Every now and then I wonder why it is that we talk about research processes as things. Our research uses interviews, not we interview. Or we opt for surveys rather than we survey. Or there are focus groups, rather than we run focus groups. These research activity-things then become ‘methods’.
When we make researching into an abstract thing, we can attribute criteria and qualities to it and we can discuss its relative limitations and advantages. We can examine its history and speculate about its future mutations. We can rattle on for pages about a method, even write a whole book – no, a hefty handbook about each different type. But this abstracting – or nouniness to quote Michael Billig – worries me just a little. Let me explain why, using a specific example.
As an ethnographer I routinely think about participant observation. There is a big literature about participant observation and this explanation is typical:
Participant observation refers to a form of sociological research methodology in which the researcher takes on a role in the social situation under observation. The social researcher immerses herself in the social setting under study, getting to know key actors in that location in a role which is either covert or overt, although in practice, the researcher will often move between these two roles. The aim is to experience events in the manner in which the subjects under study also experience these events.
All well and good? Well, it’s certainly familiar. We are very used to reading these kinds of explanations.
Let’s start to play with the terms a bit… Rather than participant observation, we might think of an observant participant. This semantic move shifts the action away from observation as the key activity. It focuses instead on the participant. “I’m going to be an observant participant” is a notion more able to be challenged by other participants – “Well I’m observant too. What’s so special about your kind of observation, is it any better than mine?” This is a good question for an ethnographer who participates in order to enter the other person’s world. Their question challenges the hierarchy of knowledge created and recreated in research practice. And it’s nowhere to be seen in the headline definition given above.
However the BIG problem with ‘’participant observation’ is that it is a nominalization, a semantic move in which an action has become a thing. The trouble with making verbs into things is that the role of the active agent, the do-er- in this case a researcher- is obscured. Their agency, and the kinds of decisions that they have to make, become more difficult to see.
So let’s think instead of ourselves as researchers who are participating and observing at the same time – participating/observing. This move immediately raises the question of how possible it is to do both together. How likely is it that participating actually stops the researcher from seeing as much if they could if they weren’t engaged? How is it possible to remember everything that one observed while participating, particularly if the participating is still going on? And what is the effect on other people if the researcher simply observes? Is it ever OK to just observe – when might this be? Or when might it be OK to just participate? Turning the method into verbs does tend to highlight these kinds of dilemmas.
Making verbs into things also has particular effects. When something is a thing it can be dealt with in a particular way, its traits and characteristics mapped and scheduled. The research methods text authors seem to find it pretty easy to construct a nice neat technical plan for we ethnographers. In ‘participant observation’ they usually note four or five stages – establishing rapport and gaining access, entering the field and generating data, exiting the field, analysing data and producing a text. Yet in reality these ‘stages’ are not linear and not separate. They are not only blurred and unable to be separated but often very messy and tangled up.
So, take rapport… Rapport is actually not an event, but an ongoing practice. The researcher must keep working on their relationships with others right through the research – maintaining trust is important. This is often not easy because people want to know what the researcher is noting and thinking. But if they are told, they may not like what they hear and then what happens to rapport… but not to tell what they are thinking means the researcher is holding back which may not be ethical, and so on… Or the people in the study may be doing something dreadful or illegal… the practice of maintaining rapport is tricky. ‘Rapport’ also often becomes important data rather than being pre a data stage. And the designated third stage, analysing data, can also be pretty problematic. When observing and participating, the researcher often starts analysing pretty early on and they may also be writing/filming texts… again do they show and tell as they are going along? And if so what does this mean if they do reveal texts, or if they don’t. And as for exiting, this may never happen, and probably it shouldn’t before the researcher has some conclusions that they want to share. And so on. All of these difficulties are hidden when participating/observing becomes a technical thing. Participating/observing always takes various particular and situated turns and twists throughout a project. It’s more helpful to be aware of the possibilities and to be thinking about what they mean, and what they mean for participating/observing, than to be guided by a set of stages and criteria about an abstract and unnaturally sanitized research method-as-thing.
So that’s my question… Why do we go on thinking that this kind of methods noun-ing is a good idea?
Yes, thank you, this is a real issue, and it reminds of John Law’s work, a sociologist at the Open University (UK), formerly at Lancaster University, who has questioned the extent to which it is possible to ‘clean up’ or ‘sanitise’ reality: http://www.lancaster.ac.uk/sociology/research/publications/papers/law-making-a-mess-with-method.pdf. Nominalising reifies reality/knowledge, which is fine up to a point, because we need to fix what we are talking about (and avoid becoming overly solipsistic), but maybe we can envisage a way of writing research that starts by nominalising it, then progressively verbalises it and hedges it, to reflect that research is agent-driven, and essentially a process of building and challenging consensus about what ‘is’ and how it relates to everything else. Maybe ….
I suspect that it is because WE (royal plural) sociologists are pretending to be objective scholars in an attempt to mix it with the quantitative researchers 🙂
Yes. I think you are right.
Great post. With students I talk about ‘modes of inquiry’ instead of research methods or methodology. Usually this disrupts the default to undue abstraction and fetshisation.
It is easier to talk about, and think about, ‘things’ than it is to do so about ‘ideas’, ‘concepts’ …