Interviewing the nonhumans

“Researching a Posthuman World” flickr photo by IaninSheffield shared under a Creative Commons (BY-NC-SA) license

Choosing to approach my research with a sociomaterial sensibility requires me to challenge the division between the human and material elements I encounter and with which I work. Since Twitter and the online world is highly mediated, it would be remiss, I’d argue, to fail to adequately account for the material actors. Bringing actor-network theory (ANT) to bear does not, however, mean that I should consider the social and then the material. Instead they are completely entangled and mutually constitutive (Fenwick, 2014), so what they are and what they do is not in isolation, but co-constituted. Herein lies somewhat of a dilemma, since at some points it might be necessary to talk about the effect of particular practice on a teacher or what a specific aspect of materiality (like a hashtag) achieves. What’s important though is not to forget that both of these are themselves actor-networks, or assemblages, and are also part of other actor-networks. For the purposes of analysis, it is sometimes necessary to narrow the focus to a single entity, provided we don’t forget the assemblage that is also being performed.

In the methods section of my thesis I’ve discussed the semi-structured in depth interviews, participant observation (as it is manifest in this context), the blog posts I read and the exchanges with their authors. In reviewing this section and how it fits into the thesis as a whole, it’s clear how anthropocentric my writing was. The transcripts, blogs and even tweets were the words of the human participants … but where was the materiality? To be fair, I hope I’ve managed to surface some of that as a result of my observations and ‘following the actors,’ but once more it’s the (my) human voice that is privileged. How then to do justice to the nonhumans? How to give them a voice?

The Uni library has just added to its shelves a copy of ‘Researching a Posthuman World’ by Catherine Adams and Terrie Lynne Thompson, and I got the privilege and pleasure of being its first borrower. If I hadn’t managed to persuade the library to secure a copy, I think I’d have had to buy my own, given how important it could be in helping me with the challenge I’ve just outlined. The central premise of the book is how one might set about ‘interviewing objects,’ although clearly not interviewing in the same way I did with my human participants. Drawing on the etymological roots of ‘interview,’ the encounter involves ‘seeing each other, visiting each other briefly, having a glimpse of.’ To interview an object then is more about capturing insightful glimpses of it in action as it associates with other human and nonhuman actors. This is perhaps no more than I’ve already been doing, but where the book moves me forward is in advancing eight heuristics which, when deployed, might enable those insights to emerge. The heuristics have been assembled as a result of the authors’ experiences using phenomenology and ANT in their research. They are offered as possible approaches to inquiry, rather than a prescriptive, fixed set of rules to follow, and as such, it may be that not all will be applicable in a single study.

Eight heuristics for interviewing objects

  1. Gathering anecdotes
  2. Following the actors
  3. Listening to the invitational qualities of things
  4. Studying breakdowns, accidents and anomalies.
  5. Discerning the spectrum of human-technology-world relations
  6. Applying the Laws of Media
  7. Unraveling relations
  8. Tracing responses and passages

These can help us attend to what objects do when folded within everyday practices, although in their summarised form above, they don’t reveal their full strength. In the book, each heuristic is of course unpacked in much more detail, and this allowed me appreciate which would be most relevant to my study. Unsurprisingly, those which had been influenced more heavily by ANT, rather than phenomenology, seemed more appropriate and drew my closer attention. For me, an especially helpful move was when the authors opened each detailed heuristic passage with potential ‘interview’ questions, so it is to them which I now turn.

1 Gathering anecdotes

Describe how the object or thing appeared, showed up, or was given in professional practice. What happened?

One obvious place to start would be to consider how Twitter arguably came to be the central medium through which teachers support their online learning, however that story simply isn’t in the data. It’s too general. Closer to the mark though are the ways in which individual people came to Twitter and how it wheedled its way into their practice. Asking how people got started on Twitter was one of the gentle introductory questions on the semi-structured interview protocol I drafted, so I do have data which speaks to that. This was also discussed in a few of the blog posts and an occasional #chat, so I had enough to help me write a section on this. (There’s brief outline in the previous post).

I was fortunate to have interviewed two people who were at the centre of ‘movements’ which started on Twitter. The #mfltwitterati has a lot to thank Joe Dale for and is marshalled largely through a hashtag and MFLTwitterers list as I briefly described here. Then there’s EduMatch, as conceived by Sarah Thomas, which has expanded considerably since its inception. Having been fortunate to interview these two people gave me one form of access to the two things with which there were involved, but I have to tread carefully, ensuring their voices aren’t the ones which solely speak for MFLTwitterati or EduMatch. I also need to find other places where these two objects do their work and influence or even enact practice.

I find myself asking, but what if I hadn’t had access to those two interviews? It was largely chance that I got to speak with people who were closely involved with some-thing appearing. Could the data with which I have become more familiar, in all their heterogeneity, have allowed me to tell the story of an object which appeared? I am considering a couple of contenders, but I’m not there yet and perhaps need to reread the data through and with this heuristic.

2 Following the actors

Consider the main practice you are interested in. What micro-practices are at work?
Who-what is acting? What are they doing? Who-what is excluded?
How have particular assemblages come together? What is related to what and how? What work do they do?Choose an object of interest. What is the sociality/materiality around it?

Applying this heuristic presents me with an embarrassment of riches. I’ve never been short of actors to follow, whether that’s the hashtag in general, specific educational hashtags, #chats, particular Twitter accounts (like @EduTweetOz), the retweet function, or the Like. In any of those cases though, it’s less about the object and instead turning to what is done, the practice that the object invokes. What is the assemblage of, say, the Twitter profile doing and how is it being maintained?

I’m minded to think that ‘follow the actors’ is one of the first principles many newcomers to #ANTheory become familiar with. Perhaps then it’s always been on my mind, and as a heuristic for interviewing objects, it comes more easily to me. The problem I have is less about how to apply it, but more about where to ‘make the cut.’ Which objects did I choose to follow more closely and why did I not follow others? Why for example did I pay more attention to MFLTwitterati and little attention to MTBoS?

3 Listening for the invitational quality of things

What is a technology inviting (or encouraging, inciting, or even insisting) its user to do?
What is a technology discouraging?

At last year’s ProPEL Conference, I was fortunate to meet Terrie Lynn Thompson, one of the authors of the book which prompted this post. The paper I presented discussed how nonhumans, hashtags in particular, coax, cajole, compel, constrain and confound what other actors do. It was only after meeting Terrie and becoming more familiar with her work that I became aware of the eight heuristics, and only after reading the book this week that I had access to more of the detail in the ways that they should be applied. I was relieved therefore when I saw this particular heuristic which seemed to exemplify the approach I’d relayed in my paper. The disappointing thing was not having been able to credit the originators appropriately, although in my defence, the book wasn’t published until after the conference.

Although I don’t take the notion of ‘listening’ too literally, I’ve found that looking for invitational quality helps too. Let’s once more return to the hashtag. When an author adds a hashtag to a tweet, whether they like it or not, Twitter takes a hand. When that tweet appears in people’s timelines, the hashtag it contains will have been imbued with interactivity by the underlying code and Twitter’s algorithms. It is turned into a hyperlink which, when clicked, executes the function to open the Twitter search page with the hashtag already added as the search criterion, and thus immediately provides a timeline of recent tweets which contain that term. Back in the tweet however, the hashtag is also coloured differently, so that anyone with even the most rudimentary web literacy will recognise that as an invitation to ‘click here.’ Depending on the platform through which the tweet is being viewed, hovering over the hashtag may also cause it to change, perhaps by becoming underlined – another invitation. The tweet author, hashtag and Twitter are inviting the reader into the assemblage, and if they take that up, become (in #ANTheory terms) translated. But what would have happened if Twitter didn’t code a colour change into hashtag when it appears to the tweet reader? That might be where the next heuristic can help

4 Studying breakdowns, accidents and anomalies

What happens if an object breaks or is unexpectedly missing? What practices then become more visible?

Picking up the previous point, what if Twitter’s algorithms didn’t automatically turn hashtags and urls into links which take you to the relevant information with a single click? Or Twitter handles, which also (at least in the browser version) provide a pop-out window with a stripped down version of the profile? If that automaticity was not built in, would readers be as likely to follow up the offered connection since it would now mean copy and pasting a url? Without the coloured highlighting, some might not even spot it.

One breakdown which might have been interesting to follow and document would have been that of the ‘fail whale,’ but it is no more. How else I wonder, are people prevented from accessing Twitter? Lack of wireless signal or Internet connectivity more generally? School Internet filtering policies? I’m not sure they constitute breakdowns though; does it really matter if you don’t have connectivity for a while? The retentive properties of the timeline mean you can usually claw back something you missed if it was that important.

Perhaps simpler breakdowns can be found as a result of mis-types, for example. If you weren’t able to use the predictive offerings provided when you begin to type someone’s handle, and you then get that wrong, the connections you were hoping to initiate will have collapsed. Mis-type a hashtag (or get the wrong one) and the tweet you launched won’t reached the audience you were hoping. Or sometimes the same hashtag become used by different people for completely different things, leading to unexpected consequences if you happen to be following a hashtag where that occurs.

All of the aforementioned perhaps do little more than disrupt the intended learning for a while. The real breakdown has yet to happen and if I’d known about this heuristic during the interview for the PhD programme, I’d have given a much better answer to the question “What will you do if Twitter goes away?” Now that would be a breakdown worth studying.

6 Applying the Laws of Media

This heuristic draws on the tetrad of McLuhan and McLuhan (1988) and poses the questions they proposed.

What does a technology/medium enhance?
What does it render obsolete?
What does it retrieve that was previously obsolesced?
What does it become when pressed to an extreme?

Applied to Twitter in the context of professional learning, it might be that it enhances our capacity to connect with people, thereby extending our reach and exposing us to a wider set of views. Some might say that renders other forms of professional learning obsolete, or that it helps bring back those points of contact lost when teachers struggle to find those times during the day for an informal chat. (It might ameliorate the loss of school staffrooms and reduction in the length of school breaks). If taken to an extreme however, time spent on Twitter might be taking from that which fruitfully be devoted to other tasks; some people have described Twitter use as ‘addictive.’

It’s important for me to acknowledge my lack of familiarity with the McLuhans’ laws here though, and that I probably settled for an easy, trite response by attending to Twitter in the whole. I’m not sure how appropriate it would be in dealing with other actors like those responsible for some of the micro practices described earlier for example. I’m also not so sure about the notion of obsolescence in here; some of the technologies and practices observed within teachers’ learning on Twitter are often discussed in terms of working alongside rather than replacing others.

7 Unravelling translations

How have particular gatherings come to be and how do they maintain their connections?
What unintended realities come into being as everyday practices unfold?
What is entrenched? Who-what is excluded?

This is an important heuristic I feel. It requires you to draw deeply on your sociomaterial sensibility, paying close attention to the way that practices unfold, are maintained and sometimes fade away. The other interview questions remind us to not lose sight of the political ‘intentions’ or consequences of the work that particular actors do. So, to draw on earlier example, Twitter’s algorithms might indeed make following a hashtag easier for us, but what is it doing for Twitter? When tens of people like an educational tweet for example, how did that happen, what are the consequences and for whom?

In my study, people have cited being able to access professional learning anywhere, anytime, and on their terms as benefits enabled by Twitter. That may indeed be true, but what unintended or ‘collateral’ realities, as Law (2011) refers to them, are brought into being? Might it become an expectation that teachers should be learning anywhere, anytime? And what if teachers have an ethical objection to the use of (apparently) free, but nevertheless commercial platforms in this way? It’s not uncommon to see tweets or blog posts stating that ‘all teachers should be on Twitter.’ Should they? And is it any different if those views are expressed by a newly-qualified teacher or an experienced and respected principal?

Interviewing objects as co-researchers

Given my academic and career history in recent years, coupled with the nature of the topic I’m researching, it’s hardly surprising that the digital plays a significant role in my study. There was therefore always going to be a section somewhere in the thesis which reflected on this aspect of the study and how it influenced the process and the outcomes. Somewhat embarrassingly, until I read this chapter, I hadn’t considered how a sociomaterial sensibility might produce something different, but altogether more appropriate and coherent with the rest of the thesis. As Adams and Thompson (2016: 89) put it:

Situating the researcher as the sole arbiter in how a research project ultimately unfolds, while superficially correct, overlooks entirely the many subtle, but sometimes profound knowledge and practice implications at stake in digital technology integration.

The digital is and has been interwoven in everything I do. The literature I’ve accessed, whenever possible and right from the start, is in electronic format which means it’s always with me, ready at hand, and hopefully reducing the impact that hundreds of printed papers might have. The notes I’ve made, the posts I’ve written, the reports I’ve submitted and the presentations I’ve given, have been produced mainly in Google docs, although often with an MS Office backup. Gathering data involved: bookmarking using Diigo and Tumblr, converting to a format which could be stored/analysed using PrintFriendly, downloading using DataMiner and TAGS, connecting using Skype, recording using Audacity, storing and managing in NVivo, and analysing with a combination of NVivo, Google docs and various visualisation tools. This is just a sample and it perhaps behoves me to document them all for inclusion in an appendix. I’ve documented how some of these tools came to be involved in various posts within this blog, but what I rarely (ever?) noted was to consider what those technologies did. Instead of Ian Guest being the ‘sole arbiter’ of how this project unfolded, the different assemblages in which I was only one actor, will have rendered different outcomes. Even a single tool like Skype will have produced a different set of data when used with video and without; I’m reminded how my interactions with participants were different depending on which we were using.

In chapter four of the book, the two examples of technologies which were ‘interviewed’ were NVivo, the QDA package used to support analysis, and an iPod used in field research. These two examples have been chosen to illustrate how this type of object can be interviewed, however, I’m faced with such a long list of candidates, I couldn’t possibly interview them all, at least not within the constraints of a thesis (although that does present a possible topic for a paper …?). Once again then I am faced with making a cut.

In the same way the empirical example in chapter four illustrated the tension that using NVivo in a phenomenological study introduces, I too need to reflect on NVivo in the context of my research. I began to rationalise that in this post, so have laid the foundations for an ‘interview.’ Like chapter four, I’d also want to apply heuristic 3, though not heuristic 5. Instead, I think heuristic 1 is important to illustrate how a QDA package came to be involved in a poststructurally influenced study, and in fact that’s partially done within this post. Heuristic 2 might also me to flesh that out by considering some of the micro-practices involved in using NVivo the way that I chose to do, and what those practices produced as a consequence.

Thinking about heuristic 4 and breakdowns, and somewhat informally, since I wasn’t aware of the book at the time, I began to interview the @PLDBot, why it arose and what it produced. The shift by Twitter from Favorites to Likes initiated this process, but bringing the PLDBot into being subsequently produced unanticipated outcomes and a further ‘accident.’ A similar anomaly arose when Anchor introduced itself in one of my interviews, enrolled me, then broke down when it changed its functionality, tipping me into using Voxer. The interviewing process unfolded through a number of posts, although again less formally. Perhaps these episodes will make it into the thesis, but firmed up a little through applying appropriate heuristics.

One aspect of my study which will be making an appearance, but with what I’ve learned from the book can now be more robust and meaningful, is my use of visualisations. Different tools and I have worked together to achieve different things. Sometimes that was in recording the paths I’d taken during participant observation, at other times in helping to analyse data in different ways, and also in presenting data, concepts and ideas. Although in these linked posts I’ve reflected on what these visualisations did, I think that applying heuristics 1, 2 and 7 (at least) will allow me to interview visualisations and what they produced … but that’s for another post.


‘Researching a Posthuman World’ came to our library rather late for me, but not too late. I fleetingly speculated on how different my research might have been had I had access to it at the start of my study (perhaps I should ‘interview’ the book?!). I suspect however, that given my limited sociomaterial understanding at the time, I wouldn’t have been in a place where I could make the most of what the book offers. It’s not too late however. I’ll be returning to some of the data in the light of what I’ve learned, especially those which have something to say about the nonhuman actors. It will also definitely influence the way that I write certain chapters or sections. The one thing I felt it helped me less with however, is something that’s been nagging away at me for a while, the notion of ethics from a posthuman perspective. If researcher and technologies become bound together in assemblage, how does that shift the ethical considerations which should be brought to bear? The book does ask the researcher to consider the cuts which are made, what is included and what excluded. Whilst some of that becomes manifest through the choices I make, depending on the tools involved, ‘I’ will not be the only one making those choices. If for example, I’m following a Twitter discussion or exchange, what I see is completely reliant on the Twitter algorithms. Did I see all the tweets or were some cropped from the exchange for some reason? If I download a corpus of tweets based on a particular algorithm, did Twitter allow all of them through? And what of the application I used to do that and any visualisation it might produce? What is it emphasising or not? Without a knowledge of what the hidden algorithms are enacting, how can I know the extent of my ethical sensibility? If I use Google docs for taking notes or reflecting, or Skype to connect with interviewees, other than through the Terms of Service, how can I be sure that the data that assemblage produces is ethically sound?

Adams and Thompson (2016: 114) suggest that the questions surrounding ethics and the reflexivity they generate can be supported by engaging the heuristics. Interviewing objects decentres the human researcher and participants and reminds us of our ‘more modest roles as co-practitioners in the performance and reshaping of practices.’ Perhaps that’s all I should be aiming for?


Adams, C., & Thompson, T. (2016). Researching a posthuman world : Interviews with digital objects. Basingstoke: Basingstoke : Palgrave Macmillan.
Fenwick, T. (2014). Knowledge circulations in inter-para/professional practice: a sociomaterial enquiry. Journal of Vocational Education & Training, 66(3), 264-280.
Law, J. (2011). Collateral realities. In F. D. Rubio, & P. Baert (Eds.), The politics of knowledge (1st ed., ) Routledge.
Mol, A. (1999). Ontological politics. A word and some questions. In J. Law, & J. Hassard (Eds.), Actor network theory and after (sociological review monographs) (pp. 74-89). Oxford, England: Blackwell Publishers.
McLuhan, M., & McLuhan, E. (1988). Laws of Media: The New Science. Toronto* University of Toronto Press.

7 thoughts on “Interviewing the nonhumans

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your 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 )

Connecting to %s