Friday, November 02, 2012

Creating an ANT research design

the question posed by Jeffrey Keefer over on the ANT fb page was:

I am wondering if anybody in our group has any experiences to share about creating an ANT research design. How did you create a research proposal or overview in an ANT-way? What challenges or opportunities did you face compared to more "traditional" research designs?
As i was studying change and the use of emergent technologies, i wanted a method that allowed me to look at change as it happened.
I looked at a few approaches: activity theory, grounded theory, and read E M Rogers diffusion of innovation, but made no progress apart from feeling they weren't doing it for me. They were not a match to my question.

After this broad reading my supervisor pointed me in the direction of Callon (1986) and scallops.that seems weird, but if you have read the article you will know it is about change, the introduction of an innovation, and that the actors involved- including the scallops- are discussed in a similar way.
As technology was so embedded within what was happening, a symmetrical analysis became important (both shaped and shaping was a continual theme for me). It was something i identified as missing in how the very human act of counselling was altered with the use of other media, specifically as it evolved with the use of sms text messaging (I did not set out to study sms, when I started multiple media were being attempted, was just that sms grew and grew for a network of reasons, and this network is what was studied).

In looking for more literature on this "new to me field", and to how to study this, my supervisor advised 'staying close to the fire' - to read the authors who were central in developing the tenets of ANT. Reading Aramis (Latour) and The body multiple (Mol) were persuasive for me. They both allowed for a multitude of actors to be involved, and they both approached what was occurring or had occurred with new insights. They both were written in ways "that sparkled". As i have written of here, having a taste of ANT, led me into a new way of seeing - or more accurately from an ANt analysis, a new way of being.

I chose to study change in a format similar to Aramis (Latour) but rather than retrospective it would be a work in progress; how things get loved rather than killed. This also sat well with the early Law stuff on hopeful monsters. And with the bush pump stuff by Mol and de Laet.

However, I also read Mol and with the body multiple my whole trajectory ripped with reality being multiple. How then to be with an entity whose identity was/is unsettled.

I found myself frustrated by descriptive accounts and wanted a 'so what' emphasis. at this stage in my ANT journeying "so what" did not loom large enough for me. I wanted to be world changing. I was still somewhat attached to earlier renditions of myself, of grounded theory making new knowledge or of critical social theory setting the world to right...

The lack of overt politics frustrated me, leading me more to Donna Haraway.
And back to what was described to me as a thinly veiled fictional discussion between Latour and Haraway (Latour,2003).
(There is a subplot here. A network of activities, of people, and of changes occurs simultaneously. tracing a beginning becomes more and more difficult when things dont/wont/cant hold still. An ANt way of being is to be lost in a kalaedescope.

But that doesn't really answer the question as set. How did i create a research design?
I recall that when i would say i would be doing an ANT analysis I was challenged by a senior academic who told me that it would only provide descriptive accounts and that it would be difficult for a thesis because it does not address the "so what" question; no recommendations, no conclusions...

I believed, with faith in my supervisor, that if other ANT researchers could do ANT research, then so could I. Plus I read some ANT blogging by Mejias a student just completing an ANt phd which gave me confidence that an ANt phd by research was possible. I also read Arthur Tatnall's (IJANTTI editor) thesis and an excellent one by Jensen, more similar to the health related field I was in. Later I also read the book produced by Spinuzzi based on his phd study (helped me understand why CHAT was not something i would be accepting. CHAT was also something another professor on my colloquium panel had suggested I should look at, so clarifying why I was "not going there" was important (lack of regard for all the actors, context not being explained, as Latour had said "the social is not a given but requires explaining"...and a lack of a positioned researcher). I also had a look at Inger Mewburn's thesis (of thesiswhisperer fame) as she too had made use of ANT.

Nonetheless early on i was ignorant/ naive. My proposal was about the influence of all actors, and that's why ANT. But the how was problematic. No ethics committee was going to accept that i would just 'follow the actors". So to some extent I had an ethics application written with imagined possibility.

The analysis was also problematic. I eventually decided to tell stories in a similar way to Latour's Aramis, and a bit of Mol's the body multiple.
And a bit of Law's aircraft stories, and more Mol with the logic of care. The writing up of data also being a bit more Law with pinboards and juxtapositioning.
None of which sits very well with a trajectory of a thesis question/argument.
Laws "After method. Mess in social science research" was useful to justify and to search for meaning being made in ways uncommon in theses. And Donna Haraway as well as Patti Lather were useful for my articulating a situated researcher.
But how to analyse a multivocal account from a network with no top nor bottom, led me to Deleuze and Guatterri, but there was still a mismatch with to justify not tracing every step of connection...much much later Law writes of this in regard to Salmon who go places humans cant.

But it been a bit of a struggle. I need markers within the paradigm of ANT. Arguing the "scientification" argument in credible academic studies meets resistance from other academics I know. Some academics, perhaps most academics, are still attached to an objective truth, yet being a situated actor telling stories of "truths' as known to the actors sharing stories with me, was leading to a very subjective analysis. But then I'm arguing that in a situated study it cannot be otherwise. This also provides my argument for not going back to participants for any validification...hows that for a new word!
Again this may not be accepted by some.
Past and present blur in ways that make bringing the past back into a future via the qualitative method of triangulation a suspect practice. There are other alternatives. There are multiple experiences shown as overlapping that provide some degree of substance to the happenings as they occurred.

And in deciding when to stop, again Law provides more useful guidance in his salmon paper...Latour's wry stop when you get to 80000 words or whatever the uni specifies seemed unlikely to be accepted by any academic i know :)

Along the way i developed a new mantra that settled my 'so what' naivety, for the whole point of ANT is political- "things can always also be otherwise". Rich descriptions bring this to light.

But what did i actually propose and how did i get acceptance of it?
I've gone back to my colloquium proposal (at Deakin University this is a 20000 word document pre data collecting that validates the study) and is defended in discussions with a panel. In this i justify a need to consider symmetry. I also provide a story that demonstrated reality as multiple, justifying looking at complexity.
I then presented data on what had changed, and of writers who had studied change noting a gap (technology poorly defined, poorly considered, despite some good sound writing on the complexities involved in change of media) and situated the study as one of change with changing media). Looking at the document now, it is not too bad :)
It did not say how I would analyse or present findings however, and i think this is something done differently with every ANT study, everyone of which is a performance...and i accounted for a performative turn within this proposed study document. I came to love being performative. The thesis as text also came to be presented in multiple ways reflecting my experience of multiple actors....some of whome "talk" very differently. And so i have a phd that includes imagery, text speak, multiple fonts (all of which also created hiccups on crossing mac/pc, word doc and pdf formats).

But more than what I proposed was also a background in networking and aligning. My supervisor was instrumental and a credible ANT person. Being a senior academic others accepted his supervision would provide the rigor required. While my proposal was liberally sprinkled with citations aligning myself within an academic tradition...and just maybe there was/is also something of a Macchiavellin tradition where having enough people/writers on side provides clout in having the proposal accepted.
Albeit with a proviso of crossing that bridge when we come to it of ethics and analysis.

Personally i have found the theory strong, the method of what to do so much harder. There are tenets of things to adhere to for a strong ANt study (see Latour's Reassembling the social) but there's no 'paint by numbers'.

Along the way I had to find other justifications; Mol on how to improve rather than prove was useful. Verran on multiple realities...Lucy Suchman on policy and documents.. Barad on identity.... and so so many more. I found the reading was not all in advance...but was pulled in as and when needed to make sense of what wasn't known as necessary until the data I gathered made it so.

A hugely iterative process.
The why, what's there, bring in data, discuss it, make some conclusions trajectory can be traced in what I produced...but really the thesis is so much more messy than this would suggest. I have a literature that is networked, it provides me a place to stand that differentiates myself and what is studied from some and how and why myself and what is studied fits better with others...
My literature review provided a place to speak from. It situates me, and it situates the research, and is written of here.

ANT is known in the doing, it may not exist in the abstract. A "thing" does not have an essence...but is itself a gathering, a unique performance on every occurrence.
Even were I to study it again, it could not help but occur differently; no-one swims in the same river twice.

Some of my references:
Callon, M. (1986). Some elements of a sociology of translation: domestication of the scallops and the fishermen of St Brieuc Bay. In J. Law (Ed.), Power, action and belief: a new sociology of knowledge? (pp. 196-223). London, England: Routledge.
Deleuze, G., & Guattari, F. (1987). A thousand plateaus: Capitalism and schizophrenia (B. Massumi, Trans.). London, England: University of Minnesota Press.
Jensen, T. E. (2001). Performing social work. Competence, orderings, spaces and objects (Doctoral dissertation). University of Copenhagen, Copenhagen, Denmark.
Latour, B. (2003). An imaginary dialogue on modernity. Retrieved January 31, 2006, from
Latour, B. (2005). Reassembling the social. Oxford, England: Oxford University Press.
Mejias, U. (2007). Networked proximity: ICTs and the mediation of nearness (Doctoral dissertation). Columbia University, New York, NY. Retrieved from
Mol, A. (2002). The body multiple: Ontology in medical practice. London, England: Duke University Press.
Mol, A. (2006, March). Proving or improving: On health care research as a form of self-reflection. Keynote address presented at the meeting of the 11th Qualitative Health Research Conference, doi:10.1177/1049732305285856
Mol, A. (2008). The logic of care. Health and the problem of patient choice. London, England: Routledge.