Writing research papers (Part 3)

Part 3: Authority, representation and responsibility

Rhetorical and literary devices are not only used to entice, engage and persuade readers of qualitative research reports, they are also employed to give them authority. Social research has always suffered of certain angst towards its own status as science and researchers have used such language devices to create texts that sound factual, objective and scientifically authoritative. Atkinson (1990, p.36) argues that textual literary conventions are greatly responsible for readers interpreting texts as factual, to persuade readers that what they are reading is the reality straightforwardly reported by the author/researcher. It follows that, from this standpoint, both text and researchers are merely, neutral conveyors of the ‘real’ world discovered/uncovered by research. Furthermore, it gives the author/researcher the authority of the one who ‘has been there’ and seen ‘how things are’. The researcher’s authority has been one of the premises upon which Western anthropology and ethnography have been built. It would be a gross oversimplification to say that early anthropologists and ethnographers had no awareness of how problematic representing others is, but later 20th century post-structuralist philosophy and calls for greater reflexivity have certainly contributed to the current view that authority in social research writing is a contested territory. Then the so-called ‘crisis of representation’ in social research. The single, privileged authoritative voice of the researcher is now under scrutiny and a call for multiple perspectives, the inclusion of the voices of those observed/researched to be included in research reports comes from those advocating for more inclusive, alternative forms of representation. As Clifford (1986) argues,

The critique of colonialism in the postwar period – an undermining of ‘The West’s’ ability to represent other societies – has been reinforced by an important process of theorizing about the limits of representation itself. (…) what is at stake, but not always recognized, is an ongoing critique of the West’s most confident, characteristic discourses. (1986, p.10)

This critique of discourse forms has lead some to advocate for novel forms of textual representation in ethnographic writing where the voice and authority of the author/researcher is somehow undermined, or at least brought to questioning, by the inclusion of other voices. New written arrangements vary from more substantial inclusion of extracts with participants’ voices in the weaving of the text to more radical text forms such as poems, drama and fictionalized narratives. While recognizing that novel textual forms of ethnographic social research representation may give a positive contribution to the field, Walford (2009) argues that they should ultimately be based on observation, field notes and research evidence. Moreover, there should be serious a commitment from the part of the ethnographer to reduce ambiguity as much as possible. For him,

Good literature, drama or poetry should certainly try to encourage ‘connection, empathy, and solidarity’. Good literature is centrally a writerly text with multiple possible interpretations designed to engage the reader in a reflexive process of new understanding. However, the reports of ethnographic research (and, indeed, all research) are surely fundamentally attempts to construct a readerly text _ one where the attempt is made to restrict multiple meanings as far as possible. For me, ethnographic reports need to be logically constructed and be clear about what empirical claims (factual and explanatory) are being made and what empirical data have been generated that support those claims. (2009: 227)

Hammersley and Atkinson (2007, p.204) had already called for caution in producing overtly fictionalized accounts of research and argued for a ‘proper balance between a totally impersonal style that elides the agency of the observer-author and an ‘exaggeratedly’ literary form in which the author seems more important than the rest of the social world.’ Such balance is not just a matter of sensibly employing rhetorical devices and opting for the most suitable textual genre, but also a matter of balancing the voices of the researcher and the voices of those represented in the text and being as faithful to the data collected from participants as possible. It is a matter of recognizing that research writing and the authorial power of representation are intrinsically interconnected and that the researcher should reflect on such issues and make them as explicit as possible when producing their research accounts. Ultimately, it is a matter of academic responsibility and an ethical stance since it is the researcher’s responsibility to provide a clear enough account of their study to allow their readers to make an informed decision on how accurate, valid, reliable and representative a research report is.


In this series of posts I have briefly considered how understandings of the nature of language also affect the way we perceive the nature of qualitative research reports. It seems to me that a responsible attitude towards the research participants, the academic community, the readers of social research texts in general and those whose lives maybe affected by our research findings is to commit to a textual representation in which arguments are based, as much as possible, on methodological rigour and appropriate evidence. As complex and debatable as notions of accuracy, validity and reliability may be, as multifarious and shifting truth may be, somewhere there is thin line between literary writing making use of research findings and research writing making use of fictional elements. Post-structuralism may have rendered the line between fiction and reality epidemiologically nonexistent, but ethically it still stands (Kearney, 1988, p.361). When deciding between writing research texts as a social scientist or as a poet we have to make the decision based on our understanding of our responsibility towards the others involved in our research projects and the ones who will be affected by it. It may not be an epistemological decision for us to make, but it is certainly an ethical one.


  • Atkinson, P., 1990. The Ethnographic Imagination. London: Routledge.
  • Hammersley, M. and Atkinson P., 2007. Ethnography. Principles in Practice. 3rd Edition. London: Routledge.
  • Kearney, R., 1988. The Wake of Imagination. London: Hutchinson.
  • Walford, G., 2009. For ethnography. Ethnography and Education, 4(3), pp.271-282.