Week 16: Pritchard (2016) Intellectual Virtue, Extended Cognition, and the Epistemology of Education

This paper explores a potential problem that arises in the intersection of virtue epistemology and modern education which is increasingly incorporating more technologies (e.g. calculators, and internet search engines, etc.). The danger is that teaching novices to use technology to achieve certain tasks might lead to what Richard Menary refers to as “cognitive offloading” (2012). This is the offloading of cognitive work from the agent onto features of their environment. In the first two parts of the paper Duncan Pritchard outlines several core distinctions in virtue epistemology before then going on to use these to defuse this problem. In the third part of the paper Pritchard goes on to introduce “the extended mind” and how he sees this as contributing to this debate – but I am less sure that this second aspect is helpful or useful.

Virtue epistemology holds that knowledge is defined as justified true belief that is attributed to the virtue of the agent’s character. Intellectual virtues are distinguisged from cognitive abilities, and we can define these as follows:

  • Cognitive abilities are specific acquired capacities which draw on cognitive traits (e.g. perception, memory) and are deployed towards specific cognitive tasks (e.g. arithmetic)
  • Intellectual virtues are general acquired cognitive traits linked to human flourishing. They have a motivational component aimed at bringing about success – e.g. the love of truth, open-mindedness, conscientiousness, etc.

A second distinction that Pritchard makes is between knowledge and understanding:

  • Knowledge is the mere acquisition of information about an event without understanding why or how this is true. Pritchard emphasises that this equates to a dead end in epistemic terms.
  • Understanding is not just the acquisition of information but where an agent also understands how to go on and use this as the basis of attaining more knowledge and understanding.

Pritchard uses these two distinctions to posit that, from a virtue epistemology perspective, that education should be aimed at enhancing both intellectual virtues and understanding (rather than just mere knowledge and the simple acquisition of cognitive abilities). These theoretical distinctions also allow Pritchard to disarm the problem of the use of technology in education. The key point is that the educator must aim to utilise the technology to develop the student’s intellectual virtues and understanding. So for example, if an agent learns to use a calculator to do arithmetic, they must also be taught an understanding of the algorithmic procedures involved. This entails that when an agent uses a calculator to complete a sum they have an understanding of how the answer has been achieved and why it is correct. In this way the use of a calculator can be used as a basis for further epistemic achievements. As he puts it:

“So long as this technology is at the service of the development of intellectual virtue and the subject’s acquisition of understanding, then this is entirely compatible with the goals of a virtue-theoretic epistemology of education” (Pritchard 2016)

I think this approach is really useful for understanding how agents interact with cognitive artefacts. Adopting terminology from Menary, I would argue that this project contributes to a focus on the crucial developmental trajectories by which an agent learns to create, manipulate, and maintain epistemic resources (2012). Menary (2007, 2014) goes on to emphasise the trasnformatory aspect that this has for an agent’s neurocognitive profile: both the impact on their neurological archictecture, but also upon what they are able to achieve at the functional level – i.e. novel behaviours. However, it is in regards to this latter point that I think that Pritchard has overlooked an aspect of an agent’s interaction with an epistemic tool. Don Norman (1991) has pointed out with a simple example of a checklist that we must distinguish between several levels of analysis:

  • at the systems level we see a “hybrid system” (Menary 2007) composed of the agent and the artefact [I will consider the rules of composition below when I discuss the extended mind] which is able to complete the task (e.g. a memory task) with greater ease/success
  • but at the individual level it becomes obvious that the individual’s performance in the task (e.g. long term memory) is not “amplified”. But we also do not need to understand the use of the cognitive artefact as mere offloading or outsourcing of cognitive work. Instead, following Menary, it becomes clear that the manipulation of a cognitive artefact changes the task space (e.g. from long term memory to one involving perception and coordination) and thus also alters the agent’s neurocognitive profile.

A core point here is to again emphasise the developmental trajectory associated with the capacity to use this device and how it enables the agent to achieve a cognitive task with a high degree of success – and potentially achieve a task that would otherwise be infeasible (e.g. the process of enculturation). We can understand Pritchard’s introduction of virtue epistemology to add another angle on this point: that this learning process must also be aimed at enhancing the agent’s broader intellectual virtues and their understanding. This therefore combines with Menary’s (2007) notion of the “transformation thesis”.

Whilst there is much to debate on this point (e.g. about whether it might be okay to teach some skills in a merely instrumental fashion; or how this edict on the aim of education combines with the work of Boyd, Richerson & Henrich (2011), discussed in week 10, who point out that much of human success is based on agents only acquiring mere knowledge and cognitive abilities in the cultural learning of very complex opaque skills…), here I will conclude with a quick rejection of Pritchard’s adoption of extended cognition to this debate.

Pritchard and colleagues (Carter et al 2014, p. 72) have distinguished between three positions:

  • Extended cognition: cognitive processing can be suprapersonal
  • Extended mind: mental processes can be suprapersonal
  • Distributed cognition: cognitive processing can be distributed across multiple agents and artefacts

And in this paper he sticks with extended cognition which he argues for on the basis of the parity principle. This argument goes back to Clark & Chalmers’ (1998) original paper and claims that we should consider a cognitive artefact to be a component of an agent’s cognitive processing system if, were it to take place inside of the skin of the individual, then we would consider it cognitive. The claim here is based on a functional parity between external and internal representations. However, there are many issues with this – which is why I do not think it helps Pritchard’s argument – which I will now list very briefly to conclude:

  1. There is not a functional parity between external and internal representations. For instance, a checklist is not equivalent to an engram. Merlin Donald (1991) – a forerunner of the debates about the extended mind – emphasised the differences between internal and external representations when he argued for the uptake of the latter. It was this very difference in functionality and attributes that made them useful for our ancestors. John Sutton (2010) has termed this position the complementarity principle.
  2. Critiques such as Adams & Aizawa (2001) and Rob Rupert (2004) have pointed out that the parity principle mistakes a causal connection for a constitutive one. One can express this formally as follows: extended cognition amounts to claiming that the connection of X and Y entails that Y is a part of X. This is obviously a fallacy. In contrast, Menary’s (2007) approach avoids this problem because it proposes the hybrid system, which can be formalised as follows: X + Y = Z (where Z is the formation of a new unit of analysis, a hybrid system, with different cognitive properties). This approach also entails that we do not mistake the cognitive artefact as part of the agent.
  3. Adams & Aizawa (2001) have also pointed out that the parity principle is insufficient for properly bounding a cognitive system and this leads to what is called the problem of “cognitive bloat”: without clear criterion for what is and is not part of the cognitive system it balloons out dangerously so that it becomes unrefined, incoherent, and intractable. Multiple proposals have been put forward against this objection but none of these repair the damage done to the parity principle itself (e.g. Kaplan’s 2012 usage of the mutual manipulation criterion from mechanistic approaches to the philosophy of neuroscience is a good demarcation principle is notable for being both agnostic in the debate and motivated based on naturalistic grounds).

Given these issues, I think that it is better to think of how we use cognitive artefacts in terms of enculturated cognition (Menary 2012, 2014; also see Hutchins 2011). This is especially pertinent here because this position places a large emphasis on learning – which I think then becomes a more natural fit for Pritchard’s position. However, to make such an argument would require a much more extended engagement with extended epistemology.







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

Carter et al 2014 Varieties of externalism. Philosophical Issues 24, 63–109.

Clark & Chalmers 1998 The extended mind. Analysis 58: 7-19.

Donald 1991 The Origins of the Modern Mind. Harvard University Press.

Hutchins 2011 Enculturating the Supersized Mind. Philosophical Studies 152, 437–446.

Kaplan 2012 How to demarcate the boundaries of cognition. Biology & Philosophy 27 (4), 545–570.

Menary 2007 Cognitive integration: Mind and cognition unbounded. Basingstoke, NY: Palgrave Macmillan.

Menary 2012 Cognitive practices and cognitive character. Philosophical Explorations 15 (2), 147-162.

Menary 2014 Neural Plasticity, Neuronal Recycling and Niche Construction. Mind and Language 29 (3), 286-303.

Norman 1991 Cognitive Artifacts. (pp. 17-38) in Carroll (Ed.) Designing Interaction. Cambridge: Cambridge University Press.

Pritchard 2016 Intellectual Virtue, Extended Cognition, and the Epistemology of Education. In J. Baehr (ed.) Intellectual Virtues and Education: Essays In Applied Virtue Epistemology. London: Routledge.

Rupert 2004 Challenges to the Hypothesis of Extended Cognition. The Journal of Philosophy 101 (8), 389-428.

Sutton 2010 Exograms and Interdisciplinarity: History, the Extended Mind, and the Civilizing Process. In R. Menary (Ed) The Extended Mind. Cambridge, MA: MIT Press


Additional links:

Here is a video of Pritchard giving an outline of the extended epistemology project.

An interesting study by Mueller and Oppenheimer (2014) claims that using different kinds of epsitemic tools can impact one’s performance in regards to note taking in lectures – pen and paper are superior to laptops.


One thought on “Week 16: Pritchard (2016) Intellectual Virtue, Extended Cognition, and the Epistemology of Education

  1. Nice summary, Alex. I really like Duncan’s distinction between cognitive abilities, cognitive traits, and intellectual virtues. I think that provides a nice way of divvying up some of what’s at play. I mentioned in the discussion that I had some concerns about the potential implications for social intelligence though. It may be true that scaffolding education with technology creates competent learners who can utilise and cognitively integrate with technology in useful ways, but there’s perhaps a cost here. It might de-emphasise the socially constructed nature of knowledge, and knowledge as a praxis embedded within communities. So much of our knowledge is contested and negotiated, not simply the result of individual learning. I feel like this has to be included in any kind of account that emphasises the intellectual virtues of understanding, as opposed to mere knowledge.

    I also mentioned there was some research that suggested increased screen time in schools may be impacting children’s emotional intelligence. I’ll link the study below. I think there’s perhaps some issues with it, but it provides some food for thought.


    Liked by 1 person

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s