Internalisation: what exactly is it good for?


Here’s the thing. We need to know what happens inside people’s heads. Damage our brains and this will almost certainly affect our mental and bodily health. Yet looking inside people’s heads won’t tell us all we want to know about the mind. Crucially, in order to understand what we mean when we talk about mentality or cognition, we need to instead examine the sorts of things people say and do and the types of practices and contexts that shape people’s behaviours. Calling yourself an enactivist means signing up to this idea.

I’m just back from an excellent workshop on enactivism organized by Fred Muller at Erasmus University Rotterdam. Caterina Dutilh-Novaes (Groningen University) and Karim Zahidi (University of Antwerp) both gave excellent presentations which, in very different ways, raised the vexed issue of internalization, the idea that we internalize what is initially an external phenomenon.

For example, Caterina ( defended her view that mathematical proof is a dialogical notion. The point of a proof, she claimed, is explanatory persuasion. Proofs can thus be characterized as having two participants with opposing goals: on the one hand, the prover, whose goal is to establish a conclusion, and on the other hand, the skeptic, whose goal (predictably enough) is to block that conclusion. Yet why don’t proofs look like dialogues? This is because the job of the skeptic has become part of the method of writing a proof. I understood this to mean that, according to Caterina, the role of the skeptic has been internalized during the formulating and writing of the proof.

In his talk, Karim ( defended an enactive account of numerical concepts. First, he laid out his view that concepts are particular types of abilities. Second, he challenged, among other things, the claim that a great variety of animal species have numerical abilities and numerical concepts. For example, he distinguished number sensitivity from number concept possession, and argued that while many animals certainly have the former, it is much less clear who, apart from humans, actually have the latter. He also offered a direct perception account of the ability to sum sequences of stimuli across different modalities, as well as describing how one could begin to give a natural history of number concepts.

Karim also discussed calculation. He argued that calculation is an activity constituted by the manipulation of public representations (symbols on a page, for example). Mental calculation, on the other hand, is what occurs when we leave these public representations out. So even if something is in fact internalized when we mentally calculate, this something is not representational. A question which then came up in discussion was: what then is internalized?

As a card carrying enactivist, my own view is that talk of internalization, no matter from what quarter, is deeply problematic. But it occurs to me (as it has occurred to others) that ‘going wide’ about mentality will always meet resistance as long as it is thought that some mental phenomena just have to be internal. Mental calculation looks to be a prime example of this. Later Wittgenstein however can help enactivists fight this resistance.

Wittgenstein cambridge
Your man himself, as they say in Northern Ireland: Ludwig Wittgenstein.

Consider one of Wittgenstein’s many thought experiments:

“Let us imagine a god creating a country instantaneously in the middle of the wilderness, which exists for two minutes and is an exact reproduction of a part of England, with everything going on there in two minutes. Just like those in England, the people are pursuing a variety of occupations. Children are in school. Some people are doing mathematics. Now let us contemplate the activity of some humans during these two minutes. One is doing exactly what a mathematician in England is doing, who is just doing a calculation. – Ought we to say that this two-minute-man is calculating? Could we for example not imagine a past and a continuation of these two minutes, which would make us call the processes something quite different?” (Remarks on the Foundations of Mathematics, VI -34)

Klagge’s (1995) description of Wittgenstein is helpful here:

“By pressing the question of what various human activities consist in, Wittgenstein hopes to demystify the mental – not by denying its existence, but by diagnosing and transcending our conception of it as an invisible reservoir. Instead of looking within the person, at the moment, for (the essence of) what constitutes, e.g. intending, expecting, or reading, we should concentrate on what leads to, surrounds, and follows from the experiences and movements with which we usually associate the activity.” (p472)

The idea then is this: when we wonder what it is that makes a given activity calculating, we encounter the primitive notion that there must be something occurring now (usually in our heads) that somehow infuses that activity with ‘calculating qualities’. Yet as Wittgenstein’s two-minute-man illustrates, strip away all context, past and future, and it becomes evident that nothing happening now (in the head, in the body, or even in the environment) makes an activity an example of calculation.

Later Wittgenstein also discusses at length the differences between calculation and mental calculation. For Wittgenstein, calculating with a pen-and-paper and mental calculation are distinguished by the different things people do and say. This of course renders the difference between these two activities one of behaviour. The twist Wittgenstein adds is that the concepts ‘mental calculation’ or ‘calculating in the head’, like many other concepts, are not about behaviour. As is well known, Wittgenstein understands the link between some concepts and behaviours as logical or grammatical. One way to understand this is to say that some concepts characterize behaviour in the sense of situating that behaviour within a particular historical socio-cultural practice. Concepts, moreover, can have primary and secondary uses. The secondary use shares important resemblances (‘looks’) with the primary use. Fogelin (1977) understands Wittgenstein to be saying that this is true of calculation, such that ‘mental calculation’ or ‘calculating in the head’ are secondary uses of the primary concept of calculation.

Together, these points – anti-presentism, the logico-grammatical link between concepts and behaviour, primary and secondary uses of concepts – provide a means to deny that mental calculation is ever internal. So, for example, when I am asked to compute a sum in my head, and in response, I furrow my brow, look pensive and give an answer, nothing that has happened inside my head at that moment explains what I just did. This is not to deny that changes may have occurred in my brain (or in my body) at that moment. What it is to deny however is that such changes capture why what I just did can be called calculating. Rather my action counts as a case of mental calculation because the secondary use of the concept of calculation is applicable to it. This secondary use characterizes my action, not through the identification of some narrow characteristic happening in me or near me at that moment, but by situating my action within the appropriate set of broader practices and techniques. This is a robustly naturalist account – it is firmly rooted in what we do and say when we calculate.

Of course, this is only the beginnings of such an account. Much more still needs to be said. Still, it confirms the merit in asking: internalisation, what exactly is it good for? The answer, I would suggest, is not much. Which raises the further question: how far can we apply this anti-internalisation agenda? To all mental phenomena? What about, say, dreams? I suspect that killing cognitivism once and for all is going to require us enactivists getting really radical…..


Removing the mind from the head. Part 1.

[The following post is the first in a series which will describe my philosophical project for the Research Foundation Flanders (FWO).]


What is the mind? Is it a thing? Can it be located? Many philosophers have endorsed the distinct but related ideas that the mind is a thing and it is located inside your head. The standard view seems to be that the mind is identical to the brain, or simply the mind is the brain. Not everyone is convinced, however.

For example, philosophy of mind and cognitive science has recently seen the emergence of two paradigms that challenge mind/brain identity. The first, enactivism (Varela, Thompson and Rosch, 1991), contends that minds are “enacted”. You think, believe, desire etc in the ways that you do because of the activities you engage in. The second, extended mind (Clark and Chalmers, 1998), claims that, under certain circumstances, minds can be “extended”. That is, the states or processes that make up a mind can, on occasion, be partially located in the environment.

Given that both paradigms challenge the standard view, then it is perhaps not surprising that both paradigms have come under sustained attack. This has led to a heated back-and-forth with, on the side, internalists, who defend the orthodox position that the mind is an internal, brain-bound phenomenon, and on the other, externalists who counter that the mind is variously an enacted phenomenon (enactivism), or an internal-external phenomenon (extended mind). At present, the literature remains divided, with an apparent standoff between the two sides. All of which raises the question: can this internalist/externalist issue actually be resolved?

In this post, I will consider one strategy that has been favoured by both internalists and externalists alike. I will suggest that consideration of this strategy indicates that we should be sceptical that this internalist/externalist issue can in fact be resolved.

The strategy in question involves appealing to what I am going to call ‘special mechanisms’. Special mechanisms are causal states or processes that constitute or compose a given cognitive and/or mental state or process. Interestingly, appeals to such mechanisms have been used to vindicate both internalist and externalist positions.

For example, Clark (2009) has argued that there are special mechanisms in the brain that demonstrate internalism about consciousness. He claims that consciousness depends on high speed (or high bandwidth) information processing. Such processing is brain-bound, according to Clark, because, first, this processing depends on the synchronous activation of neural populations in the brain (Singer, 2003), and, second, the body acts a low pass filter and so slows down the transfer of information (Eliasmith, 2008). If consciousness is dependent on high-speed information transfer and, as a matter of contingent fact, this can only occur inside the brain, then the states or processes that realise or made up consciousness do not extend outside the skull. Simply put, consciousness is an internal phenomenon.

However, Thompson and Varela (2001) use similar empirical data i.e. data about the synchronisation and de-synchronisation of oscillating neural populations, to claim that consciousness is in fact an interactive, brain-body-world affair. They argue that brain mechanisms are “a paradigmatic example of self-organisation” (ibid, p419). That is, they understand the synchronous activation of neural populations in the brain, not in terms of the binding together of bodies of information, but rather as a self-organising emergent feature of the brain. They claim that emergence through self-organisation entails that neural, bodily and worldly elements can interact to produce emergent global organism-environment processes. Consciousness is consequently an emergent phenomenon, one that “cut[s] across brain-body-world divisions” (ibid, pp421-424). In other words, consciousness is external.

I offer this brief comparison between Clark, Thompson and Varela to illustrate the point that the same empirical data e.g. data about neural synchrony, can be used to defend opposing views on consciousness. Yet if appeals to such data can support such opposing views, then arguably this data cannot be used to settle the internalist/externalist debate.

(A further example of the same back-and-forth about special mechanisms could be recent Predictive Coding accounts of the brain, since such accounts have been used both to vindicate an externalist position (e.g. Clark, 2015) and various internalist positions (e.g. Milkowski, 2015; Gladziejewski, 2015)).

This offers us grounds to think that appeals to special mechanisms are not decisive in the internalist/externalist debate. I think this should encourage a certain scepticism that the debate can be resolved. But we can motivate this scepticism further by considering another means of settling this debate.

For one might take the alternative view that rather than going ‘top down’ – that is, from the mind down to the causal states or processes that accompany mentality – we should instead go ‘bottom up’ – that is, from the necessary and sufficient conditions needed to regard any state or process as mental or cognitive to the mind itself. This would then enable us to demarcate the boundaries of the mind and so answer the question as to where the mind is located. I shall consider this view in my next blog post. However, as with the appeal to special mechanisms, I shall argue that this view encounters its own set of problems.