Challenges: ontology and generalizability

However, what at first glance looks like a fairly straightforward process is replete with challenges. The most fundamental is the choice of the ontological and episte- mological framework: in other words, the set of ideas used to conceptualize what exists in the world at the most basic level and how we can produce knowledge about it. This choice will not only influence the final analysis and theorization but also the way the research is carried out. In qualitative research, this usually boils down to whether the researcher chooses a naturalist/realist or interpretivist framework. The ontological issue is at the heart of this choice: is the human world constituted by structures, forces and processes external to individuals and their cultures, or simply a set of cultural meanings and practices? The former, realist position suggests that human experience is underpinned and shaped by more fundamental influences exter- nal to culture. For instance, in his study of young people’s experiences of the holiday industry in Ibiza, Briggs (2013) firmly connects what seems like compulsory exces- sive hedonism to the demands of the industry and the circulation of capital. The latter, interpretivist assumption, which suggests that culture is the bedrock of society with nothing below it, and that human experience cannot be reduced to external influences, is also common in ethnography. For instance, Katz (1988) focuses princi- pally on cultural meanings and social micro-interactions in his study of the moral and sensual attractions of doing crime. However, interpretivism tends to ignore or down- play external forces such as the natural environment, economy, politics, ideology and history (see Gilmore, 1990), or even common biological drives and conditions of insecurity and anxiety (see Hall et al., 2008). Or, as yet another alternative, were early French structuralist ethnographers correct in their assumption that meanings and practices are themselves ordered in universal structures of language external to the individual yet mirrored in internal patterns of thought (Lé vi-Strauss, 1973)? Can the researcher employ moral and epistemological certainty to make clear judgements about what they experience, or should s/he be completely flexible, celebrate moral pluralism and epistemological uncertainty, make no clear judgements at all and leave interpretation to the reader? The obvious way out of these tough choices is to take account of all these possibilities, but assembling a compound position can be diffi- cult. O’Reilly (2009) argues that a reflexive-realist position helps to overcome the rigidity of realism and the defeatism and relativism of endless interpretivism.

Another fundamental problem is the generalizability of knowledge, analysis and

theory. Can the researcher’s account of what occurs in the chosen research setting be transferred to other research settings that might be different? Do Wakeman’s (2016) ‘moral economy’ of drug users, Ellis’s (2016) violent males or Smith’s (2014) hedon- istic night-time consumers, which these researchers found in their respective research settings, also exist in the same forms in similar research settings in similar structural locations elsewhere? Williams (2000) argues for moderatum generalization, which can offer tentative generalizations that leave room for future reformulation. His solu- tion is to offer each analysis and theorization together as a tentative hypothesis to be

confirmed or modified as alternative research settings are explored, which adds a deductive element to the otherwise inductive ethnographic approach. The politics of research becomes very influential here. Ethnographies are expensive, funds are extremely limited and quantitative studies soak up most funding; therefore ethno- graphic explorations of further research settings might never materialise. In this case, early explorations, if the accounts they produce are attuned to the dominant or sub- dominant cultural and political values of their time and thus achieve initial popularity (see Hall and Winlow, 2015), might remain immune from refutation and establish themselves in the discipline as received wisdom for decades.