Collection Date: 12th January 2009
Scholar #1 Info: (if more than one scholar's process is described, copy this set for each scholar)

• Name: anonymous
• Email:
• Title:
• Institution/Organization:
• Field of Study/Creative Endeavor: English / Linguistics

Collector Info (can be the same as "Scholar" above):

• Name: Martin Wynne
• Email: martin.wynne@oucs.ox.ac.uk
• Title:
• Institution/Organization: University of Oxford

Story for Bamboo

A researcher in an English department is studying metaphor in literary and non-literary texts. They approach the phenomenon from vaarious points of view and with various approaches. They are interested in describing the forms and functions of metaphors in texts, in identifying and classifying metaphors, and understanding what these findings can contribute to general theories of grammar, style and cognition.

In order to conduct a more well-founded and intensive empirical study, the researcher sets up an international interest group, which selects a text, each individual identifies and categorises all metaphors in the text, and then they meet together (in one physical location) approximately once per year and compare and discuss their analyses, thus testing, refining and further developing the analytic scheme, and discussing the theoretical implications of their findings, as well as their problems and differneces. They publish regularly, individually, and (more rarely) in jointly authored papers. This has taken place over the past 10 years or so, and goes on. No external funding, or computational tools were necessary for this work, although the researchers realised that online collaborative analysis tools would make more of this work possible.

In order to take things a step further, and to conduct a wider empirical study, allowing comparison of text genres, and some intimation of frequencies of metaphorical forms, the researcher decided to apply some corpus linguistic resources and methods. Significant external funding was acquired, a big team of graduate students hired, and corpus linguistics consultants co-opted. Together with the consultants, a small corpus (a subset of an existing standard reference corpus, using well-documented, standard formats) was adopted, and a formal system of tagging metaphors was devised, using well-documented standards (TEI). The consultants (including the current author) selected the corpus texts, devised the tagging formalism, adapted the corpus XML schema to incorporate the metaphor tags, indexed the corpus for an analysis tool (concordancer), and ran a training session for the project team (including IT support staff).

Outcomes:
- test and refine the analytical scheme by having to account for every occurrence in a corpus
- train young researchers and provide data and projects for PhDs
- produce reusable dataset for internal and external use (e.g. for discussion in international research group)
- analysis of annotated corpus, providing frequency profiles of metaphors in different text types
- capacity building: a team of young researchers and IT support staff who know how to work with XML, TEI, Oxygen, Xaira;
- concrete deliverables for funding bodies (the annotated corpus).