Week 26: John Sutton (2008) Material Agency, Skills and History: Distributed Cognition and the Archaeology of Memory

This week we were kindly joined by John Sutton who discussed his work on distributed cognition, skilled performance, and the study of memory. We used this paper as a jump off point on these topics (you can see the breadth of John’s interests and research here and here at his website).

The paper focuses on cognitive archaeology – in particular, an analysis of the work of Lambros Malafouris (whose work we encountered in weeks 21 and 22). Sutton outlines a number of intriguing examples of exograms – external memory devices – throughout history: e.g. cloth, and table-books. Crucially, what is interesting about this latter case is that rather than being permanent (a supposedly usual feature of exograms), they are erasable. And it is claimed that this led to new ways of remembering and forgetting in medieval England.

But although Sutton agrees to some extent with Malafouris’ project of trying to expand beyond the original symbolic representational static account of exograms, he thinks that Malafouris goes too far towards radical enactivism. Sutton argues that the debate about representations is a separate issue and there is no need to reject them in order to have a more full-blooded account of what he refers to as cognitive ecology: the interplay of bodily, neural, material, gestural, etc. resources in dynamically composed cognitive systems engaged in cognitive tasks.

Beyond these general points there are three features of this paper that I would like to draw special attention to because they are mainstays of Sutton’s research and they vital points: skills, history/time, and case studies.



Sutton has consistently argued that we should not see thinking and doing and that skilled action should not be seen as entirely divorced from declarative thinking (pp. 48-49). A large swathe of philosophy and cognitive science has regarded expertise in skill as divorced from declarative memory and thinking. For instance, Hubert Dreyfus’ phenomenology has been incredibly influential: it proposes that as the agent becomes more enskilled in a particular activity they move from declarative and conscious deployment of rules and norms in guiding actions to an unconscious habitual “skilled coping”.

dreyfus skill model

On this account, norms and rules are just training wheels to be discarded. Sutton has repeatedly challenged this view. As he points out, a strict division between knowing and doing does not make sense because:

“What is striking about the exercise of kinaesthetic memory in such complex acquired skills is that there is never a simple repetition, but rather a contextually appropriate distinctive felt movement dynamics “that is at once both familiar and yet quintessentially tailored kinetically to the particular situation at hand”” (p. 49. My emphasis)

As such, Sutton claims that genuine expertise in a skilled activity is not just skilled coping but also the capacity to intercede on one’s actions in a deliberative and conscious manner. Furthermore, he importantly points out that if this point is not recognised, and the division of thinking and doing is maintained, then it renders it rather curious how experts do in fact consciously influence their behaviour (p. 50; also see Sutton 2007 for a more extensive account of this in relation to batting in cricket).


Importance of history and time

One of the strongest and most important contributions of Sutton’s work is his emphasis on the importance of time for considering questions around distributed cognition (also see Biette & Sutton 2015; Williamson & Sutton 2014 for recent accounts of this in regards to the studying of memory and collaborative teams). This is at the forefront of this article which begins by stating that how cognition is distributed takes place in time: dynamically, developmentally, and historically (a point previously discussed here).

biete &sutton on time

[diagram from Bietti & Sutton 2015, p. 422]

Furthermore, Sutton goes on to also stress that distributed cognition is not an “after the fact” event, nor is it a recent event (pp. 37-38). I.e. there is no point for the human organism in which they are totally isolated from the cultural niche – it is always there for them; and in some ways is a part of them (working out how to cash this out is the tricky part). Additionally, distributed cognition is not a post-humanist position. As the focus on cognitive archaeology and paleoanthropology shows – the way in which the human mind is shaped by and shaping its environment (including tools and techniques) has been occurring for millions of years (also see week 10).

scales of dc.png


Emphasis on case studies/”no enemy of motley”

Sutton proposes that the conceptual resources for considering distributed cognition are now sufficient and he suggests that any necessary modifications are likely to arise through applying the framework to case studies rather than attempting “further theoretical radicalisation” (p. 40). Similar views have expressed by Bryce Huebner (2014) who has argued that we should focus on concrete case studies and develop rich ethnographic case studies in order to properly explore distributed cognition.

I entirely agree with this proposal. But it is noteworthy that in early work Sutton noted that Adams and Aizawa raise the plausible worry that simply cataloguing lists of interesting case studies will lead to “an unscientific motley” (2001, p. 62). In response, Sutton has stated that although he is “no enemy of motley” [a friend of mine recently told me that this is one of her favourite statements published in academic philosophy], that this is a legitimate issue. Proponents of distributed cognition cannot just offer endless examples, but must also offer a systematic approach to this state of affairs (2006, pp. 235-236).

motley and systems.png

In these regards it is noteworthy that he and his colleagues (Sutton et al 2010) have devised dimensional approaches to this problem. There have been a number of proposals in the literature that have recently been nicely taxonomised by Richard Heersmink (2015) – who has also gone on to show how the dimensional framework can make sense of cases of distributed cognition in philosophy of science (2016) and in relation to the internet (Smart, Clowes & Heersmink 2017).

However, despite all of this excellent work, I think that the questions that Sutton raises in this paper in regards to the dimensional approach are still incredibly pertinent (p. 38). I have broken down from each of these questions a number of principles or sub-issues which I think are important for proponents and researchers of distributed cognition to consider:

– “What are the synchronic and diachronic principles of coordination between diverse components?”
a. Time: how do we parse this? Is an examination of longer time scales properly possible for discreet systems? Do we have to combine both micro-processes with much longer diachronic processes? If so, what kind of framework can handle such a broad scope?
b. coordination of interactions between parts: how do we identify and properly measure and explicate the actual mechanisms between the putative parts of distributed cognitive systems as they dynamically change?

– “What are the different forms of coupling, involving distinct forms of availability and use of external resources?”
c. different kinds of coupling: what is coupling? How is designated?
d. the availability of cognitive resources: what kinds of resources are there? How do we taxonomise availability?

– “How truly interactive are particular emergent systems and how durable?”
e. rates of interaction:
f. emergence: how do we define this tricky notion?
g. durability: how durable does an integrated or distributed cognitive system have to be in order to count as one? Rob Rupert (2010) has pointed to the extreme brevity of some amalgations of brains, body, and environment as opening up a new issue in which cognitive systems are coming in to and out of existence very radidly (one could perhaps characterise this as a conceptual cousin of the cognitive bloat problem). On the other hand, Michael Anderson and colleagues (2012) have emphasised the transcience of cognitive systems as “soft-assemblages”. Additionally, Colin Allen (2017) has recently pointed out that we are interested in the cognitive processes of smaller and shorter lived organisms and even bacteria – as such, we can infer from this that the “fleetingness” of a cognitive system should not be a concern for naturalists, but is instead an interesting puzzle.


This list is an excellent – but not exhaustive – set of questions from which to begin a case study of distributed cognition. It is interesting to note that some headway has already been made in regards to some of these questions – e.g. it is almost standard now within the distributed cognition literature to use Bill Wimsatt‘s conditions of the failure of aggregativity to define emergence (see Poirier & Chicoisne 2006; Theiner et al 2010). But in many ways I think that we may have to take a dappled or archipelago approach to these cases. And treat individual case studies or cognitive domains as special cases on their own terms.






Adams, A. & Aizawa (2001) The bounds of cognition. Philosophical Psychology, 14, 43-64.

Allen, C. (2017) On(not) defining cognition. Synthese. [published online], 1-17.

Anderson, M. L., Richardson, M. J. & Chemero, A. (2012) Eroding the Boundaries of Cognition: Implications of Embodiment. Topics in Cognitive Science 4 (4), 717-730.

Bietti, L. M. & Sutton, J. (2015) Interacting to remember at multiple timescales: Coordination, collaboration, cooperation and culture in joint remembering. Interaction Studies 16 (3), 419-450.

Heersmink, R. (2015) Dimensions of integration in embedded and extended cognitive systems. Phenomenology & Cognitive Science, 14, 577–598.

Heersmink, R. (2016) The cognitive integration of scientific instruments: Information, situated cognition, and scientific practice. Phenomenology and the Cognitive Sciences 15, 517-537.

Huebner, B. (2014) Macrocognition: A Theory of Distributed Minds and Collective Intentionality. Oxford: Oxford University Press.

Poirier, P. & Chicoisne, G. (2006) A framework for thinking about distributed cognition. Pragmatics & Cognition 14(2), 215–234

Smart et al (2017) Minds Online: The Interface between Web Science, Cognitive
Science and the Philosophy of Mind. Foundations and Trends in Web Science 6(1-2), 1-232.

Sutton, J. (2006) Distributed cognition – Domains and dimensions. Pragmatics& Cognition 14(2), 235-247.

Sutton, J (2007) Batting, Habit, and Memory: The Embodied Mind and the Nature of Skill. Sport in Society 10(5), 763-786.

Sutton (2008) Material Agency, Skills and History: Distributed Cognition and the Archaeology of Memory. (pp. 37-55) in C. Knappett & L. Malafouris (eds.) Material Agency: Towards a Non-Anthropocentric Approach. London: Springer.

Sutton et al (2010) The psychology of memory, extended cognition, andsocially distributed remembering. Phenomenology and the Cognitive Sciences 9(4), 521–560.

Theiner, G., Allen, C. & Goldstone, R. L. (2010) Recognising group cognition. Cognitive Systems Research 11, 378-395.

Williamson, K. & Sutton, J. (2014) Embodied Collaboration in Small Groups. (pp. 107-133) in C. T. Wolfe (ed.) Brain Theory: Essays in Critical Neurophilosophy. Basingstoke: Palgrave Macmillan.


Extra Stuff:

Here is a lecture on transactive memory systems and distributed cognition in relation to memory from 2014.

Here is a lecture on memory studies at Warwick University from 2011.

One thought on “Week 26: John Sutton (2008) Material Agency, Skills and History: Distributed Cognition and the Archaeology of Memory

Leave a Reply

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s