Collaboration Between Computer Science, Social Science and Humanities
Creating a Module on AI in Society that Allows for Collaboration
There is a lot of talk of interdisciplinary collaboration or transdisciplinarity. Therefore I thought it would be good to explore these terms. First recognising as an anthropologist that these terms are likely not as neat as they are presented. However we can look at these and how they are perceived.
There are three distinctions that are spoken of often:
- An academic discipline
Each in turn has a common definition.
(1) “An academic discipline or field of study is a branch of knowledge, taught and researched as part of higher education. A scholar’s discipline is commonly defined by the university faculties and learned societies to which they belong and the academic journals in which they publish research.”
(2) “Interdisciplinarity or interdisciplinary studies involves the combining of two or more academic disciplines into one activity. It draws knowledge from several other fields like sociology, anthropology, psychology, economics etc. It is about creating something by thinking across boundaries.”
(3) “Transdisciplinary Research is defined as research efforts conducted by investigators from different disciplines working jointly to create new conceptual, theoretical, methodological, and translational innovations that integrate and move beyond discipline-specific approaches to address a common problem.”
It is said that transdisciplinary research is place based. Different disciplines must work together to solve a specific problem. Exactly how it is multi-, inter- or transdisciplinary seems like an open question.
Kirsten Hastrup talk of cross-disciplinarity in fusing different styles of reasoning by: “integrating the points of view and unspoken certainties of their partners in the field into their analysis.”
Therefore an economist working with a computer scientist to integrate calculations on a project brief into the framework of how a project is shaped: what type of disciplinary is that?
Collaboration is important, Hastrup argues too that knowledge generated on the edge of one’s familiar disciplinary territory may both expand and intensify the anthropological field.
Hastrup calls this collaborative moments that: “…emerge through the co-presence of several analytical perspectives in the field.”
Although we conceive of these definitions at times in a structurally bounded manner it is not hard to grasp that they are more fluid. These different definitions emerge from moments.
One such moment that is often considered important is the Macy Conferences or “Cybernetics. Circular, Causal, and Feedback Mechanisms in Biological and Social Systems” (one could understand why Macy was easier). It was a collaborative moment that is attributed with a contribution to the advancement of technologies we consider important today such as the Internet and systems theory.
If we want to be operating with such collaborative moments we have to accept the differences that we may have. Samuel Gerald Collins a Professor in Anthropology writes this:
“In the ten Macy Conferences held between 1946 and 1953, many of the tools associated with AI — neural nets, von Neumann architecture, Shannon’s quantitative definition of information — coalesced into the what some have called “strong AI” including the hegemony of mechanical models of cogitation….”
As chair of this set of conferences, Warren McCulloch had responsibility to ensure that disciplinary boundaries were not unduly respected.
Perhaps a module on AI in society must be about how this respect for different discipline can be nurtured or how it may work in practice?
Practice does vary across regions, places, people and such even locally within a building. In the academic environment where I am situated it is not shameful to say that some floors operate almost in complete isolation from one another with a few exceptions.
This is not necessarily bad as working with different disciplines is not always needed or not all the time, however we may have reason to believe that developing AI requires to a high degree.
Contrary perhaps facilitating these collaborative moments could be a possibility. Then again this may have been put into systems already in a way through structured methods ‘lean’, ‘agile’ or other innovation methodology. We can picture on the other hand that we are able to move beyond this, however it is hard to conceive of how.
Should we be encouraging students to write their master’s degrees together with other disciplines, or must the student be ‘disciplined’ enough in their own discipline before they are allowed to cross boundaries? The standards by which students are judged can seem grounded in a particular field or area of study.
A module that is structured as a collaboration between computer science, social science and humanities has to be structured in a way that creates the opportunity for different collaborative moments. In a way the strength of different students as academics as well as practitioners may shine through, however it requires facilitation. It must be structured in such a way that firstly brings in different students from different backgrounds, so this must be recruited for — and as well make it so that the working topics are structured in a constructive manner. There are disciplinary boundaries that those present are situated within, but with knowledge — knowing the edge of one’s familiar disciplinary territory and crossing it.
This is #500daysofAI and you are reading article 162. I write one new article about or related to artificial intelligence every day for 500 days.