healy on “the performativity of networks”

Kieran Healy has a wonderful new article on The Performativity of Networks, just published in the European Journal of Sociology. The article is useful for both its empirical claims about how social network analysis has transformed the world, and for its cogent summary of the performativity of economics as laid out in MacKenzie’s work on finance. Throughout, Healy takes a somewhat ambivalent tone towards the performativity thesis itself. That is, he carefully argues that the evidence for the performativity of network theory is as good as the case for financial economics, but does not make a strong claim that such evidence is overwhelming. Anyone interested in the performativity thesis or in the history of social network analysis should give the paper a read. I won’t spoil the empirics here, read the paper for the details. Instead, I want to focus on how we think about performativity.

Beyond the simple joy of reading the paper, I had two connected reactions. The first was about my lingering concern that MacKenzie and Callon are up to slightly different things, and thus that MacKenzie’s performativity is not Callon’s. I made a version of this argument in blog form a long time ago*, but the quick version is something like: MacKenzie is interested in whether or not theories can make themselves true (the “Barnesian” performativity, see Healy 2015: 177-181 for a compelling summary). Callon, in contrast, is not so interested in the truth or falsity of economic theories, but in the capacity for economic theories (broadly construed to include marketing, accounting, finance, etc.) and calculative devices to create economic objects and make economic action possible (his newer work on “economization” makes this more explicit). For paradigmatic cases of how economic theories and tools make economic action, I think of work by Bruce Carruthers and Wendy Espeland on double-entry bookkeeping as much as I think of MacKenzie’s story of Black-Scholes. Fabian Muniesa’s recent book adopts a similar approach, looking at everything from “traditional” performativity topics like derivatives to perfume marketing focus groups. Healy, following the debate in mainstream economic sociology in the past decade, focuses pretty narrowly on the MacKenzie version and the question of how much evidence we have for Barnesian performativity. This is important, and interesting, but I think is not the whole story of performativity.

My second reaction was about the way MacKenzie’s work on Black-Scholes has become the paradigmatic case for performativity, literally the basis of comparison for Healy’s account of the performativity of networks. Though Black-Scholes is certainly our best documented example, I’m not sure it’s our best one, especially if we are interested in both MacKenzie’s and Callon’s claims. Specifically, I wonder if we shouldn’t be as or more focused on the case of game theory and market design. Black-Scholes and the options market is a wonderful story in part because of its complexities – I love how MacKenzie tells the story of the Chicago traders uniting together as a mini social movement to lobby regulators and mobilize their peers to both legitimate and kickstart the options market. But those complexities also raise significant empirical challenges to some of the simplest versions of the “theories making themselves true” claims.

Although I have not studied its history in detail, I wonder if market design might be a cleaner case. It helps that the subfield literally calls itself ‘market design’ – no pretensions here of a simple camera! In 2012, Al Roth and Lloyd Shapley won the Economics Nobel for their work on “the theory of stable allocations and the practice of market design.” Applications of their work included the revised system for matching medical residents and hospitals, and matching kidneys and patients. It’s also interesting how part of the point of market design is to prevent strategic action and force the revelation of true preferences. In other words, a well-designed market saves actors the trouble of having to make complex calculations – they just reveal their beliefs to the market mechanism and the right allocations are made. These examples showcase another side of economics, the side that starts from the assumption of strategic interaction rather than anonymous market transactions and tries to design a market robust to shenanigans. Market-like behavior has to be coerced, and the calculative tools of game theory can help to do just that.

How would our understanding of the performativity of economics change if we spent more time on market design and related endeavors and less on finance? Where else should we be looking to see the performativity of economics?

* I’m pretty sure 2009 qualifies as “a long time ago” in blog years.

Author: Dan Hirschman

I am a sociologist interested in the use of numbers in organizations, markets, and policy. For more info, see here.

2 thoughts on “healy on “the performativity of networks””

  1. I think these claims are empirically overblown. We’ve found a handful of instances of academic models finding widespread diffusion, and then reasoned from first principles that those can be extrapolated to the widespread influence of academic ideas.

    The gender literature has the same problem as the humanities leaning economic sociology literature, because both draw heavily on symbolic interactionism.

    Sociology has landed I think on a rough consensus about the reproduction of durable social relations through interactions and ideological repetition in process. Plausible enough. But this is just a theory and extremely difficult to show. So we extrapolate from a few cases at ideologically concentrated social locations, like finance (Black Scholes) or a high-school weight room (Dude You’re a Fag) to claim that gender or economic ideologies are exponentially bigger than what we’ve actually observed.

    Some part of that is driven by wanting to lay blame at ideologies and culture we don’t like, but can’t observe. That’s paranoid, but Kieran’s paper looks like a nice break from that trend.


  2. Nice post. You may like to know that some of the issues you raise regarding performativity after market design were discussed in this workshop we report here http://easst.net/easst-review/easst-review-volume-34-1-march-2015/studying-the-failures-of-markets-for-collective-concerns-a-workshop-report/ . My guess is that market design does not only challenge MacKenzie’s version of performativity but also Callon’s (or so I speculated here: https://estudiosdelaeconomia.wordpress.com/2013/04/08/are-markets-matching-callon-and-roth/ ). All best, José


Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s