The PhD project A Pedagogy of Machines addresses the key role played by digital tools in shaping and framing our knowledge system and the practices of learning and teaching. The debate on the tools for education has a long history that precedes the rise of the so-called “platform education”1. The research builds upon tracing an “alternative” history of Distance Education, one which does not necessarily depart from the recruitment strategy and economic performance promised (and unfulfilled) by Silicon Valley and Massive Open Online Courses. Conventionally, this history starts off with the success in attendances by the three 2011 Stanford University MOOCs on machine learning, databases, and artificial intelligence2 . A rendering of distant learning through the lens of labour history – a subset of the larger field of Social History of Education – shall help to shed new light on today’s growing reliance of universities’ curricula over taxonomies and data-drive ontologies intended to measure in real-time the fluctuation of the labour market3.

Guiding questions

  1. Which epistemological assumptions the concept of “interpretability” entails?
  2. For whom is interpretability intended?
  3. What pedagogical aspects are embedded in the AI Forensics toolkit?

Current transformations in Higher Education have remarked how the process of “catching up” and “not to be left behind” with a model – mostly the one of American university – is still a powerful force in education policy across regional and national borders4. It is also important to recognize how the university in Europe – and even more in the US – has long survived on the appropriation and incorporation of “externalities” from foreign countries – mostly those defined as “anthropological areas”5. For Wendy Hui Kyong Chun, in the HE sector, the utopian dream surrounding computer technology is fundamentally disembodying: “it assumes the material presence of the teacher immaterial”6. This is a discourse based on the fantasy of a direct “flow of information from material to student that automatically posits the teacher as an impediment to learning”7.

Today’s debate on machine learning revolves around the normative premise that full automation will be achieved when the machine learns “by itself” in a fundamental asociality considered in contradiction with the context of human learning. Revisiting Lev Vygotsky’s theory of concept development8, Reigeluth and Castelle9 stress how in the future it will become increasingly common to understand that human learning and machine learning co-evolve and therefore a “social theory of machine learning” will become necessary. Contrary to the assumption that a learning process can be best understood in isolation, which is the capacity of an agent to solve a problem without the help of peers of technical mediations, a social theory of machine learning would argue for the mutual acknowledgment in the development of mental and technical models10.

Themes and material of investigations

  1. Machine translation as a technology of bordering
  2. A Labour History of Distant Learning
  3. The Digital Walls of the University
  4. Universities in Translation and Technology Transfer
  5. What Kind of Learning after Machine Learning?

In order to formulate teaching and learning machines able to respond to the complexities and the challenges of the present, we must rethink how we translate technology across social, economical and cultural fields. The project’s research hypotheses intend to reverse the assumption that “AI is transforming education” and to investigate how “education transforms AI”. This is precisely the attempt to reframe the impact of machine learning on education and the urgency to understand how machines learn into a wider social framework committed to articulate a pedagogy of machines.

References

  1. B. Williamson, Making markets through digital platforms: Pearson, edu-business, and the (e)valuation of higher education, Critical Studies in Education, vol. 62, no. 1, pp. 50–66, 2021. doi:10.1080/17508487.2020.1737556
  2. F. Hollands and D. Tirthali, MOOCs: Expectations and reality. Center for Benefit-Cost Studies of Education, Columbia University, 2014.
  3. R. Rentzsch and M. Staneva, Skills-Matching and Skills Intelligence through curated and data-driven ontologies, In Proc. Proceedings of the DELFI Workshops 2020, 2020
  4. B. de Bary, Eds., Universities in Translation: The Mental Labor of Globalization. Hong Kong University Press, 2010. [Online]. Available: https://press.uchicago.edu/ucp/books/book/distributed/U/bo37851556.html [Accessed: Dec. 17, 2024].
  5. N. Sakai, Translation and subjectivity: on “Japan” and cultural nationalism. Minneapolis: University of Minnesota Press, 1999.
  6. W. Chun, Imagined networks: Race, digital media and the university, in Universities in translation: The mental labor of globalization, B. Bary, Eds. China: Hong Kong University Press, 2010, p. 341–354.
  7. E. Mcwilliam and P. Taylor, Teacher Im/Material: Challenging the New Pedagogies of Instructional Design, Educational Researcher, vol. 27, no. 8, pp. 29–35, 1998. doi:10.3102/0013189X027008029
  8. L. Vygotsky, Thinking and speech, in The Collected Works of L. S. Vygotsky: Problems of General Psychology, R. Rieber and A. Carton, Eds. New York: Plenum, 1987, vol. I.
  9. T. Reigeluth and M. Castelle, What kind of learning is machine learning?, J. Roberge and M. Castelle, Eds. Cham, Switzerland: Palgrave Macmillan, 2020, pp. 79–115.[Online]. Available: https://doi.org/10.1007/978-3-030-56286-1_3 [Accessed: Dec. 13, 2024].
  10. P. Damerow, Abstraction and Representation. Dordrecht: Springer Netherlands, 1996. [Online]. Available: http://link.springer.com/10.1007/978-94-015-8624-5 [Accessed: Dec. 17, 2024].