Sunday, 7 February 2016

Extended minds and wisdom sits in places

Are minds extended?

This is a reaction piece to 'The Extended Mind Paper' (Clark and Chalmers, 1998) co-authored by Andy Clark and David Chalmers with references to some of the more recent work done in this area. I am sure that my impression might change next week but such is the nature of my 'mind'.

When reading the extended mind readings for this week I was intrigued and stumped, in equal measure, as it sat awkwardly with some of my intuitions, and comfortably with others. Are cases of people using memory aid tools such as diaries, other people, Evernote (in a reliably coupled systematic way) to function in their day to day activities, to be thought of as cases of minds being extended e.g. a protestant minister who can no longer form new memories, following brain damage, who employs Evernote to successfully go about his duties (Clark, 2014)?

I am not wholly convinced by the premise of minds as extended, though I am sympathetic to the views extolled. So is it something useful to think with, but little more, or are they genuine cases of cognition? There is an issue concerning the lax usage of the terms cognition and mind in this piece which reflects a more general problem in cognitive science which I will not be able to tackle today (that mind is a contested concept, not to mention that cognition is too). For some thinkers, minds are synonymous with cognition and that cognition only takes places within the brain, which is the classic cognitivist view, of which there many variants. For others it is the view that the mind is thought to be embodied, that the cognising agent is the brain and body in dynamic interplay with one another. Enactivist approaches go further noting the dynamic relationship between the agent and the environment that is integral to cognition. I do all of these diverse theoretical outlooks a great disservice by passing over them in this way but I'll leave it to the readers to explore these avenues. Is this simply a case of different definitions of the same term being used to describe radically divergent activities, for which Clark and Chalmers should be taken to task, for their muddying of the water? I suspect that the water has been muddied long before this. With no expectations of putting the 'what is mind', 'what is self' dilemma to rights let's explore the Extended Mind hypothesis (EMH) a little.

Basic intuitions about the self

The EMH goes against my intuition of what might be called the introspective self. I grew up, perhaps mistakenly, entertaining the intuition that I am an individual whose deepest thoughts are private, cut off from the world except by some sort of mediation through linguistic, aural or visual representation. For some of us this intuition may lead to the conclusion that we are self-constituted through sheer will, a Nietzschean style self-overcoming or mastery, so as to explain our agentive actions in the world. To talk of minds as extended seems to go against this intuition. Despite once favouring this intuition, which unchecked descends into an unproductive idealism or solipsism, I would not put too much stock into the idea of the individual cut off from the world. Anti-realism of this sort, whilst interesting to think with, doesn't lead anywhere. Given my anthropological background, I am very much aware of the complex interplay between the individual and the environment, that together constitute that person. From a holistic approach, the individual is necessarily an integral part of a society and a natural environment (again this draws the nature/culture divide but it is heuristic in this case) and is involved in contributing to and reconstituting the social and natural environment around them. This recognition of the complex and largely inextricable relationship between individuals and their environments at least makes me open to the EMH. But is it useful to think with? Does it yield any explanatory power or lead to good questions? On those grounds I am not sure.

The extended mind; is it a productive thesis?

If we are to say that minds are extended do we dilute the concept of mind to the point where it refers to too many situations and coupled relationships? It appears to be a case of once knowing the EMH you begin to see minds being extended everywhere. This seemingly ability of the EMH to be extended (pun intended) to many cases is a weakness rather than a strength as it can largely explain away counter-arguments by appealing to what their, rather loose, definitions of what constitutes cognition and minds [See Popper on falsification Falsifiability]. It doesn't seem to generate testable questions or robust definitions. Would it be harsh to say that it is a gloss rather than the foundations of an explanatory theory of mind? The Extended Breath (Thagard, 2013) parody is particularly good at exposing some of the flaws in the EMH's logic, though the piece does embodied cognition and enactive approaches an injustice in its lazy characterisations.  

There is a definitional problem regarding minds. This is not to mention the difficulties of defining cognition, and what this entails. Are minds the product of cognising systems? In spite of the bold claims of the paper Clark does not seek to have minds extended everywhere, especially if you look at Clark’s more recent work (as discussed in this video lecture in Edinburgh Clark 2014). Clark and Chalmers refer to only specific cases which satisfy the necessary conditions, as was the case for the protestant minister. At this point Clark and Chalmers diverge as Chalmers has went down the rather more radical route of pantheism (I  link here a great philosophically themed comic 'Existential Comics' Chalmers and the Panpsychists). Clark refers to minds as fluid but what he argue for mostly, especially if you look at his more recent work, is the embodied mind approach with specific cases meeting the EMH criteria. He rightly takes the view that is not fruitful to study minds as disembodied from brains and bodies. The mind, if one takes this view, is the brain and body in dynamic interplay with one another and cannot be separated from one another without reducing the power of description and inevitably closing off avenues of study. Clark goes further to extend this definition of mind to include aspects of the environment such as the diary or Evernote. This seems plausible according to his definition of mind and what constitutes cognition but one problem emerges immediately. The word mind will be used to describe a plethora of possibly related but markedly diverse phenomena. In one case, for example, the term mind is used to describe a person performing a memory task, such as remembering where a place is, and in another case the term mind is used to describe a person using technology to augment their memory and recall. So would we have to say in one case we are talking about MindType1 and in another MindType2 and so on so as to be able to differentiate between them? The EMH, on these grounds, poses definitional problems and obscures rather than clarifies the issues. Now I will briefly look at one example from anthropology that appears to be amenable to investigation using the EMH. 

Wisdom sits in places; place, language and extended wisdom

"And what about socially-extended cognition? Could my mental states be partly constituted by the states of other thinkers? We see no reason why not, in principle" (Clark and Chalmers, 1998)

"Knowledge of places is closely linked to knowledge of the self, to grasping one's position in the larger scheme of things, including one's own community, and to securing a confident sense of who one is a person. " (Keith H. Basso, 1996)

Clark and Chalmers 'Sea of Words' idea, that is, that we are born into a social network united by language and comprised of other people, is an interesting argument for the EMH. It is reasonable to argue that much of how we think is reliant upon, or perhaps it is better to say enabled and constrained by, the affordances made by the social world around us. This particular iteration of the EMH prompted me to think of a book I read seven years ago on the linguistic practices of the Western Apache. Anthropologist Keith H. Basso (Keith H. Basso, 1996) presents a compelling and beautifully written ethnography, centred principally around four key persons, in his book 'Wisdom Sits in Places' (I will leave it to your interest to follow up on this book). He conducted his fieldwork in Cibecue, Arizona, amongst the Western Apache. Their principally oral way of life led him to emphasise the relationship between language, place and culture in forming a people's way of life in the world. Place is a key feature of how this group of Apache organise themselves socially, see and move in the world, and record history i.e. through oral histories, with reiteration of place names, playing a key role for memorisation. In this case metalanguage e.g. using a place-name to index a story related to that place, and thus invoke a moral or instructive lesson, toponymy i.e. place-names, and physical places in their environment, form an integral part of memory, and more broadly social identity for this group. The EMH could be used to argue that minds or cognition e.g. in this case memory, for the Western Apache, are extended here to language, specifically metalanguage, and the physical environment. So why don't I want to bite? Because the EMH doesn't enrich the study of this cultural group and their practices. The importance of place in this particular group can be described robustly in terms of toponymy, environment and metalanguage, without invoking the idea that this is a case of extended minds. There is evidently a dynamic relationship between the members of this group, and their environments, made possible by particular patterns of social interaction and linguistic convention, which merits a holistic approach but the EMH doesn't seem to provide the necessary framework.

In conclusion Clark and Chalmers have rightly drawn attention to the problems of viewing cognition and/or minds as being consigned to individuals, whether this is skull bound or brain/body bound etc. But rather than providing a robust set of definitions to do some science with they only serve to confuse the issue even more. So I would say that it is a piece to think with rather than a prompt for a radical overhaul of our understanding of minds, cognition, and human behaviour. This radical overhaul will, if it does, come from approaches in embodied cognition and enactivist accounts of human intelligence and behaviour in the world.


  1. Nice post Sam. It should be noted that the EM hypothesis seems to be adopted in one form or another by most folk within the embodied and enactive schools, at least, but I suspect that it is often interpreted in a manner that Clark, at least, would be uncomfortable with. Once a thesis is loosed upon the world, the originator does not get to decide how it will be received.

    1. Thanks Fred. Yeah it's interesting to see Clark discuss his ideas on the EMH now, and then see how the EMH is used independently of his work throughout others' research and informal blogging. The hypothesis can be seen 'extending' everywhere.