Working across disciplines

In this research methods conversation interdisciplinary working, in particular in academia, was discussed. The group consisted of 24 researchers from different fields, such as Sociology, Physics, Archaeology, Education, Computer Science, and others. The discussion was set off with some observations from the field of Scientific Computing, where researchers from different STEM subjects work together and effective communication is very important: How people have different names for the same thing (be it a method, or an entity in a method), or sometimes even the same name for a different thing. How they might describe and understand the same method in different ways. How different fields claim ownership of a method (the question whether this usually means that they reinvented the method in different fields, or each field branched off from the same invention, stayed unanswered). 

Across all fields present, people agreed that they can see similar things in their disciplines, and in particular that the REF can reinforce boundaries between disciplines, where there are concerns that a "percentage of pure science in our field" is required to be "valid for REF". However, as one participant pointed out, this can be less of a problem for academic units that are seen as a "field" rather than  "discipline".  But especially for young researchers it can be problematic: When appointing new academics, departments want "their" people. It is also more difficult to earn a reputation if you publish in a number of disciplines, than if you concentrate on one narrow field. Outside of academia, in contrast, there is a lot of interest in highly qualified interdisciplinary people.

Additional problems noted were finding a journal for interdisciplinary work, as well as the refereeing: Often referees will want to stress their particular field in the publication, and are less interested in the contributions in the other fields.

Examples where interdisciplinary projects worked out well had two things in common: There was sufficient money available, and they had a common problem that they wanted to solve. Rather than allowing a sense of "problem ownership", these projects started off by analysing the problem from their different angles. This approach can be very enriching for all fields involved, and all researchers can benefit from the collaboration. 

For writing grant applications, the advice was to embed the interdisciplinarity from the start, and not as an afterthought. Write the application together, and start, in the best case, with a small seedcorn project, so that people can start the communication and get to know the terminology and ways of thinking of their colleagues from different disciplines. Then, in the next step, they can apply for a big project together.

A side note at this point was there are different levels:  multidisciplinary, and crossdisciplinary, interdisciplinary, and transdisciplinary projects, with increasing level of proximity of the disciplines.

Another question raised was whether research methods need to be interdisciplinary, or only the research itself. Sometimes this needs to be the case, if the output of one method is the input of the other, one participant observed.

But how do you find the overlap between fields or disciplines? Boundaries between fields can be links, and many researchers feel they have more in common with their interdisciplinary colleagues from a "neighbouring" discipline than with many colleagues from their "home" discipline.  An interesting suggestion was to implement a tool that finds which papers are cited by papers from different disciplines. Running this over general publication databases should create an interesting graph/map of connections. Here's an example for the interdisciplinary field of cultural evolution by Youngblood and Lahti (2018).

Discussing the interdisciplinary environment at Durham University, facilities such as the Institute of Advanced Studies and the Institute for Data Science are mentioned on the plus side. Interdisciplinary study courses, that exist in varying forms for decades, are very popular and show that students want and chose interdisciplinarity (although, at this point it is noted that the real interaction between departments and disciplines is minor in these courses).  However, in the departments and schools it is still difficult to start a new interdisciplinary collaboration. Researcher have to prove themselves, fit into the boxes, and experience identity threat. 

One participant reflected on her career, and noted how she changed disciplines, but is still using methods and literature from previous disciplines she worked in. For her, interdisciplinary work was and is very valuable. A young academic, in contrast, is struggling to publish papers and wonders whether it would have been easier to publish if she had stayed in one field.

So, in summary, most of the participants were in favour of fostering interdisciplinary research but recognised the significant challenges in doing so, particularly for early career researchers.

This blog expresses the author's views and interpretation of comments made during the conversation. Any errors or omissions are the author's. 

This post was written by Marion Weinzierl, who is Research Software Engineer at Advanced Research Computing, Durham University