This entry is prompted by a question I received about PIA (Poststructural Interview Analysis) (Bacchi and Bonham 2016). I was asked if it would be acceptable to change the order of the “processes” listed for PIA. I will list the processes simply so that we all know what I’m referring to:

Process 1: Noting “What is Said” 

Process 2: Producing Genealogies of “What is Said”

Process 3: Highlighting Key Discursive Practices

Process 4: Analysing “What is Said” 

Process 5: Interrogating the Production of “Subjects” 

Process 6: Exploring Transformative Potential 

Process 7: Questioning the Politics of Distribution

What I have to say about PIA in response to the query about changing the order of the processes applies equally to the “seven forms of questioning and analysis” in WPR (see Table in Bacchi and Goodwin 2016, p. 20). 

My reply to the query was an enthusiastic “Yes, of course”. I have mentioned many times that the WPR questions are interconnected. The same is the case in PIA. And, as a result, there is overlap and hopefully fruitful conjunctures. 

When Jennifer Bonham and I developed PIA, we hoped that the language of “processes” would convey the kind of fluidity and movement between forms of analysis that we wished to encourage. However, the imposition of a listing and numbering of “processes”, as in PIA, or “questions”, as in WPR, can undo the best of intentions. There can develop a tendency to “tick off” certain numbered items and then move on, failing to reflect on the mutual imbrication of all the processes/questions. The processes/questions feed off one another. Treating them as separate activities imposes limitations on what can be seen. 

Allow me to attempt an analogy, though I am wary of analogies. Both PIA and WPR can be conceptualised as trips to places less well known and less often visited. However, the focus is on the travel itself, not on the destination. I think of it as a sightseeing trip with the joys of discovery along the way. The questions/processes serve as prompts to encourage certain lines of thought. They are not a recipe with either processes or questions as ingredients added sequentially. 

An example of the “listing” peril

I need to tell the sad story of “Step 7” in WPR to illustrate the dilemma of developing a research “tool” (I prefer the word “approach” or analytic strategy). If you look back to Analysing Policy (Bacchi 2009), the book that lays out the WPR questions systematically (Chapters 3 and 4), you will see that each time the WPR Table appears (pp. 2, 48) there is a list of six numbered questions, followed by the comment (with no number): “Apply these questions to your own problem representations”.

The Table was set up in this way because the last comment could not be listed as a question since, grammatically, it was an instruction not a question (on the challenges this situation throws up, see the Research Hub entry on 30 August 2023, called “Applying WPR to WPR”). It followed that WPR came to be described as consisting of six questions and everyone ignored the reference to “self”-problematisation. Since this reference did not have a number, it disappeared from people’s analyses (see Research Hub entries on 21 Oct 2018 and 5 November 2018). 

I decided I needed to do something about this “disappearing” of “self”-problematisation. In 2016, in the book with Sue Goodwin (Poststructural Policy Analysis: A Guide to Practice), the Table (p. 2) now includes a “Step 7” – “Apply these questions to your own problem representations”. Again, since the statement is not a question, I decided to call it a “step”. 

Over the next little while, it became clear to me what had happened. Making the “self”-problematising undertaking a “Step” implied that you could somehow embrace this practice easily. Making it Step 7 seemed to imply that you should undertake this exercise at the end of your analysis or paper, once you had done everything else.

The layout of a Table with numbered questions imposed a structure that had all sorts of unintended consequences. There is a reference in Poststructural Policy Analysis (p. 24) regarding the need to maintain a “self”-problematising ethic throughout one’s analysis. However, the placement of “Step 7” at the end of the Table undermines this commitment. 

The practice of “choosing” some questions

One effect of the “listing” peril is that the “questions” in WPR are thought of as separated from one another. Such a separation leads to the practice of choosing some questions as the basis of one’s research and removing others. I am well aware that, both in 2009 and in 2016, I stated that it was “possible to draw selectively upon the forms of questioning and analysis” in WPR (Bacchi and Goodwin 2016, p. 24). 

I apologize for this misleading phrasing. In the 2022 Keynote at Karlstad ( KEYNOTE ADDRESS – CAROL BACCHI – 18 August 2022 I note that “a researcher can foreground certain WPR questions and bypass others”. What I ought to have said is that it is possible to foreground some part of the analysis in a research paper. As an example, the researcher may want to highlight the politics of subjectification (from Question 5) or a particular silencing practice (from Question 4). 

As the WPR approach has matured, I take every opportunity to stress the linkages among aspects of the approach, the interconnections, the interactions, rather than the separations. This is clear in the 2016 reference to maintaining a “self”-problematising ethic, already mentioned (Bacchi and Goodwin 2016, p. 24). And in the Karlstad keynote address, I highlight the importance of developing a “genealogical sensibility” as an integral part of WPR. Hopefully, these interventions make it possible to think about WPR and PIA as “ways of seeing” rather than as lists of “steps” to perform. 

Producing an “integrated analysis” 

To acquire some grasp of the kinds of analyses these suggestions promote, I would recommend looking at the last five chapters in Analysing Policy (Bacchi 2009). Each of these chapters applies what I call an “integrated analysis”. If you read those chapters, you will see no separate listing of the WPR questions and no imposed sequential analysis. Instead, you will observe how the aspects of WPR/PIA thinking influence an argument. To achieve this effect, I experimented with inserting the notation, Q1, Q2 etc to signal when the thinking behind a particular question is being applied. There is no intention in these chapters to suggest that researchers ought to use similar notation, although it may prove useful as a theoretical exercise.

Why are the questions in the order they appear currently? 

What you see in the current listing of WPR/PIA questions/processes is one way of thinking through the material being analysed. The current listing illustrates the thinking processes as I developed them. This approach does not rule out other ways of “ordering” the processes/questions. 

One interesting suggestion sent me by a researcher who draws on WPR suggests making genealogy the priority and then seeking problem representations throughout this analytic process. The argument here is that such a repositioning of genealogy would make it easier to identify a plurality of problem representations. I am still thinking through the implications of this suggestion but would certainly encourage further experimentation of this kind.

Other possible models?

I have been thinking about the other sorts of visual representations people use to convey “directions” in a research protocol, other than a numbered list as in WPR currently. There are many possible models, often with circles or speech bubbles linked to each other with arrows. In a 2014 article Jennifer Bonham and I produced a diagram that attempted to map the network of relations which might apply to any field of statements, with statements used in a Foucauldian sense (Bacchi and Bonhan 2014, p. 186). The diagram is intricate and yet Jennifer and I felt it still did not invoke the “action, fluidity and potential variability of continually enacted relations” (p 185, fn 85) that we wanted to capture. To achieve that effect, we decided, the connected lines in our diagram need to be in motion! Tricky!

In the event, the discursive practices diagram has been taken up by very few researchers. In contrast, the WPR questions form the basis of numerous research articles in a wide range of fields and across many regions/countries. Simplicity, it appears, trumps elegance! 

How to handle the peril of lists

Can you use the questions as listed? Yes, but the message in this entry is to try to become so familiar with the questions/processes that you can hold them up together like juggling balls. My suggestion is to avoid treating the processes/questions as a formula and to resist the implication in their listing that they follow a certain order. The suggestion that Step 7 is something to tack on at the end of the other questions (if there is space) in an analysis indicates one of the clearest ways in which WPR can be misleading. 

I take some responsibility for the misunderstandings. The WPR approach has matured and continues hopefully to reflect the insights generated by the many researchers who find it useful. I do my best to find ways to make WPR more responsive and flexible. I genuinely appreciate your comments, suggestions and questions as prompts to make improvements. So please keep in touch at carol.bacchi@adelaide.edu.au

References

Bacchi, C. 2009. Analysing Policy: What’s the Problem Represented to be?Frenchs Forest: Pearson Education. 

Bacchi, C. and Goodwin, S. 2016. Poststructural Policy Analysis: A Guide to Practice. NY: Palgrave Macmillan.

Bacchi, C. and Bonham, J. 2014. Reclaiming discursive practices as an analytic focus: Political implications. Foucault Studies, no. 17, pp. 173-192. 

Bacchi, C. and Bonham, J. 2016. Poststructural Interview Analysis: Politicizing “personhood”. In C. Bacchi and S. Goodwin, Poststructural Policy Analysis: A Guide to Practice. Palgrave Macmillan: NY. pp. 113-121.