Fields and Infrastructure? A comment to Fligstein on orgtheory

As a guest blogger on orgtheory Neil Fligstein started a series of posts about his and Doug McAdam´s new book “A Theory of Fields ” (Oxford 2012). It seems to me that to continue the debate on Instiutionalism and Infrastructuralism we should take his analysis that most of what is called NI today is – although continuously citing Meyer/Rowan, Dimaggio/Powell and of course the Powell/Dimaggio book – no longer working out the “old neo institutional” programm but try to deal with problems of ongoing activity, constant and gradual change and overlapping fields. In Fligsteins words:

One critical argument of the “new” new institutionalism is that actors are always jockeying for position in existing fields. They are always trying to better their situation and in doing so, can create change in both their position and the underlying order of the field. This produces two distinct kinds of change, the change whereby a new institutional order comes into existence and the more common situation whereby change is more gradual and continuous.


But this view of the world posits two radically different states, one where we can be agents and make our world and the other where we can do little about it. “A Theory of Fields” undermines this entire line of argument by asserting that actors are always acting and this means they are always struggling. They are in a battle for position and the game is always being played. This means that “A Theory of Fields” is part of a “new” new institutionalism that honors actors, sees purposes, interests, and identities, and allows for stuff to happen all the time.

Although I would reflexively add that what Fligstein seems to do here is itself the activity of an instiutional entrepreneur – trying to change the rules of the field or – if that turns out to be impossible – prepare the setup of a new field, I guess there are some common problematiques that both the proposed new new institutionalism and the infrastructuralist agend have in common: a focus on practice (The Theory of Fields draws heavily on Bourdieu), a focus on constant change, a curiosity for the question of how in a world of constant change patterned activity is produced and the focus on struggles (although we might be critical about the psychological undertone of the term struggle and rather speak of trials of strength).

On microfoundations and Barley

Seems the comment tool on Posterous is not working today, so here is my comment to Jan-H. about Barley, the neo-I crowd, and technology.

While I have always liked Barley’s work on technology, there is something I have got to get off my chest: in his super famous 1986 paper, the one about radiologists and new workplace technology (ASQ?), what was he really showing in that paper that made him so famous?

He showed that technology does not have a straightforward consequence when it enters a work place and instead can have different consequences in different workplaces. Well, I’d say “of course”. After all, the paper provides a counter point to a non-issue. Even in the literature at the time, it was almost unfair to ask: what is the single consequence of this technology for all work? Thus, his argument was pinned against is flimsy one, even for org studies.
Here is why: they have been searching for that answer since the 1960s with Woodward who assumed one could “unlock” the consequences of industrial manufacturing technology, namely, the assembly line, for work in general, laborers, managers, etc.
Guess what? She couldn’t. There was no single consequence.
Tech did not uniformly speed things up.
Tech did not shape each organization the same.
Tech did not … and so on.
Woodward assumed it would, or hypothesized as much, but Woodward was, in fact, learning what Barley would later test as a hypothesis the other direction. Barley assumed it the other way — that there would not be just one consequence of technology — and, unsurprisingly, found it. Well, in a way, of course he did; he found what scholars had been finding (but not looking for) for decades.

Microfoundations, institutions and two ways of studying technologies

Quite some time ago we had a couple of posts on the possible links between STS and Neo-Institutionalism (see here, here, here and here) and about how both camps can be fruitfully matched in their attempts to get a grasp a the black-boxed, taken-for-granted or institutionalized character of modern practices.

One of the basic lines of linkage we identified back then was this: while Neo-Institutionalism is great at pointing out the empirical details and explaining the diffusion and isomorphisms of patterns that are taken-for-granted (institutions), they lack (following Powell and Colways 2008) a perspective on the respective microfoundations. THAT on the other hand is something that (most) STS approaches are quite good at – but they on the other hand – see for example the underdetermined concept of black-boxing – lack an understanding of how the “functional simplification” (Luhmann 1997) that technology enacts is comparable to other forms of making something taken for granted: habitualization (in the bourdieuian sense), embodyment, signification, formalization, institutionalization.

After reading Barley´s and Tolbert´s 1997 paper in Organization Studies on Institutionalization and Structuration and after reviewing Barles´s research on technologies at workplaces I wondered a. if and how the Powell and Colways argument about the missing micro-foundations has ever been valid in institutional theory given the amount of thought that Barley and Tolbert are investing in designing their concept of scripts and the methodology to analyze them and b. why STS approaches to technology do not seem to play a large role in institutional analysis that deal with technologies on the one hand and why these institutional approaches to technology on the other hand do also not play a significant role in STS? Any thoughts?

Legit Infrastructure?

An interesting, new-ish scholar to look into is Ben Cashore, Professor, Environmental Governance & Political Science; Director of the Governance, Environment, and Markets Initiative at Yale (GEM) and Director, Program on Forest Policy and Governance.

Currently, I’m reading his co-authored paper on establishing legitimate non-state governance infrastructure, in his case, regarding the voluntary self-regulation for the development of forestlands and the sale of such harvests on the global market.

The paper is very similar to his other work, but it raises two great points worth considering:

1. While there is much ado about “governance without government” most governance seems to be about government in some way or another; put another way, most non-governmental goverance is in fact quite governmental in terms of its origins, functions, and composition. For example, the International Monetary Fund (IMF) was established by legitimate states and its body is mainly composed of individuals in or with strong ties to existing governments and it was established by states as a way for states to deal with international issues. This is good hidden-in-plain-sight observation: there is such a thing as goverance without government, but the “government” never gets too far from sight.

2. Because non-state mechanisms, such as voluntary self-regulation for the development of forestlands, have little or no power to enforce standards/norms and authority to penalize non-compliant firms, the legitimacy of these operations becomes paramount to understanding them. Their authority or power, which tend to be limited, are contingent on their legitimacy (note: this is a bit strong-handed, so please read the paper for a more nuanced interpretation).

Now, for the readers of this blog, this raises the issue of legitimate infrastructure. It is relatively rare to see work on infrastructure raise the notion of legitimacy, which is a concept that has many meanings and numerous analytical trajectories in various disciplines. In sociology, the new institutionalism is where I was first exposed to legitimacy arguments. Thinking back to Cashore’s work now, the development of non-state mechanisms such as voluntary self-regulation is influenced by the perceived legitimacy of the mechanism (and its relation to other mechanisms, conceivably), thus, might non-state infrastructural development follow the same underlying dynamic?

Sergio Sismondo on black-boxing and taken-for-grantedness

Concern over the relationship between processes of black-boxing and gradual taken-for-grantedness has been expressed a bunch of times on this blog — here, here, and here.

Gearing-up to teach STS to mainly engineering students today has me reading Sismondo’s intro text — and in Chapter 11, on the topic of “controversies”, he lays out the terms as follows:

Science and technology produce black-boxes, or fact and artifacts that are taken for granted; in particular, their histories are usually seen as irrelevant after good facts and successful artifacts are established (2010:120).

It is nice that the world of ideas in science is not auotomatically labeled “taken for granted” (when facts are momentarily settled) and the world of things in engineering is not automatically labeled “black boxed” (when artifacts are momentarily settled) so that the distinction is not reified (i.e., that facts are only taken for granted and that machines are only black boxed).

However, the two terms seem to be synonyms to Sismondo — do you agree with Sergio?

Greatest thing to happen to STS since the Bijker/Pinch paper

New interest in the micro-foundations of institutions has got to be one of the best things to happen to STS since the Bijker/Pinch paper…

The new institutionalism in organizaitonal analysis has been a well-spring for research. A quick summary of neo-I that Fabio Rojas and I wrote (in a paper on museusms):

The hallmark of the ‘new institutional’ school is the relentless focus on how life inside organizations is regulated by stable social practices that define what is considered legitimate in the broader external environment in which an organization operates (DiMaggio 1987, 1991, DiMaggio and Powell 1991b, Meyer and Rowan 1991, Scott 2000). The influence of institutions on organizational behaviour is supposedly most obvious in organizations like museums – organizations that new institutional scholars label as ‘highly institutional and weakly technical’ (Scott and Meyer 1991: 124). By this, scholars usually mean the following: that the organization’s leadership is highly sensitive to the expectations and standards of its industry; that the organization of work within the bureaucracy depends on broader ideologies and cultural scripts found in modern societies; that managers are likely to copy the practices of other organizations, especially high-status organizations; that professional groups are the arbiters of organizational legitimacy; that rational organizational myths and rules structure work practices; and that the ultimate performance of an organization’s set of tasks does not depend much on tools like assembly lines, computers, and the like (see also DiMaggio and Powell (1991a, DiMaggio and Powell 1991b).

The new approach/point of emphasis for neo-I folks is laid-out by Walter Powell and Jeannette Colyvas in their 2008 chapter in “the big green book” of organizations and institutions — copy of the paper is available in draft form at right here.

And so the story goes:

1. Older research is cast as calling for “the need to make the microfoundations of intitutional theory more explicit” (p.276). This is something that institutional theorists have had much success with — positioning papers to create the feeling that this idea is both something new and exciting but also that the call for micrcofoundations is an old one (that we need to now make good on). The opening lines of D&P’s 1983 paper does a good job of saying “that was then” and “this is now.”

2. The upshot: “much analytical purchase can be gained by developing a mirco-level component of institutional analysis” (p.276) which would link “micro-concepts, e.g. identity, sense making, typifications, frames, and categories with macro-processes of institutionalization, and show how these processes rachet upwards” (p.278).The invocation of “hierarchy” or “upward” levels is somewhat disconcerting for those of us set on flatter analysis, but there is likely room to show (and convince) that even the tallest, most stable actors and actions occur locally and laterally on a flat surface of interactions.

3. How can we, in STS, get some purchase on this?

A. Emphasize the interpretations of contexual factors (p.277) rather than assuming them (as has happened now and again in organizational theory devoted to field-level analysis — these are assumptions that occasionally must be made in order to do the diffusion studies so common in neo-I).

B. Display the on-going micro-maintenances of apparently stable institutional forms in daily practice AND/OR discover how stable institutional forms in daily practice result in change over time such that they transform the forms they are intended (in the behavioralist sense) to prolong.

C. Enliven analysis of actors — old new institionalism (let’s say) emphasized two types of actors, “cultural dopes” or “heroric ‘change agents'” the reason being that action was essentially assumed to operate at a level unnecessary to fully capture during large-scale field studies (i.e., so managers simply sought legitimacy at all costs, we assumed, and mimicked their peers) OR in the move to caputre the actions of real actors (instead of assuming organizational entitivity) the studies overwhelmingly invovled entrepreueurs and celebrated/worshipped their field-altering accomplishments, respectively. The new emphasis (of, let’s say, new new institutionalism) sort of smacks of STS lab studies where we saw the how the mundane facets of scientists’ behaviors in labs resulted in field-altering science. Now, neo-I wants to avoid momentous events, or, at minimum, show how seeming huge events were a long time in the making and like all experiments involved loads of failure, which demands of writers the ability to show how local affairs prompt shifts in conventions (locally or broadly) (p.277).

Why is this so good for STS? We have already done much of this type of work, and have oodles of folks committed to these axioms for analysis. The only thing we really need now is a bridge between these two camps — while STS could not break into neo-I on the topic of technology, Powell and Colyvas might have just opened the door to an new institutionalism in STS…


the role of reviews in the social science

All this discussion of new infrastructuralism and the infrastructural relation between “black-boxing” and “institutionalization” makes me think about the role of reviews in social science.

The question is: is a concept what it is, or does how it has been used constitute what it (now) is?

I think this for the following reason: Latour, in Actor Network Theory and After goes into something of a littany regarding the ways that ANT has been used and abused over the years since he and Callon (and Woolgar, honestly) thought of it. Of his many points, a meta-point matters for this post: he basically states that as his ideas spread, they increasingly got used in ways he did not expect and then Latour makes something of a value judgment in suggesting that some research, which appears to be relatively more current as compared to his original works, don’t do ANT right. Of course, Latour takes some blame in saying that perhaps the entire moniker including, A, -, N, and T were not perfect, it still seems like an odd point to hear from Latour. About 120-ish pages into Science in Action, Latour reviews as part of the translation model of how things spread (i.e., diffuse, though he considers this a dirty word) he demands that spread requires change — that a technology, for example, must change as it enters into new hands. This was a counterpoint to diffusion of innovation literature (that he hardly cites) and their supposed assumption that diffusion, as an idea and model, only works so long as we assume the innovation is “constant” over time (meaning that it does not and will not change). Getting to the point: ANT was going to have to change to spread so widely, and the ideas would necessarily be used in ways unintended and perhaps unacceptable to its originators.

Again, then, the question is: is a concept what it is, or does how it has been used constitute what it (now) is?

Latour contributed to the notion of “black-boxing” as much as perhaps any scholar of the last 30-ish years, and given his disappointment with how some of us have used his concepts, does it really matter? (i.e., this value judgment) Or, does it matter more for science not to judge how concepts have been used and instead document how they have been used because the way they have been used is effectively what they are?

Returning full circle, in all this discussion of new infrastructuralism and the infrastructural relation between “black-boxing” and “institutionalization”, what would make the best review paper? Review the terms as if they are not artifacts changing hands in order to conceivably arrive at some core meaning of these concepts, or review how the terms have been used and that this will tell us more about the operational meaning of the terms? 

Institutionalism and Infrastructuralism – some first thoughts on differences

In a recent post following Nicholas’s thoughts about blackboxing and taken-for-grantedness and about what that could mean for discussing the benefits of STS and neo-institutional theory, I asked: what are the difference between institutions and infrastructure? Nicholas and I discussed that today for the first time in detail and we thought it might be worth to post it to see if it makes sense.

Neo-Institutional theory is – to tell a very long story short – based on the question of how many different things (organizations, models, cultural forms) become similar over time. This is the basic problem in DiMaggio/Powell (1983): understanding the institutional isomorphisms if the impetus of rationalization is taken away. It is the problem that Strang and Meyer works on when studying the institutional conditions of diffusion (1993). It´s central focus was – like Powell argued in 2007 – on “the field level, based on the insight that organizations operate amidst both competitive and cooperative exchanges with other organizations.” (2007). DiMaggio (1988) and Powell (1991) both noted that this was a bit too smooth and that institutional arguments would need a more detailed perspective on contestation, fragility and struggles. Nevertheless the framework provided a fresh and new way to understand institutions – so productive that it framed a discipline or two.

Infrastructure studies, on the contrary, focussed on how things can appear systematic and highly-integrated but are actually implemented in many heterogenous, historically contingent local processes (Bowker/Star 1996/1996; Star/Ruhleder 2996). In some ways, diffusion becomes less important as implementation takes a more central role. Infrastructures are not build by system makers, but screwed together loosely by complex arrangements of interfaces, gateways and work-arounds, as Edwards has shown in 2003 and in his fabulous book on climate models (2010). However there seems to be a tendency to focus on normalizing and standardizing effects of classification systems implemented in large infrastructural settings – this is something like the Weberian “iron cage” of infrastructure studies visible already in “Sorting Things Out” and very strong in the works of Hanseth and Monteiro (1997, Monteiro 1998).

The link seems obvious, doesn´t it? Neo-Institutionalism starts looking at heterogenous stuff and finds it similar – too similar perhaps, so that it is missing the complexity of the social world sometimes. But it is a great framework for strong explanations. Infrastructure studies look at systems and find them fragile and fragmented inside. But they seem to lack the “big explanatory” power, which leads to giving up the focus on local multiplicity and emphasizing standardization/normalization instead. Could the strengths of both be added to get a good grasp at the installation of social order under (high) modern conditions?

Technologies/Black-Boxing and Institutions/Taken-for-granted – A question of levels?

This post started as a comment to Nicholas´ post on the museum but became so long that I decided to make it a new post. The debate on black-boxing and “taken-for-grantedness” (or STS & New-Institutional Theory) takles some very important points. They remind me of T.Pinch´s (2008) paper on Technologies and Institutions. Pinch´s focus is on the problem of “skill” and he argues that (new) institutional theory  – for example in its micro form as in the works of Fligstein – is basically only focussing on a very small aspect of ways to make an institution materially stable. Technology, he argues, adds at least a second way because it is black-boxed, not just taken for granted.

The reason why I think this is only partially true is that proposing such an argument is only possible by conflating levels of “taken-for-grantedness”: Sociology knows a whole spectrum of ways to make social order become taken for granted: from the taken-for-granted stream of everyday routines and interactions that make up Schütz´ life world to Mauss/Bourdieu´s techniques of the body that constitute the habitus, from B&L´s (or also Gehlen´s) processes of institutionalization to Foucault´s epistemé, Polanyi´s tacit knowledge and Ryles “knowing how”.Technology, if we follow this route, could be added to the book of tactics to make practice become taken-for-granted – through a very distinct process that has been described as “black-boxing” in which some aspects are packaged and sealed away, others delegated to specialists (for example for maintenance and repair). It is distinct from at least two other tactics exactly because of the form of this process. Embodying habits and skills for example is a process of becoming taken for granted by routine and repetition. Discursive closure is a matter of rhetoric, persuasion and concealment. 

Institutions and Infrastructures I suppose are strategies of “taken-for-grantedness” on a different level: they are hardy stabilized by just one of the discursive, habitual or technologcial tactics just described. Neither can an institution be based just on skills, nor on legitimizing and reglulating discourse, nor on technology. Hey, we know from a long time of STS research that not even technology alone can rely on technology alone. Institutions and infrastructures are complex installations – hybrids or monsters if you will. They both rely on a fragile architecture of “taken-for-grantedness” – plug-ins. What is the difference, then?

The "lighthouse" (re: Coase) in new institutionalism is the museum

Per recent discussions of black-boxing and institutionalization, a paper that Fabio and I wrote seems useful to remember. It was a piece we wrote to interrogate the use of “the museum” by new institutionalists of the organizational analysis-bent.

Sociologists that study organizations often analyze the museum from a cultural
perspective that emphasizes the norms of the museum industry and the larger
society. We review this literature and suggest that sociologists should take into
account the technical demands of museums. Drawing on insights from social
studies of technology, we argue that museums are better understood as
organizations that must accomplish legitimate goals with specific technologies.
These technologies impact museums and the broader museum field in at least
three ways: they make specific types of art possible and permit individuals and
organizations to participate in the art world; they allow actors to insert new
practices in museums; and they can stabilize or destabilize museum practices.
We illustrate our arguments with examples drawn from the world of contemporary

The black-boxing of a technology is different than the incremental emergence of “taken-for-grantedness”

Below is an excerpt about black boxing and taken-for-grantedness, which I wrote with Fabio Rojas years ago:

The black-boxing of a technology is different than the incremental emergence of “taken-for-grantedness,” which comes from writers such as Schutz (1967) and Berger and Luckmann (1966). They argue that knowledge in everyday life is taken for granted by individuals as reality, “but [that] not all aspects of reality are equally unproblematic” (p. 24).  They provide an example germane to this discussion:

  • …suppose that I am an automobile mechanic who is highly knowledgeable about all American-made cars. Everything that pertains to the latter is a routine, unproblematic facet of my everyday life. But one day someone appears in the garage and asks me to repair his Volkswagen. I am now compelled to enter the problematic world of foreign-made cars (p. 24).

Technologies, they contend, like those related to car repair, get taken for granted over time and through expertise. But by looking at technologies as black-boxes, scholars can gain a fresh perspective on the institutionalization of technology by emphasizing how stable technologies stabilize human networks, rather than how routinization results in a technology’s disappearance for organizational actors. Returning to Berger and Luckmann’s example, as a black-box, automobiles and the networks of dependence and exchange built-up around them are concealed (or ignored). Further, the patterns of human behavior that make a mechanic’s garage the one place to fix broken cars is missed because of the emphasis on how technologies get “taken-for-granted.”

Seems Berger and Luckmann’s (1966) old work on the social construction of reality might find new use distinguishing blacking-boxing from institutionalization. Also, please note: If there was one thing that early Latourian thinking and the new institutionalism in organizational analysis were looking to unlock was how something gets sealed-up and stabilized over time to the point of being taken-for-granted as real, true, or rational. Look back at the early pages of Latour’s book on Pasteur — it opens with the image of Rue Pasteur and asks how did we get this … a good question, no?

Black box" and "taken-for-granted

I recently asked:

When are the processes that bring about a black box the same as those that bring about — in the institutionalist frame — the notion of taken-for-grantedness … and, when are the processes that bring about either of these notions incapable of producing the other?

Seems, upon further reflection, to be an obvious paper, which might bridge some of the thinking about technology and institutional arrangements. Restated as a couple of thesis statements, it would go:

Q1. What circumstances/processes do the concepts of “black box” and “taken-for-granted” both meaningfully capture?

Q2. What circumstances/processes does the concept of “black box” meaningfully capture that the concept “taken-for-granted” cannot?

Q3. What circumstances/processes does the concept of “taken-for-granted” meaningfully capture that the concept “black box” cannot?

Seems like an interesting review piece to see where organizational theorists and STSers have historically overlapped and where they have diverged, with the caveat that each might learn something if orthogonal points of divergence where re-considered in the respective lines of research.