Decennial conference on computers in Context: Joining Forces in Design. Aarhus, Denmark, Aug 14-18, 1995.

Callon, Michel (1986a). Some elements of a sociology of translation: domestication of the scallops and the fishermen of St Brieuc Bay. In John Law, ed. Power, Action, and Belief: a New Sociology of Knowledge? Socioogical Review Monograph 32. London: Routledge and Kegan Paul, 1986, p. 196-233.

Callon, Michel (1986b). The sociology of an actor-network: the case of the electric vehicle.In Callon, Rip, and Law, p. 19-34.

Callon, Michel, John Law, and Arie Rip, eds. Mapping the Dynamics of Science and Technology: Sociology of Science in the Real World. London: Macmillan Press, 1986.

Chaiklin, Seth, and Jean Lave, eds.( 1993). Understanding Practice: Perspectives on Activity and Context. Cambridge U press, P. 144-178.

Cohen, Don (1998). "Report on the First Annual UC Berkeley Forum on Knowledge and the Firm." California Management Review 40:3 (special issue).

Digital Libraries '98: Proceedings of the 1998 ACM International Digital Libraries Conference. ACM.

Hess, David J. (1997). Science Studies: an Advanced Introduction. New York University Press.

Hutchins, Edwin (1995). Cognition in the Wild. MIT Press.

Latour, Bruno (1986). "Visualization and Cognition: Thinking with Eyes and Hands", in Knowledge and Society: Studies in Sociology of Culture Past and Present, 6, 1-40.

Latour, Bruno, (1987). Science in Action, Cambridge, MA, Harvard University Press.

Latour, B. and Woolgar, S. (1991). Laboratory life: the construction of scientific facts, intro by Salk, J., Princeton, NJ, Princeton University Press..

Lave, Jean (1988). Cognition in Practice: Mind, Mathematics, and Culture in Everyday Life. Cambridge: Cambridge University Press.

Lave, Jean, and Wenger, Etienne. (1991). Situated Learning: Legitimate Peripheral Participation, Cambridge, England, Cambridge University Press.

Law, John (1986). The Heterogeneity of Texts, in Callon, Law, and Rip, p. 67-83

Law, John (1990). Technology and Heterogeneous Engineering: the case of Portuguese expansion. In Bijker, Wiebe, and others, Eds., The Social Construction of Technological Systems: New Directions in the Socioogy and History of Technology. Cambridge, MA: MIT Press, P. 111-134.

Law, John (1992). "Notes on the Theory of the Actor-Network: Ordering, Strategy, and Heterogenity." Systems Practice 5:4 p. 379-393.

Law , John, and John Hassard , eds. (1999). Actor Network Theory and After. Blackwell Publishers.

Rochlin, Gene I., (1997). Trapped in the Net: The Unanticipated Consequences of Computerization.  Princeton NJ: Princeton University Press.

Shapin, Steve (1995). "Here and Everywhere: sociology of scientific knowledge." Ann Rev Sociol 1995:21 p. 289-321.

Star, S. Leigh, ed. (1995). Ecologies of Knowledge: World and Politics in Science and Technology, State University of NY Press.

Suchman, Lucy (1987). Plans and Situated Actioins. Cambridge University Press,.

Suchman, Lucy (1994). Working relations of technology production and use. Computer-Supported Cooperative Work (CSCW) 2: 21-39.

Suchman, Lucy A. 1990 Representing practice in cognitive science. In Representation in Scientific Practice, ed. By Michael Lynch and Steve Woolgar. MIT Press,. P. 301-322

Suchman, Lucy (1997). "Centers of Coordination: a case and some themes", in Resnick, Lauren, ed. Discourse, Tools and Reasoning: essays on situated cognition.  Springer, 1997.

Suchman, Lucy A. And Randall H. Trigg (1993). Artificial intelligences as craftwork. In: Seth Chaiklin and Jean Lave, eds. Understanding Practice: perspectives on activity and context. Cambridge University Press, P. 144-178.

Van House, Nancy A., Mark H. Butler, Lisa R. Schiff (1998). Cooperative Knowledge Work and Practices of Trust: Sharing Environmental Planning Data Sets. CSCW '98: The ACM Conference On Computer Supported Cooperative Work. Seattle, WA, November 14-18, 1998. Proceedings. ACM, p. 335-343.

Van House, Nancy (in preparation). "Digital Libraries and Collaborative Knowledge Construction." In: Ann P. Bishop, Barbara P. Buttenfield, and Nancy A. Van House, eds. Digital Library Use : Social Practice in Design and Evaluation, MIT Press.

Wenger, Etienne (1998). Communities of Practice: Learning, Meaning, and Identity. NY: Cambridge University Press.
 

1. There exist a variety of related (and sometimes feuding) research areas and terms, including science studies, science and technology studies, social studies of science and technology, and sociology of scientific knowledge production.