In our previous posts, we have so far focused on: (1) clarifying our understanding of Ur-intentionality – REC’s positive proposal for understanding the thesis that basic cognition lacks content; (2) reviewing the problems faced by classic teleosemantic theories that motivate adopting REC’s proposal; and (3) detailing some of the theoretical consequences that adopting REC’s proposal has for how we think about and describe the contributions that brains make to cognition.
One advantage of going contentless about cognition is that it eliminates the need to deal with what we dub the Hard Problem of Content that haunts naturalistic theories of content (Hutto and Myin 2013, Ch. 4). A theory of content attempts to account for the existence of states that represent the world as being a certain way such that, in good cases, those states can be, e.g., true, accurate or verdical. Naturalistic theories of content are constrained in the resources they can employ in giving their accounts of representational content: they can appeal only to naturally occurring entities, states and processes – namely, naturally occurring phenomena that have their scientific credentials securely in order.
The most promising candidates among existing naturalistic theories of content are those that try to account for content in terms of some combination of informatonal covariance, structural isomporphism and biological function. While such notions can help us to understand contentless forms of cognition, if we are correct in our assessment, naturalistic theories that only employ such resources will fall short of delivering what is required for naturalizing representational content per se. There are compelling reasons to think that information-bearing states of affairs and structural isomorphisms – which are plentiful in nature– do not on their own, suffice for representational content. Crucially, understood as self-standing natural phenomena, they lack built-in norms and standards of the sort that is required for that task.
In adding biological functions into the mix, teleosemantic theories of content seek to overcome this problem by showing how to “ground the needed norms in evolutionary biology” (Millikan 1991, p. 151). Such theories ascribe representational content to internal structures that have been selected for their correspondence-preserving properties precisely precisely because such structures are used by systems in the service of specific kinds of world-directed biological ends.
There are two principle reasons for thinking that teleosemantic theories still come up short. First, as discussed in our blog on Ur-intentionality and argued at length in Hutto and Myin (2013), it is possible to explain these world-directed capacities in full without making any appeal to representational content. Second, though appeal to biological functions is enough to account for the normativity of world-directed cognitive acitivty, more is needed to account for the existence for norms and standards required for there to be contentful representing. The latter requires special forms of normativity of the sort expitomized by language users when they “bind themselves, in principle, to standards beyond themselves” (Price 2013, p. 37).
In light of the above considerations, and given that we that are not eliminativists about representational content, we are obliged to offer an alternative proposal of how and where it arises in nature. This is the project of accounting for the natural origins of content, or NOC. In Evolving Enactvism, we propose that origins of content can be explained by giving an account of how special forms of enculturated practices arose in phylogeny and how they are mastered in ontogeny, without presupposing the prior existence of contentful states of mind.
REC’s account of Ur-intentionality does important work in this regard. It can be called upon to defuse the worry that any socio-cultural account of NOC is simply a non-starter. Some hold that there is an essential tension in trying to give such an account of the origins of content if the following propositions all hold true: (1) Participating in and mastering socio-cultural practices requires cognition; (2) Cognition entails intentionality; (3) Intentionality entails representational content. If all of these premises are true then any socio-cultural account of the origins would hopelessly need to presuppose what it seeks to explain. However, REC’s account of contentless Ur-intentionality holds true then the third premise is false. For, in that case, there is more than one variety of intentionality and not all forms of intentionality entail content (for a fuller discussion see Hutto and Satne 2015).
Evolving Enactivism engages in further philosophical work that is needed to lay secure foundations for developing a detailed socio-cultural account of NOC. Such work is necessary because there are other arguments afoot that purportedly scupper our positive program before it even gets up and running.
One, allegedly crippling, objection that has been levelled against the prospect of providing a workable naturalistic account of the origins of content in REC’s terms, is that in distinguishing between contentless and content-involving forms of cognition REC presupposes a kind of psychological discontinuity that is at odds with evolutionary continuity (Menary 2015, Clowes and Mendonça 2016). In Evolving Enactivism, we show that the explanation of the socio-cultural origins of content only introduces kinks, but no fundamental discontious breaks into our story about the natural origins of content. We propose a multi-storey explanation – one that assumes that content-involving cognition could have arisen through the mastery of special socio-cultural practices but without changing the fundamental character of cognition. If we are right, a gapless account of evolutionary continuity is consistent with our working asssumptions and is possible without assuming psychological continuity.
Another argument against REC’s prospects carrying out the NOC program attempts to use the Hard Problem of Content againt us. Our critics insist that the Hard Problem of Content is a universal acid. They hold that if it works against any naturalistic theory of content then it works with equal force against all of such theories. Building on this assumption, it has been argued that the problems we identify as facing other naturalistic theories of content must apply with equal force to our own preferred socio-cultural account of the origins of content (Alksnis 2015, Shapiro 2014, pp. 217–8). Yet we see no inconsistency in holding that the Hard Problem of Content is hard, even impossibly so, for some naturalistic theories of content but not others. This will be so just in case different explanatory resources yield different explanatory prospects. The Hard Problem of Content is only impossibly hard for naturalists who limit themselves to using overly narrow resources when trying to explain the natural origins of content. By contrast we can draw on resources of a much richer set of empirical sciences than our rivals when developing our NOC account. These include: anthropology; developmental psychology; comparative psychology; and social neuroscience, among others.
What entitles us to this wider set of explanatory resources in pursuing our version of the NOC project? To answer that question, in a series of publications we have been making the case for adopting a relaxed as opposed to a strict and reductive form of scientific naturalism (Hutto and Satne 2015; 2017; 2018; In press). The relaxed naturalism we favour regards philosophy and science as mutually constraining, without assuming that their methods and tools reduce to one another. As such, with respect to the NOC project, its philosophical work consists mainly in clarifying key theoretical issues and synthesizing empirical findings so as to develop sketches of plausible explanations.
Assuming our ground-clearing efforts succeed, the way is open for fresh investigations into the natural origins of content, building on REC’s account of basic cognition. In Evolving Enactivism we sketch, in broad outline, what that sort of positive account will need to address and work is already underway in developing the account more fully (see Satne 2014; 2015; 2016; Satne and Salice 2016; In press).
Alksnis, N. 2015. A dilemma or a challenge? Assessing the all-star team in a wider context. Philosophia. 43:669–685.
Clowes, R., and Mendonça, D. 2016. Representation redux: Is there still a useful role for representation to play in the context of embodied, dynamicist and situated theories of mind? New Ideas in Psychology. 40 (A): 26–47.
Hutto, D. D. and Myin, E. 2013. Radicalizing enactivism: Basic minds without content. Cambridge, MA: MIT Press.
Hutto, D. D. and Myin, E. 2017. Evolving enactivism: Basic minds meet content. Cambridge, MA: MIT Press.
Hutto, D.D. and Satne, G. In press. Naturalism in the Goldilock’s zone: Wittgenstein’s delicate balancing act. In Wittgenstein and naturalism. Cahill, K. and Raleigh, T. (eds.). London: Routledge.
Hutto, D.D. and Satne, G. 2018. Wittgenstein’s inspiring view of nature: On connecting philosophy and science aright. Philosophical Investigations.
Hutto, D.D. and Satne, G. 2017. Demystifying Davidson: Radical interpretation meets radical enactivism. Argumenta. 3: 1. 127-144.
Hutto, D.D. and Satne, G. 2015. The natural origins of content. Philosophia. 43: 3.
Menary, R. 2015. Mathematical cognition—A case of enculturation. In Open MIND: 25 (T), ed. T. Metzinger and J. M. Windt. MIND Group.
Millikan, R.G. 1991. Speaking up for Darwin. In Meaning in mind: Fodor and his critics. Loewer and Rey (eds.). Oxford: Blackwell.
Price, H. 2013. Two expressivist programmes, two bifurcations. In Expressivism, pragmatism and representationalism. Cambridge University Press. 22-45.
Satne, G. 2016. A two-step theory of the evolution of human thinking: Joint and (various) collective forms of intentionality. Journal of Social Ontology. 2:1.105-116.
Satne, G. 2015. The social roots of normativity. Phenomenology and the Cognitive Sciences, 14:4. 673-682.
Satne, G. 2014. Interaction and self-correction. Frontiers in Psychology. 5.796. https://doi.org/10.3389/fpsyg.2014.00798.
Satne, G. and Salice, A. 2016. Helping behaviour and joint action in young children. Phenomenology and Mind, 9. 99-106.
Satne, G and Salice, A. In press. Shared intentionality and the cooperative evolutionary hypothesis. In Minimal Cooperation and Shared Agency, Fiebich, A. (ed). New York: Springer.
Shapiro, L. 2014. Review of Radicalizing enactivism: Basic minds without content. Mind. 123 (489): 213–220.
What you call the ‘hard problem of content’ is at once the thing that most excites me about your work as well as the thing that most troubles me. It excites me because the underdetermination perpetually plaguing attempts to solve content is so obviously a crucial data point. But when it comes to chronic underdetermination, ‘content’ is by no means alone. The hard problem here is the hard problem of intentionality, and not just content. As a result, we should expect that ‘ur-intentionality’ will run afoul the underdetermination plaguing all intentional posits.
It really could be the case that your deflationary formulation of intentionality might be the one possessing the explanatory resources required to eventually compel consensus, to solve the hard problem of intentionality, but ‘raw directedness’ strikes me as susceptible to endless reinterpretation/recontextualization as ‘raw feels.’ What do you see blocking a pessimistic induction like this?
There’s much in the distinction between contentful and contentless cognition that I like, so I hope my criticisms will be taken up as suggestions to an improved account rather than objections.
1. I’d still like a better understanding of what you mean by content. At times Radicalizing Enactivism and Evolving Enactivism read as though you were saying that Fodor was right about what content is, and we just need to supplement a Fodorian account of content with an enactivist account of contentless world-tracking dynamics. But at other times (as here) the emphasis is put on the role of scaffolding socio-cultural practices. That would seem to align your project much more with a neo-pragmatist account of content, such as Brandom et al. An account of content in terms of inferential states instituted by normative statuses seems (to me) far more compatible with enactivism than a strictly cognitivist account.
2. I’m puzzled by how the REC position is related to Price’s new bifurcation thesis. You seem to claim Price as an ally. But Price is quite clear that he’s distinguishing between two quite different kinds of representation, i-representations and e-representations. It would seem that the REC position would have to be that Price is mistaken about e-representations: there aren’t any! The dimension of cognitive activity that does the work of information-sensitive feature-tracking or affordance-tracking is not representational at all, so there aren’t any e-representations. Rather, on the REC view, all representations are i-representations. So it seem that you would need to either be far more critical of Price that you have been thus far, or keep the alignment with Price but say that Ur-intentionality does involve e-representations after all.
3. The “duplex” or perhaps “multi-storey” view here is quite similar to the distinction I urged between discursive intentionality and somatic intentionality in Intentionality and the Myths of the Given. Mark Lance suggested that my view has a problem: by pushing these two capacities so far apart, I ended up in the position of saying that perceptual judgments involved the conjoint operation of two distinct capacities. He used chessboxing as an example: chessboxing is a sport that involves players going back and forth between two different activities, boxing and playing chess. But this simply does not cohere with the phenomenology of perceptual judging. We don’t take ourselves to be doing two different things, but one thing. I worry that the REC view, like mine, pushes contentless cognition and contentful cognition so far apart that it becomes a Mystery as to how they are functionally integrated. (Consider this a neo-pragmatist/enactivist descendant of the Cartesian mind-body problem!)
4. There is still the big question as to how socio-cultural practices evolved from contentless minds. I would like to suggest that you take a close look at Joseph Rouse’s Articulating the World. Rouse tries to explain how we get from merely Gibsonian creatures to genuinely Brandomian rational animals. The answer, he thinks, lies in niche construction. He does not think that Gibsonian creatures exhibit Ur-intentionality, but that is (I suspect) because he is, following Haugeland, a Heideggerian about intentionality, and REC builds on a much more Merleau-Pontyian story that has its theoretical basis on Husserl’s late work on “operative intentionality”. Nevertheless I think that Rouse is someone that you should be in conversation with as you develop the duplex view. (He calls his theory “two-dimensional normativity”: the normativity of biological teleology and the normativity of socio-linguistic commitments and entitlements.)
CS, do you know if anyone in the Brandom camp is trying to incorporate the research into cognitive-biases and the like?
I’m not aware of anyone making those connections. In my own thinking, the link between the empirical research on cognitive biases and the Pittsburgh program of explicating the space of reasons is provided by Hugo Mercier and Dan Sperber’s The Enigma of Reason. They make no mention of Brandom but I think that their view of reason as an essentially social process is perfectly consistent with the more generic account that Brandom develops.
thanks I was thinking about Dan & Hugo when reading yer comment, reminded me of Rorty making the case that we are always already manipulating our environs.
Apologies to you all for not yet replying to your excellent comments and queries. I have been busy completing a new paper on basic emotion theory – https://www.academia.edu/35844753/A_New_Better_BET_Rescuing_Basic_Emotion_Theory. I am now working flat out on another, with Erik, on whether debates about mental representations are required for cognition are pointless and/or merely semantic. I hope to post replies within the next few days.