Originally posted: Thu, 09 Jun 2005 20:46:30
I finished this book a couple of weeks ago, but just haven't had the time to blog it until now. It's certainly bloggable, though, particularly since it represents another instance of the intersection between activity theory and actor-network theory (my current focus).
Knowing in Organizations was published in 2003, growing from papers presented at the 1998 Academy of Management meeting. Its focus, of course, is "knowing in organizations," a broad focus that draws in papers from four literatures: knowing as culture and aesthetic understanding; communities of practice; activity theory; and actor-network theory/heterogeneous engineering. The latter two have traditionally worked in different areas, but are beginning to run into each other more frequently. Activity theory is primarily a theory of distributed cognition, and focuses on issues of labor, learning, and concept formation; it?s primarily used in fields such as educational, cognitive, and cultural psychology, although it?s also making inroads in human-computer interaction, computer-supported cooperative work, communication, and anthropology. Actor-network theory is primarily an ontology ? an account of existence ? and focuses on issues of science, politics, rhetoric, production of facts, agreements, and knowledge; it?s primarily used in science and technology studies, philosophy, and sociology. But recently the edges of these projects have begun to meet and contend with each other. Activity theory, in its ?third generation," is attempting to move from the study of individuals and focused activities to the study of interrelated sets of activities, and thus into work organization. Actor-network theory is expanding from studies of scientific knowledge into popular science and technology, and from there into work organization as well. So it?s not surprising that advocates of the two approaches are beginning to grapple with each others? approaches, particularly in the pages of this book.
Niccolini, D., Gherardi, S., and Yanow, D. "Introduction: Toward a practice-based view of knowing and learning in organizations."
In the introduction, the editors discuss "the centrality and strategic importance of knowledge in postindustrial organizations" (p.4). An emerging "knowledge-centered" discourse has resulted in increased interest in "the issue of organizational learning and knowledge creation and management" (p.5). At the same time, there's widespread disagreement about what actually constitutes a useful understanding of knowing and learning; these categories and their vocabularies must be rethought, and "the main assertion of this book is that a practice-based vocabulary is a promising candidate for such rethinking" (p.7). The authors identify and discuss three main trends of thought on practice that undergird the work in this book: "Marx's work, phenomenology and symbolic interactionism, and Wittgenstein's legacy" (p.7). Then they go on to identify the "four practice-based ways to talk about knowing and learning" (p.12). (I disagree in that actor-network theory may talk about knowing, but it doesn't really attempt an account of learning.)
I won't review each chapter here, but I'll discuss highlights of the ones most relevant to my current work:
Wenger, E. "Communities of practice and social learning systems."
Wenger builds on his earlier CoP work here, providing detailed taxonomies of boundary workers, boundary objects, and boundary interactions. His detail stems from the fact that he sees knowledge work as becoming more distributed. "In a knowledge economy, sustained success for any organization will depend not only on effective participation in economic markets, but just as importantly and with many of the same players, on knowing how to participate in broader social learning systems" (p.98). The more distributed knowledge work becomes, and the more central knowledge work becomes to the economy, the more important it will be to understand and describe how workers interact across increasingly porous boundaries. Wenger sees these interactions in terms of learning.
Blackler, F., Crump, N., and McDonald, S. (2003). "Organizing processes in complex activity networks."
Unlike Wenger, Blackler et al. work within activity theory to conceptualize this sort of boundary work. In particular, they are working within AT's "third generation," in which activity networks are conceptualized and explored through boundary spanning, dialogue, and polycontextuality, and their changes and interactions explained largely through contradictions. Activity networks are typically pictured in two ways: either as modular, with each activity providing an "output" that functions as the "input" for another activity, or as overlapping, in which several activity systems might interpenetrate. Blackler et al. portray their activity network in the latter way (which I tend to think is more sensible). In this case study, Blackler et al. study a high-tech defense contractor with a special eye toward perspective making, perspective taking, and perspective shaping (p.127).
Blackler et al. introduce the notion of activity networks in an interesting way:
The question arises of how activity theory can be adapted for organizational analysis. One approach might be to analyze particular organizations as activity systems in their own right. However, partly as a result of the complex division of labor that exists in work organizations, participants' understanding of the links between their actions and the overall activity system of which they are a part can become obscured. While a level of internal differentiation between individuals and groups is inevitable in activity systems (of any size), complex organizations can easily become segmented and fragmented. (p.129)
So "second generation" activity theory is not enough for dealing with complex organizations. I would add that complex organizations such as the one Blackler et al. describe tend to be multidisciplinary, meaning that very different activities are "spliced" together. Accountants at the high tech firm, I suspect, have much more in common with accountants at other companies than they do with engineers or custodians or executives at their own firm. Furthermore, these activities have their own social languages, their own chronotopic senses (ideas of space and time), their own genres.
Rather than analyzing organizations as single activity systems, it is more satisfactory, therefore, to analyze them as networks of overlapping activity systems or, for simplicity of expression, as activity networks. The units that make up such networks can be labeled "communities of activity"; such communities can be loosely defined in terms of the extent to which members recognize shared work priorities, work with a common cognitive and technological infrastruture, and support each others' activity. Relations betweeen activity systems bring the (sometimes difficult) issues associated with multidisciplinary work into sharp focus. Collaboration across different systems of activity raises issues concerning priorities, identities, and operational methods as well as questions about relative authority and influence. (p.131)
The authors provide a modified activity system triangle meant to represent the entire activity network -- a very different diagram than, say, Bodker or Miettinen, or Engestrom would produce, one that emphasizes the activity network as thoroughly interpenetrated activities (p.132). Blackler et al. make the following substitutions around the triangle:
- subject -> community of activity
- community -> related communities of activity
- object -> emerging collective object of activity
- mediating artifacts -> contextual innovations (perspective shaping)
- rules -> boundary innovations (perspective taking)
- division of labor -> domain innovations (perspective making)
The terms in parentheses above are defined as follows:
- perspective shaping: "How the community of activity understands current priorities and imagines the future."
- perspective taking: "Relations with other communities of activity"
- perspective making: "The contribution of each community of activity"
I find this formulation to be quite interesting, particularly the thought that Blackler et al. have put into naming and theorizing the interpenetrations among activity systems. They have avoided depicting the activity network as modular, and consequently they have had to confront how these activity systems overlap at every point. The substitutions above are clearly ways to describe such overlaps or interfaces, highlighting the innovation work in which people are inevitably engaged as they negotiate these interfaces. Localizing the perspective work on the triangle also makes a lot of sense to me, and it delivers on what is too often a vague discussion of perspectives in activity theory. Perspective work is mediatory: "Cooperative relations between communities of activity are mediated by the processes of 'perspective making,' 'perspective taking,' and 'perspective shaping'" (p.134, their emphasis). Furthermore, the discussion leads to a principled understanding of expertise, which is not always easy when you mash together several different activities. "Through the concept of 'activity system,' expertise can be studied as a collective, heterogeneous phenomenon" (p.133, their emphasis). By the end of the case, we can see why the authors believe that 'perspective making,' 'perspective taking,' and 'perspective shaping' are "the core organizing processes" within activity networks (p.143).
In the footnotes, the authors provide a useful three-paragraph overview of the development of activity theory (p.147). They also repeat Miettinen's criticism of actor-network theory as emphasizing "fixed patterns of power and domination" (p.148).
Engeström, Y., Puonti, A., and Seppänen, L. (2003). "Spatial and temporal expansion of the object as a challenge for reorganizing work."
This chapter provides an introduction of the notion of object in work, based on Engestrom's developmental work research, before overviewing three cases. (One is Seppanen's dissertation work; I reviewed the dissertation elsewhere on this blog.) Objects of work are important here because activities are oriented toward objects; "there is no such thing as objectless activity" (p.152). Indeed,
We argue that a new, more interesting insight into the developmental dynamics of timing and spacing in work organizations can indeed be gained if we shift the focus of analysis onto the objects of work. We suggest that the ongoing historical transformations in objects of work are best conceptualized as expansion rather than compression. (p.152)
Objects are constructed and invested with meaning by means of cultural tools. Such mediating tools operate not separately but in complex constellation we call instrumentalities. Emerging new objects call for and generate new instrumentalities. (p.152)
The authors then turn to Victor and Boynton, whose book has been showing up a lot in Engestrom's work lately. Victor and Boynton
suggest that we can examine the evolution of work in capitalism as a succession of five major types: craft, mass production, process enhancement, mass customization, and co-configuration. The last one of the five, co-configuration, is particularly interesting from the point of view of the spatio-temporal expansion of the object. (p.153)
Co-configuration involves building both a continually adaptable product and an ongoing relationship with the company. (Engestrom et al. make co-configuration sound much more concrete and reified than it is; Victor and Boynton present it as a speculation on where work is going next, and do not confidently present a case study of it, as they do with the other forms of work.)
Engestrom et al. follow up with the case studies, which are worthwhile examples of object construction.
Suchman, L. Organizing alignment: The case of bridge-building.
Lucy Suchman's case study of bridge building is categorized within the sociology of translation, although I usually don't think of her as an activity theorist. But in this case study she grapples with the issue of multiplicity that inheres in large projects:
This story of bridge-building points as well to the multiplicity of perspectives involved in such large modern projects. A view of artifact construction as heterogeneous engineering emphasizes issues of stabilization of human and nonhuman networks as central. Along with the contingencies of this process as seen by engineers, however, one can catch glimpses of other perspectives, colllected generally under the heading of "residents" or "citizens." In a real sense there are at least two different artifacts at issue, with associated networks of stabilization, that must somehow be aligned. Project engineers are immersed in a history and daily order of professional practice and practical exigencies. Their orientation is to moving the project forward accoording to the order of phases and timetables, toward the production of an artifact within budget and with appropriate projections of maintainability and durability. Residents, on the other hand, are working on a different order of stabilization: that of their daily lives. ... These two different 'stabilizations' -- of artifact, careers, professional networks, on the one hand, and of daily life, property, and so forth on the other -- comprise different, only partially intersecting fields of knowing and acting. (p.200)
Very nice. Suchman's characterization of "stabilizations" and multiplicity are indeed in line with a sociology of translations, and they highlight what translation has to offer to organizational studies.
Law, J. and Singleton, V. "Allegory and its others."
This last essay is about the same vintage as Law's After Method, and covers some of the same ground. Law and Singleton draw on a case study of "typical patients" at a medical center (and they use the term "typical" ruefully, recognizing that there is no such thing). Although the piece is written in Law's start-stop style, which I find grating, the piece itself is worth it particularly for the criticism of the notion of object. Although they don't use the term "object" quite as Engestrom et al. do, the shoe still fits:
Perhaps there is simply something diffuse about the object itself, the compartment of alcoholic liver disease, alcoholism, alcohol abuse. Perhaps it simply slips, slides, and displaces itself. Perhaps its boundaries move about from one location to another, and do not stay still. Perhaps they ebb and flow. But if this is the case, then something similar goes on, too, for the patients, clients, citizens who experience this condition (or set of conditions). (p.240)
Despite Law's tendency to use "perhaps" when he obviously means "certainly," this passage is useful. Law and Singleton are discussing multiplicity here, and you may recall that multiplicity escapes dialectic. If the object and subjects both have uncertain, fluxing boundaries -- something that is implied by a sociology of translation -- it becomes hard to confidently stabilize either one enough to talk about spatial and temporal expansion of the object or development of the subject. Again, although activity theory and actor-network theory are increasingly being brought to bear on the same problems, we can see why ANT is unsuitable for providing a developmental account -- and why AT is having to work so hard to provide an account of polycontextuality! (AT seems to be having more success at its end, though.)