Why Scholarly Writers Should Learn to Write Dialogues

-->

When I teach qualitative research methods, I often bring in Ernest Hemingway’s short story, Hills Like White Elephants. While it has some anachronisms in its language and gender attitudes, the piece is still a brilliant example of how writing well-crafted dialogue can tell the whole story—only 257 words of the 1465 are not dialogue as Hemingway powerfully establishes a scene between a man and a woman at a train station in Ebro, Spain. Readers recognize the tensions between them and the dilemma they are facing without these ever being spelled out.

I juxtapose Hemingway’s dialogue with a page of transcript from one of my research projects. While it is clear that these are real people speaking, the students are surprised at how this ‘real’ conversation sounds less ‘real’ than Hemingway’s characters. This activity opens the discussion for what evidence transcripts might offer a research article, but it also underscores the shortcomings of such transcriptions. We consider how the representation of data might require the same kind of skill that literary writers bring to their work.

The first response that some students bring to this discussion is a concern that I am suggesting that they ‘make things up’ in their research. Such an attitude is understandable given their years of schooling where teachers may have made black and white distinctions among genres such as highlighting the ‘untruth’ of fiction and the ‘truth’ of non-fiction. But characterizing the difference in such a way does not recognize the discernment that good researchers and writers bring to their work.

Hemingway imagines the scene between the individuals in his story, but there is no doubt he knows the place he is writing about and the dynamics between individuals—all real details. A researcher studying human interactions could have noticed many of the same features that Hemingway portrayed. Where the skills of the researcher and the writer come together is in the careful noticing and choosing of details that represent their insight, details that invite the reader into the scene to recognize, understand and then respond.

In qualitative research—especially narrative research—we talk about the importance of verisimilitude—something that gives the appearance of being real. It is no coincidence that this is a literary term. Most of us know that a story is not a recounting of something just as it happened; neither is a research article. Using well-recognized and accepted methods, the writer and the researcher use a process of interpreting and constructing an artifact that represents their thinking whether that is an imaginary scene in Ebro or a reporting of an exchange in a classroom. Only when we can offer our readers the specific language that shares our thinking can they then meaningfully respond.