1 On making an impression: A response to our critics John Boswell & Jack Corbett When we first wrote “Embracing Impressionism: Revealing the Brushstrokes of Interpretive Research” (Boswell and Corbett 2015a) we did not anticipate receiving any, let alone five, published responses. Our aim was more cathartic: to reflect on our experience of conducting interpretive research and contrast that with what we had learned about the method, both from the literature and our training. We wanted to highlight a disjuncture we felt between the notion of interpretive research as having a distinct ‘system’ analogous to other ways of doing political science research, and our own experience with the method. We struck upon the analogy of impressionism as a provocative way to do that. We sought to defend a version of interpretive research as a consciously ‘impressionistic’ craft—one that remains unsettled and uncomfortable for its practitioners and one that leads to interesting and important, but also partial and stylized, findings. Here we reply to some of the concerns raised in these five critical responses. It is not possible to do justice to them all. So, we extract what we feel are the key criticisms and respond to each in turn. Before we begin, however, we state from the outset that the volume and caliber of responses are indicative of the willingness of this journal, and the field it represents, to remain open to critique and self-reflection. In that spirit, we thank each of the five contributors for taking the time and effort to engage with our thoughts on this topic. Certainly, the discussion has helped us sharpen our thinking. Critique 1: Embracing impressionism is dangerous. This is perhaps the most obvious critique of our argument. For Wagenaar (2015) and Dubois (2015) in particular, embracing impressionism risks undermining interpretivism as a robust approach to social scientific inquiry. Gaining credibility in the eyes of this mainstream audience has been a long and hard fought battle against labels like ‘impressionistic’. This concern, we maintain, rests on the oppositional identity that pervades much interpretive policy analysis, whereby interpretivists continue to define themselves in relation to their dominant, positivist ‘other’. We understand why the scholars who fought that fight might be hostile to our argument. But we remain skeptical that continuing to reproduce an oppositional counter-identity is always useful or appropriate (see Boswell and Corbett 2014). Moreover, as Dubois hints, clinging tightly to this oppositional identity risks muting methodological reflexivity and internal critique—two of the very qualities that an interpretive orientation is supposed to engender. So, by employing the impressionistic label, we sought to 2 distance ourselves from an unhelpful opposition between interpretivism and its ‘other’. On this point, Mandel and Barbosa dos Santos Raxlen’s (2015) critique of our account as, in fact, positivism-affirming was perhaps the most surprising but also the most thought-provoking. Our ‘confessional’ tone, for them, simply reinforces the deviant nature of our work and thus reproduces positivism as the orthodoxy. While clearly not our intention, we concede that their reading of our argument has some merit. This poses a broader question for interpretive researchers working in political science about how we construct the genealogy of our approach; must we always adopt the caricature of ‘resistance’? As Dubois notes, our move to invert the label impressionism so that it invokes a positive identity is one way around this. But this approach clearly has limitations, too. We are open to better suggestions. Nevertheless, simply by generating further discussion we consider our intervention worthwhile, despite any dangers it may pose. Critique 2: The novelties of our impressionistic account are ‘trivial’. The second key criticism – that this is really just old wine in new bottles – seems incompatible with the first: while our claims are dangerous, they are simultaneously ‘trivial’, too. This echoes through several of the responses but it comes through loudest in Dubois’s critique. After all, impressionism is not new to the sort of organizational ethnography (see van Maanen 1988) that he has helped pioneer. Indeed, we have reflected elsewhere (Boswell and Corbett 2015b) that scholars who self-identify as anthropologists seemingly have just one qualm with our argument: that it is, in Dubois’s turn of phrase, ‘pushing at an open door’. And so, we are reminded that interpretivists already acknowledge the recursive, dynamic, open-ended, creative nature of doing interpretive research, and told that our attempt to re-label ‘systematic’ as ‘impressionistic’ is just semantics. Our retort here is a simple one, and one we hope interpretive scholars—given the rich history of scholarship in this vein on the importance of discourses, metaphors, categories—are hardly likely to object to; labels, descriptions and categories, and the meanings they infer, matter. Indeed, it seems that Dubois and Wagenaar appreciate this point acutely, or else why would they be opposed to our alternative label? For us, embracing impression was a more authentic way of representing our experience. We remain agnostic about whether researchers should wear it on their sleeves—and certainly we do not see that as universally useful—but recognising the substance is important, both as a way of framing research findings but also of passing on the craft. For us, the key strength of the interpretive method is reflexivity, and it is this commitment that underpinned our initial essay (despite Dubois’ strange assertion to the contrary). Reflexivity is contingent on a level of intellectual honesty from practitioners. The ‘systematic’ account we were given did not resonate with us, and so we could not in good conscience pretend it did. As 3 reflexive scholars we felt the need to admit our role as ‘fabricators’, to use Mandel and Barbosa dos Santos Raxlen’s phrase. No doubt this reflects our own insecurities, especially as early career scholars. But, we hoped that our experience might aid others who have a similar experience. Thinking of interpretive work as following a system or as being akin to putting together a puzzle might work for some. It didn’t work for us. We do not expect that all researchers will find our alternate analogy useful, but we will be pleased if it prompts further refinement and discussion (as in Wagenaar’s account of ‘systematic improvisation’). Critique 3: The impressionistic/systematic dichotomy is reductive. A more nuanced critique of our analogy questions the dichotomy we present between systemacity and impressionism. A common refrain on this point is our ‘convenient’ reading of the impressionist movement, but Durnova takes the substance of this argument farthest. For her, our impressionistic reflections are valuable and important, not dangerous or trivial. But she stresses that embracing impressionism need not come at the cost of rejecting systemacity, either. Researchers may develop a system of sorts to gather and record their evidence, but they cannot escape their impressions from the field; or they may seek to convey their impressions, but they cannot do so without a systematic appreciation of the contexts they operate within. Durnova, in this sense, sees impressionism and systemacity as two sides of the same coin. We think there is much of value in this critique. It is also, we think, not that far removed from what we had in mind. We think that this represents a promising way forward for rekindled debate on the meaning and value of interpretive research. We will return to the issue of ‘making an impression’ in the sections below but we cannot help but wonder whether such a debate is contingent on an initial provocative statement. Impressionism relies, after all, on the play of such bold contrasts to portray a vivid and arresting image. Critique 4: An impressionistic orientation is limiting. Marsh is least bothered by our analogy. His concern remains, however, that an impressionistic orientation still sees us wedded to a constructivist ontology. This methodological debate between interpretivists and critical realists is by now well worn, and we won’t rehash that here. The broader point to extract is that being ‘impressionistic’ seems to risk detaching the analyst from theoretical concerns that dominate mainstream disciplinary debates. This was certainly never our intention. Indeed for us, Marsh—a larger-than-life figure in the small Canberra circles where we both began learning this craft—is less an oppositional figure than a methodological inspiration. Underlying our advocacy for an impressionistic orientation is a desire to engage in the very biggest of questions and most current of debates, including or perhaps especially with those who do not share our ontological perspective (as Marsh himself so 4 often does). Interpretive researchers should feel confident—and certainly we have felt so—to draw on their data and their deep contextual knowledge to engage in important conceptual debates, even if they are somewhat removed from the issues and questions that have motivated their inquiry all along. Indeed, as we have written elsewhere (see Boswell and Corbett 2015b), it is not about having to have ‘seen [our subjects’] world’. We just need to be confident of ‘having seen some of their world’, and about being able to convey our rich, interesting and convincing impressions of that world to speak to theoretical concerns. The prevailing image of interpretive research as a systematic craft— the pursuit of the Eureka! moment where the jigsaw falls into full and final place—risks reinforcing the paralyzing perfectionism that can grip early career scholars in particular. By embracing impressionism, the standard moves from a seemingly impossible one of ‘somehow capturing’ the complexities of political life, to a more achievable one of representing a plausible, interesting and important account to the audiences who receive it. Critique 5: The reception of interpretive work remains ‘out of our hands’. In concluding our initial essay, we argued that embracing impressionism can strengthen interpretivism from within (especially for newcomers to the craft), but also enhance its reception with broader scholarly and practitioner audiences. Wagenaar argues that these purported benefits are spurious. How audiences read our work is, he says, ‘out of our hands’ anyway. On one level, we agree. We have little control over how our interpretations of interpretations will be received by readers. Certainly, we can point to instances in which others have used our empirical work in a manner that we feel misrepresents our claims. But, this is not the aspect of doing research that we were referring to by employing the label impressionistic. Rather, just as we expect that our interviewees actively anticipate how we will represent their position or experiences when they talk to us, so to do we actively anticipate how different audiences will receive our work. We spend a lot of time thinking about what our contribution to existing theory or practice will be and how our data might be best represented so as to maximize the novelty of our claims and, by extension, improve our chances of getting published in the best outlets or realizing policy impact. We carefully consider the standards and norms of different journals and their editors, anticipating reviewer comments and later revising in ways that we imagine will assuage their concerns. Likewise, we work out which practitioners will be receptive to our work and tailor it to best fit their needs and commitments. As such, while we acknowledge that we cannot influence the reception of our work post publication, we reject the idea that this does not influence how we represent our data. An impressionistic orientation recognises the subjectivity of both audience and author. This might, as we suggested in the initial essay, encourage researchers to interact with their audiences in innovative ways that 5 invite them into the process of interpretation. But on further reflection we must concede that—if, for example, the researcher wants to speak to mainstream political scientists or policy scholars—such an orientation might also lead them (and has led us at times) to suppress all traces of subjectivity, too. In any case, reception may be ‘out of our hands’, but it is never far from our thoughts. After all, the whole point of developing and publishing research findings is to ‘make an impression’. References Boswell, J and Corbett, J 2014, 'An Antipodean history of interpretation', Australian Journal of Public Administration, vol. 73, no. 3, pp. 296-306. Boswell, J. and Corbett, J. 2015, ‘Embracing Impressionism: Revealing the Brushstrokes of Interpretive Research’ Critical Policy Studies, vol. 9, no. 2, pp. 216-225. Boswell, J. and Corbett, J. 2015, ‘Who are we trying to impress? Reflections on navigating political science, ethnography and interpretation’ Journal of Organizational Ethnography, vol. 3, no. 2: XX-XX. DOI 10.1108/JOE-08-2014- 0030. Dubois, V. 2015, ‘Embracing impressionism: Giving the game away, or pushing at an open door?’ Critical Policy Studies, Vol. 9, no. 2, pp. 237-240. Durnova, A. 2015, ‘Between the individual and the collective: understanding the tensions of the interpretive inquiry. A response to Boswell and Corbett’ Critical Policy Studies, Vol. 9, no. 2, pp. 241-245. Mandell, A. and Barbosa dos Santos Raxlen, J. 2015, ‘On Boswell and Corbett’ Critical Policy Studies, Vol. 9, no. 2, pp. 234-236. Marsh, D. 2015, ‘Two cheers for Boswell and Corbett’ Critical Policy Studies, vol. 9, no. 2, pp. 230-233. van Maanen, J. 1988, Tales from the field: on writing ethnography, University of Chicago Press, Chicago. Wagenaar, H. 2015, ‘Systematic improvisation. Some misunderstandings in Boswell and Corbett’s ‘Embracing impressionism’’ Critical Policy Studies, vol. 9, no. 2, pp. 226-229. Critique 1: Embracing impressionism is dangerous. Critique 2: The novelties of our impressionistic account are ‘trivial’. Critique 3: The impressionistic/systematic dichotomy is reductive. Critique 4: An impressionistic orientation is limiting. Critique 5: The reception of interpretive work remains ‘out of our hands’. References