Pedagogy and technology shape each other

Tim Fawns, writing in Postdigital Science and Education, argues that the relationship between pedagogy and technology is not sequential at all. The starting point for his thinking is a critique of two positions that dominate educational discourse. The first is technological determinism β€” the idea that technology drives change, that outcomes follow more or less inevitably from the tool. The second, and more common in educator culture, is pedagogical determinism β€” the idea that the teacher is fully in control, that technology is neutral, and that good pedagogy applied well will produce the intended results. Fawns (2022) argues that both are illusions, because both assume a level of independence between elements that simply does not exist in practice. That is precisely what draws me to this body of work: it moves past the binary ways of thinking, and in doing so opens up a much more nuanced and honest conversation.

In his model of entangled pedagogy, technology, methods, purposes, values and context are continuously shaping one another (Fawns, 2022). Pedagogy is not a plan that precedes the technology β€” it is the whole unfolding situation, including the tools, the people, the institution and the specific moment in which teaching takes place. Outcomes are therefore not determined by any single element, but emerge from the relations between all of them. This means that teachers hold real but only partial agency. As Fawns (2022) puts it, they may lead the choreography, but they have limited control over how the dance plays out.

A similar logic might be found in self-regulated learning. Goal-setting, monitoring progress and adjusting strategies do not just happen in a fixed sequence nor do they happen in isolation from the context around them β€” they inform each other continuously, in response to what is actually happening. You do not simply plan, then act, then reflect and are done with it. The same fundamental iterative, mutually shaping dynamic is how I understand what Fawns argues applies to pedagogy and technology.

This sits within a broader postdigital perspective, which takes seriously the idea that the digital, material and social are already woven together in all educational situations (Fawns, 2019; Jandrić et al., 2018). The postdigital is a refusal to treat digital tools as a separate layer that can be added to or removed from an otherwise intact pedagogical situation. Every classroom, every course design, every moment of learning is already a combination of the physical, the social and the technological. Recognising this is simply a more accurate starting point (Fawns, 2023).

That all has implications for how we think about design in practice. Adam Matthews (2019), drawing on Actor-Network Theory, argues that educational designers are better understood as bricoleurs β€” people who work with and respond to the full range of materials, tools and relationships available in a given situation, rather than applying a predetermined design from above. The design that results is always an assemblage, shaped as much by what is at hand as by what was originally intended. This is not some pessimistic view of design β€” it is a realistic and holistic one, and it follows naturally from taking the postdigital entanglement seriously.

Question also raise about language. Fawns (2019) makes a nuance worth holding onto here. Terms like “educational technology” or “digital education” are not necessarily wrong to use. When a new tool genuinely demands closer attention β€” when its novelty prompts educators to rethink their design, question their assumptions, or look more carefully at what they are trying to achieve β€” naming it separately can be useful. The distinction earns its place when it opens up inquiry. The problem arises when it closes inquiry down instead: when “digital education” becomes a brand, when “technology” becomes a category that is either celebrated or feared as a thing apart from teaching, or when naming something as an “ed-tech solution” substitutes for thinking carefully about the actual situation. In other words, the label is a tool too β€” and it is subject to the same entanglement logic as everything else.

What generative AI makes visible

There is a simple analogy I use to try make all this tangible, if not just for myself and although lacking most of the intricacies. Yet, I find this analogy helps at least start the discussion. If you want to build a chair, you can design it first and then pick the tools to realise that design. That is the sequential logic β€” and it works well enough for a chair. But imagine you have access to a CNC machine. The precision and repeatability it offers might lead you to design a modular, flat-pack chair that could never have come from a handsaw and a workbench. And then something else happens: the CNC’s capacity to produce identical parts at scale makes you realise you could design a whole furniture system, not just a chair. The tool changed the design. The design changed the goal. What started as “make a chair” became something else entirely β€” not because the original purpose was wrong, but because engaging with the technology revealed new possibilities that were not visible from the planning stage. The tool did not follow the design. It shaped what became possible to think in the first place.

Generative AI in higher education works in much the same way. A course team designs a writing module around developing critical thinking through iterative drafting. The pedagogical intent is clear. Yet once students begin working with generative AI tools, the act of drafting might start to mean something different. Generating ideas and evaluating AI-suggested text are not the same cognitive process, even when they look similar on the surface. Detection tools introduce a layer of suspicion that reshapes trust relationships between students and teachers. Unequal access to more capable models creates new equity gaps. And policy ambiguity leaves many teachers navigating the distance between what they value and what the institution has actually decided.

The technology did not wait for the pedagogy to settle. It entered the situation and began shaping it β€” the trust relationships, the equity dynamics, the meaning of the task itself. And it is not only about generative AI as a writing tool. The rapid emergence of AI tutors and AI agents capable of guiding students through problems, providing feedback, or completing work on behalf of students altogether introduces entanglements that go far beyond any initial pedagogical intent. Whether those tools support genuine learning or subtly replace the productive struggle that makes learning stick is not a question that can be answered by the pedagogy alone β€” it depends on how all the elements interact in a specific context. The OECD’s Digital Education Outlook 2026 confirms this is not a local problem: generative AI is already operating across very different educational scenarios, each producing its own configuration of purposes, constraints and unintended consequences (OECD, 2026). The entangled view helps explain why β€” because in each of those scenarios, the technology is not sitting passively behind a pedagogical decision. It is part of how that decision unfolds.

A more productive question

The entangled view does not offer a cleaner answer to the original question. What it offers instead is perhaps a better one: rather than asking what comes first, it asks what everything is doing to everything else β€” and keeps asking as the situation unfolds. That ongoing attentiveness to how purposes, values, context, method and technology are all in motion at once feels like the more productive place to work from.

References and Further Reading

  • Fawns, T. (2019). Postdigital education in design and practice. Postdigital Science and Education, 1(1), 132–145. https://doi.org/10.1007/s42438-018-0021-8
  • Fawns, T. (2022). An entangled pedagogy: Looking beyond the pedagogy–technology dichotomy. Postdigital Science and Education, 4(3), 711–728. https://doi.org/10.1007/s42438-022-00302-7
  • Fawns, T. (2023). Postdigital education. In P. JandriΔ‡ (Ed.), Encyclopedia of Postdigital Science and Education (pp. 1–11). Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-35469-4_52-1
  • JandriΔ‡, P., Knox, J., Besley, T., Ryberg, T., Suoranta, J., & Hayes, S. (2018). Postdigital science and education. Educational Philosophy and Theory, 50(10), 893–899. https://doi.org/10.1080/00131857.2018.1454000
  • Matthews, A. (2019). Design as a discipline for postdigital learning and teaching: Bricolage and Actor-Network Theory. Postdigital Science and Education, 1(2), 413–426. https://doi.org/10.1007/s42438-019-00036-z
  • OECD. (2026). OECD digital education outlook 2026: Exploring effective uses of generative AI in education. OECD Publishing. https://doi.org/10.1787/062a7394-en