Last week I got a phone call from an anxious student.
I get lots of these calls each year. Luckily for me, most problems research students encounter have happened before so processes or tools exist to solve them. I like to think of myself as a mountain guide. All I have to do is help the student to find a form, person, book, website or service which can be used to solve their problem (actually, it’s precisely this kind of work which got me the nick name of ‘Thesis Whisperer’ in the first place).
This student’s problem was difficult and, to be honest, had me kind of stumped.
This student, let’s call him Bob, is an engineer studying production processes within the transport industry. Specifically he was trying to solve some difficult problems to do with construction, design and safety. Now most studies of this kind use a ‘qualitative’ approach. Qualitative methods, as you are probably aware, usually involve techniques like interviews, observation and the study of documents.
Unfortunately this student is an engineer and trained to solve problems with math – and so are his supervisors.
To his credit, Bob had trained himself in qualitative techniques without much guidance. He did interviews and observed people at work. He devised his own method of taking fieldnotes and used this data to develop a framework. The framework described, in detail, how the production processes worked and where the problems were occurring.
Bob is now fast approaching the end of his thesis, so he duly presented the framework at his completion seminar (a presentation we ask students to do before they submit for examination). As Bob tells it, the panel assessing his presentation were not happy because he didn’t have enough ‘theory’. The panel advised he put off submitting his thesis until he fixed this problem.
Bob found he just couldn’t action this advice. In fact, he had to admit he didn’t even understand what the panel meant by ‘theory’. Hence the phone call to me.
The advice to “add more theory” is, on the face of it, pretty useless. A thesis is not like a petrol tank – you can’t just ‘top it up’ with some missing ingredient. But by the end of the phone call I could kind of understand why the panel gave such confusing advice. I don’t know if you have ever tried it, but it’s terribly difficult to explain what a ‘lack of theory’ in a thesis means.
I hear that scientists have trouble explaining magnetism without resorting to complicated jargon. So it is, I discovered with ‘theory’. I blathered on a bit about theory being a framework or a lens through which you see the world. I talked about how some theories can explain and others can predict. I tried to use feminism, gender and patriarchy as examples of the differences between movements, labels and theories. I told Bob he needed to get ‘above’ the details and stop merely ‘reporting’.
In short, I failed miserably.
Bob was nice about it, but I’m certain I left him even more confused than before. The problem was : I was suddenly confused myself. Somehow I managed to catch a dose of ‘theory anxiety’ off Bob over the phone. If I had such trouble explaining it, did that mean I didn’t understand what ‘theory’ meant either?
I turned, as I usually do, to Twitter for some advice. I was relieved when most of the 50 odd replies I got back were versions of the same things I had already said. It was nice to feel I was with the majority, but I wasn’t any closer to helping Bob. So I fired up the magic Google machine and discovered two handy papers in, of all places, the organisational management literature. Both are about the role of theory in journal publications; one by Sutton and Staw called “What theory is not” and a follow up / rejoinder by DiMaggio called “Comments on What theory is not”.
[late addition: These links go through RMIT Library. If you can’t access them the full references appear at the end of the post]
Sutton and Staw’s idea was both modest and clever – they aim to explain theory in the negative. First of all they point out that adding more references is not the same as adding more theory. If you include references to other people’s theories you need to explain how they relate to your work and the argument(s) you are trying to make, not, as Sutton and Staw put it, use citations as a “smoke screen”. Staw and Sutton believe that theory should be specific and built (or adapted) for the purpose at hand – not just ‘imported’ into a paper to make it sound more profound.
The second important point Sutton and Staw make is that “data are not theory” – and here they describe Bob’s problem exactly:
“Data describe which empirical patterns were observed and theory explains why empirical patterns were observed or are expected to be observed.”
Likewise Sutton and Staw caution that “lists of variables or constructs are not theory”: by themselves they do not explain why. I think this is the crux of the panel’s problem with Bob’s framework: it’s a tool which describes or represents the situation without really interpreting it or making meaning. As Staw and Sutton put it: ” A theory must also explain why variables or constructs come about or why they are connected”. Likewise, they go on, a hypothesis is not theory because these are “… concise statements about what is expected to occur, not why it is expected to occur”.
Sutton and Staw include a lively section against theory and ask whether, in fact, we expect too much theorising, which I found enlightening, but might save for another time because I’m working on the assumption Bob needs more theory, not an elaborate argument about why he might not need it at all.
So far so good, but the accompanying commentary article by DiMaggio provided what I think are important additional considerations. DiMaggio points out that there is more than one idea of what Good Theory might be. He offers a couple of different ways of thinking about theory which were new to me and quite exciting (you know, in a nerdy way):
- Theory as covering law
- Theory as enlightenment, and
- Theory as narrative
‘Covering law’ theory is, as DiMaggio puts it, a set of “generalizations that,taken together, describe the world as we see (or measure) it. He points out that Sutton and Staw are implicitly rejecting this type of theory – but I think this is the kind of theory which most scientists are aiming to build (I could be wrong in this because I am not a scientist, so please feel free to correct me in the comments).
Theory as enlightenment is where theory is treated as a ‘surprise machine’ which clears away “conventional notions to make room for artful and exciting insights”. This is the use of theory I myself am attracted to (now I am starting to wonder why…). By contrast ‘theory as narrative’ attempts to provide models for why people (or machines, or cells for that matter) behave the way they do. DiMaggio goes on to suggest that the most successful theorisations are probably hybrids of these types. He also goes on to make some wonderful observations about how theory is taken up (or not) in academic cultures, but I am already 300 words over my post limit of 1000 so I will have to save that for another time too.
So what should Bob do to solve his theory anxiety? Given that he has a framework already I suspect the answer lies in the last point DiMaggio made about models. But I’ll leave that up to him to decide. If you are interested, I strongly recommend these two papers as starting points for thinking about how to use / develop theory in your thesis, but they should be read with caution for those of you not in the humanities.
Since I still have some teacher anxiety I’m wondering – how useful was this post to you? Do you have any objections, qualifications or extensions to offer? I’d love to hear from you in the comments – always the best part of any Whisperer post!
Sutton, R. I., & Staw, B. M. (1995). What theory is not. Administrative Science Quarterly, 371–384 http://haas.berkeley.edu/faculty/papers/stawtheory.pdf
DiMaggio, P. J. (1995). Comments on“ What theory is not.” Administrative Science Quarterly, 40(3), 391–397 http://web.ku.edu/~jleemgt/MGMT%20916/PDF/DiMaggio1995ASQ.pdf