Rethinking “Inferred” Thought Presentation: some reflections on Palmer's “Social Minds”
Elena Semino · Style · 2011
Reading Alan Palmer's target essay on Social Minds made me think again, and in a new way, about ah aspect of my earlier work that I had not reconsidered for some time. Between the mid-1990s and the m id-2000s, I worked with Mick Short and several other colleagues on a project that involved analyzing a quarter-of-a-million-word corpus of fictional and non-fictional narratives for what we called writing and thought (see Semino and Short). Our data included 120 2,000-word extracts from 20th century British fiction, (auto)biographies, and news reports. We inserted tags for different forms of speech, writing, and thought presentation in electronic versions of our data, so that we could carry out quantitative comparative analyses within and across the different genres. While annotating the corpus for thought presentation in particular, we felt that we had to make a distinction that we had not previously anticipated we would need. Our approach to thought presentation, which, l admit, largely took what Palmer ah internalist perspective, was prototypically concerned with stretches of text in which the narrator/reporter presents thoughts or internal states that they know about by having direct to the mind of the relevant story participant: from our point of view, this applied to omniscient third-person narrators and to first-person narrators presenting their own thoughts as a character in their own story (although we recognized the differences in the nature and reliability of access to mental states between these two situations; see Cohn 144). We therefore had to pause when we noticed not only that first-person fictional narrators routinely talk about the contents of the minds of other characters, but also that narrators/ reporters in news reports, biographies and autobiographies do the same thing with the participants in their own stories. The following, for example, is ah extract from a biography of W. H. Auden by Humphrey Carpenter: His predecessor in the job, Day-Lewis, had found his Glaswegian accents difficult to get used to, and presumably Auden experienced the same problem. But he discovered he had a real liking for the business of teaching. He enjoyed the boys' company (Twelve year old boys are the best people to talk to,' he remarked to a friend) and be was sympathetic to the rebels in the school, having been rebellious in his own schooldays. Above all he liked influencing his pupils and telling them what they should do with their lives. The job of teacher, in fact, gave a proper outlet to two dominant aspects of his personality: the dogmatic schoolmaster and the would-be healer. The boys, on the other hand. did not find him the ideal teacher. (Carpenter 193: 111-12) In this extract, Carpenter talks about the internal states of Auden's predecessor in the job, Auden himself, and Auden's pupils, in a way that is typically associated with omniscient third-person narrators, even though he did not have what my colleagues and I called direct to the relevant minds. In the course of our project, we used the label thought presentation for this phenomenon and added ah additional element to our annotation system in order to be able to study the characteristics and distribution of the inferred' instances of our thought presentation categories. Having read Palmer's groundbreaking work, I now realize that we were trying to deal with the textual evidence of different forms of what Palmer calls intermentality I should clarify from the outset that our notion of inferred thought only overlaps in part with Palmer's intermentality. However, the areas of overlap are interesting, particularly because two thirds of our corpus consisted of non-fictional narratives. In the final sentence of the extract above, a particular attitude is attributed to Auden's pupils, who can therefore be described as a medium-sized intermental unit, in Palmer's terms. …