Bringing ideas to life

“We need to talk about Actor Network Theory”

Published in Method, Research Students by on May 14, 2013

A few weeks ago, these eight words were enough to catapult me straight into full-on panic mode…

When I say panic mode, I am of course exaggerating a little. However, this suggestion, which was raised in a recent supervision meeting, made me a little uneasy. This is mostly because I have some niggling doubts about employing an ANT approach for the analysis of my data. I am worried that if I choose to go down the road of a pure ANT approach, I may miss some of the more ‘human’ aspects of policy mediation and struggle. The particular policy process that I am researching can be regarded as highly emotive, involving much contestation, debate, argument, mediation and disagreement as well as agreement – after all, to achieve an end result some sort of consensus must be reached, but the process by which this has occurred has been complex. ANT can map all of this complexity and messiness; however, it ignored the role of human consciousness and intent, and cannot explain why particular things happen, or why they don’t.  This concern (and subsequent solution) will form the basis of my next post so I will drop that particular grumble for now and continue with the dilemma of picking a theory…

As you can probably guess, one issue that I have been grappling with lately is whether or not it is good practice to continue to use a theory when you disagree with some of its claims. This concern has been exacerbated by my irrational fear of hard-core social-material theorists, and like my fellow student Dave, I am worried about the consequences of not remaining ‘true’ to the authentic version of the theory that is being applied.

For me, social theory has only really started to come to life now that I have some data and I am trying to work out how to analyse it. It is therefore only in its application that it is beginning to make sense. I think that this is because I can now begin to see ANT concepts come to life within my data; ANT terms appear less abstract and confusing, and are instead taking on a whole new meaning, and to my initial surprise, some are becoming quite useful (but not all). However, I am pretty sure that this is a less than ideal way to go about applying social theory to research, and in a sense I may be working backwards.

As research students, we often hear that a theoretical framework, or research paradigm, should shape our research design, including the way in which data is collected as well as analysed and understood. However, it often doesn’t work out like this (and it certainly hasn’t within my research!). There is a lot of pressure on doctoral students to ‘pick a theory’ (see my colleague Michele’s recent post on ‘Theory Shopping’) and to try to find ways to tie it in with the phenomena that they want to research. However, the following characteristics of doctoral research can make this a difficult process:

  • Your research questions can change: the focus of your research shifts over time… Linking this with a suitable lens can be tricky when you’re not exactly sure what it is that you’re trying to look at
  • You discover new social theory as you go along, some of which may appear much more appealing than your original choice
  • The number of different activities and tasks which make up doctoral research and development are vast, overlap, and usually don’t work out exactly as planned: reviewing relevant literature, reading around the topic, research design, securing access to participants, data collection, data analysis, article writing, presenting at conferences, seminar and conference organising, the odd bit of teaching and marking…. This list is endless. Locating an adequate amount of time to devote to reading and understanding enough social theory to justify a choice is near impossible!

So my questions are, what if, following data collection, you come to realise that your chosen theoretical framework may be more restrictive than enlightening? What if there is a better theory out there for you and your research and you just haven’t come across it yet? Should you just ignore them and stubbornly continue with your chosen framework, oblivious to the fact that there is a whole other world of exciting ideas and perhaps more suitable theories out there?

This is the stage that I was at a few weeks ago. But I have now decided exactly how I would like to use ANT, and picked out the concepts that I find most useful. I have combined these with elements from another social theory, which help to illuminate the areas of my research that ANT ignores. Essentially, I am cherry-picking the concepts and ideas that best describe and explain the phenomena that I am researching – I will write more about this in my next blog post.

But this does make me wonder, should a good knowledge of social theory be deemed essential for being accepted as a doctoral student in Social Sciences? Or should we be more open about the uncertainty, confusion and concerns that surround the processes of choosing the most suitable theoretical framework and then applying it? Perhaps we should admit that it’s not the end of the world to change, adapt and improve your theoretical framework in line with the shifting focus of your research questions and on discovering something you didn’t quite expect in your data.

The process of choosing and applying social theory is not straight forward – I wish I had known this at the start!


Enhanced by Zemanta
Spread the love!

About the author /


Anna is a final year PhD student in the School of Education, University of Glasgow. Her main research interests centre on educational change and policy making/implementation. With a background in Psychology, she is particularly interested in the way in which teachers shape and engage with educational change.


  1. Huw Davies
    Huw Davies

    Hi Anna, I helped organise a workshop on social theory in Paris last month. This guy was our brilliant keynote speaker.
    He runs Latour’s lab – he said ANT was “not a theory it was a method”. If you have any questions on ANT he’s an expert and he’s very approachable, I’m sure he’d help you out.

  2. Mark Murphy
    Mark Murphy

    Thanks Huw – but then why don’t they call it ‘Actor Network Method’ then?? Can it not be a theory and a method at the same time?

  3. Huw Davies
    Huw Davies

    I don’t know, I’ll ask a colleague who knows far more about ANT than I do – he was also at the workshop. While I was in Paris I also went to talk by Latour where he basically said the identification of a network is a construct because we have to “shave” it and detach it from its other nodes and connections – we create the entity we are analysing. He called this the “illusion of zoom”. I’ll try and find a record of it somewhere.

  4. Avatar
    Anna Beck

    Hi Huw,

    Thanks for your comments and the link to Dominique. I will definitely check him out!

    I’m not sure about creating the network that we want to analyse – or why we have to create it if it is already there – sounds dangerous! If you manage to find it, my email is: Thanks for your help!


  5. Avatar
    Harry Dyer

    Only just found this. I find that being honest and open about the shortcomings of any theory is a good thing. I’m adopting ANT after a rather traumatic realisation that it was necessary; I was hesitant at first but quickly realised it was great, and that I could augment it into a hybrid theory that adequately suited my ever changing research questions. As long as you can leave yourself wiggle room with a less prescriptive methodology (which ANT actively encourages) then you can be reactive to the situations that occur during research. ANT allows you to “follow the actor”; keep the methodology loose and you’ll be happy!

    • Avatar
      Anna Beck

      Thanks for your comment Harry! Nice to hear that you have had the same experience. Yes I agree – a bit of wiggle room is necessary when applying ANT.. (I also have ever changing research questions!). Can I ask what you’re researching with it?

Post your comments

Your email address will not be published. Required fields are marked *