Description
In this panel, we aim to explore how some corpus research can lead to dead ends, result in detours, and ultimately end in disappointment. Gloomy though this may sound, the spirit in which this panel is planned is a positive and encouraging one. For as many experienced researchers do not hesitate to admit, ‘failure’ not only comes with the territory, but usually tells us a great deal about a lot of things, including the nature of the data we are dealing with, our methods, and foundational questions about epistemology.The great majority of published work glosses over these stressors, however. The publishing process means that when we write up projects for publication, we construe the research journey in such a way that it appears entirely straightforward – as if the results simply jumped from the data. This is as much for our benefit as for our readers’, as the research story must be as easy as possible to understand. Yet by simply ignoring complexity, we may be doing ourselves and our field a disservice. Junior scholars may be fooled into thinking that the process is mechanical, and feel discouraged when their data doesn’t play ball. What is more, they may quite rightly feel that being open about failure, and dealing with it, is a privilege that is unique to tenured faculty. Yet the latter too are subject to institutional and reputational pressures.
In this panel, we invite four established researchers to draw upon their own experiences and reflect upon projects of theirs that struggled to make progress. First, Paul Baker will discuss issues that he has personally encountered: projects which had to change direction half-way through; data which turned out to be faulty and threw up strange results; cases where the research simply resulted in uninteresting results; and challenges with applying old methods to new questions. He will offer some practical guidance on dealing with such dead-ends which might, in some cases, simply be to let go altogether. Charlotte Taylor will similarly draw on her personal research experience to discuss the ‘detours’ of research journeys. This is a space to think about the indirect progress of research projects, which is often in sharp contrast to the linear image given in published pieces coming out of them. Afterwards, Clyde Ancarno will focus on issues encountered in her collaborative research combining corpus and anthropological methods. She will also provide preliminary insights into specific problems encountered in the study of African languages, caused by, for example, the prevalence of code-switching, and inconsistencies in the way words/phrases are truncated and spelled. Finally, Susan Hunston will discuss the essentially comparative nature of most corpus linguistic work: the issues with establishing ‘baseline’ corpora, and the problem that researchers are faced with when they simply don’t have two corpora available to compare.
We hope that the panel will kick-start a renewed sense of transparency and honesty in the field (ironically, an honesty that is often found in PhD theses, but not published work). As Baker (2018: 289) writes, ‘nobody talks about “failed corpus research” […] I wonder if we could collectively learn as much from our mistakes as our successes and perhaps we should worry less about our images.’
| Period | 3 Jul 2023 |
|---|---|
| Event title | Corpus Linguistics 2023 |
| Event type | Conference |
| Location | Lancaster, United KingdomShow on map |
| Degree of Recognition | International |