Taking Talk Seriously: Large-Scale Methodologies

Taking Talk Seriously:Large-Scale Methodologies Leslie Howsam (bio) The challenge for "talk studies" is the ephemerality of sources. Because speech leaves no direct trace, scholars must scramble for indirect evidence—of words spoken, their context in conversation, their tone and flavour, their very sound. For the nineteenth century, such sources mostly take the form of reports in text, whether in correspondence, memoir, biography, or official record. In the nature of things, such reports are quite ephemeral in their own right, and only rarely does an extensive diary survive, providing the rich detail of a prolonged conversation among close colleagues working, laughing, and arguing together on a joint project, as was the case with the Henry Silver diaries used by Patrick Leary for The Punch Brotherhood. As valuable as such idiosyncratic sources are, however, it's crucial that "talk studies" not depend on chance survivals. Instead, I want to make a case for bold large-scale methodologies that require collaborative teamwork, significant funding, and rigorous review. With that in mind, it might be salutary to note that "talk" is not the only impossible subject now under consideration. The Odeuropa consortium is capturing evidence of odour in historical texts. Scholars are exploring "how to make computers read, see, and organize smells," using strategies derived from artificial intelligence (AI) "to extract smell-related information from text and images and encode them in visual graphs that can allow historians to follow olfactory practices over time and space" ("AHA Activities"; van Erp et al.). Odeuropa is a massive interdisciplinary and collaborative effort organized on a very large scale. Isn't the search for evidence of people in the past talking among themselves at least as important as discovering data about their experience of smell? End Page 217 My paper for the 2022 Belcher Colloquium in Oxford mined a small private database for evidence of the role of talk in one well-documented academic project that flourished from about the 1860s through to the early twentieth century. A group of men who identified themselves as properly serious historians wanted nothing less than to see their scholarly preoccupations—about the past and how to conceptualize it—recast from a largely literary pursuit into a rigorously scientific discipline. One aspect of their ambitious venture was to convince publishers that historical writing would repay investment in one or more academic journals as well as in a range of newly researched books, from monographs to reference works to printed primary sources. The publishers resisted, knowing that this austere scholarship was unlikely to attract the kind of general reader who liked authors such as Thomas Babington Macaulay and J.R. Green for their lucid prose and strong narrative (Howsam, Past into Print 49–75). I knew about this project because the historians and publishers wrote each other dozens of letters; I could use it to contribute to the "talk" initiative because I had transcribed many of those letters and organized them into a digital database. Treating the correspondence as data for a study of Victorian talk revealed that these intellectual innovators resorted to face-to-face meetings and conversation when the subject matter was awkwardly personal or political. Sometimes, for example, one member of the group had to be disinvited so that the others could discuss how to recruit him for a post, or perhaps how to discourage his participation on grounds of class or lack of seriousness. But exciting as it was for me to search for terms like "talk," "conversation," "chat," "discussion," "lunch," and "dinner"—and then use the results to construct a narrative and an argument—I knew my database couldn't be replicated. It had been created neither collaboratively nor systematically, but merely to serve the needs of my project. Nevertheless, the exercise has proved useful for identifying and articulating the massive problems of research methodology that stand in the way of the great promise of "talk studies." Without being an expert on the methodology, or even a practitioner, I want to argue that digital methods offer the opportunity to scan, data mine, and analyze the texts of published letters, memoirs, and essays. At one level, research could focus on identifiable projects, debates, or disputes...