Pattern and Being (part 1/intentionality), post-discussion
on Tuesday, I presented some findings by Gergely and Csibra suggesting that 1 year-old infants may be able to attribute goals to agents and assessing the rationality (i.e. efficiency) of their actions without interpreting them as fully intentional beings. Gergely and Csibra argue that this points to the existence of a “teleological stance” emerging prior to the full-blown Denettian “intentional stance”. The question that I posed to the group then was, how that fits in with Haugenett’s ideas of the intentional stance being constitutively governed by a commitment to interpreting actions as rational. There seem to be two possibilities to cope with these results from Haugenett’s point of view: One posibility would be to argue that there is something additional to commitment to rationality going on that we need to take the intentional stance (i.e.: you can’t have inentionality without rationality but you can have rationality without intentionality). A second possibility would be to criticise Gergely’s and Csibra’s interpretation of their data and claim that what children actually do in these experiments is to take the intentional stance.
Regarding the first possibility:
In the discussion, it was repeatedly pointed out that Haugeland understands pattern recognition as a skill, which would therefore also apply to recognising patterns in other people’s behaviors that afford interpretation of these people as intentional agents. However, I pointed out that a simple hand-wave to skillood in this context seems unsatisfactory as what we really would like to know is how this skill is actually possible and what we need over and above a commtiment to the constitutive standards of rationality to execute it. On this point, Haugeland has a short passage, which I also read in class but which is still entirely mysterious to me:
“The constitutive standards for a given domain – the rules of chess, for instance – set conditions jointly on a range of responsive dispositions and a range of phenomena: if they are both such that the former consistently find the latter to accord with the standards, then the former are recognition skills and the latter are objects in the domain. But such eventual concord is anything but vacuous: it is rare and, in general, a considerable achievement.” (p. 279)
That recognition is an achievement doesn’t add much to the general picture, however, and thus I am still left to wonder what actually goes on when we take the intentional stance over and above committing to a constitutive standard of rationaliy. It was also pointed out in discussion that we should take Haugeland to refer to the Dreyfusian notion of “skill” when he uses the term. That may be true, but in this it would still be desirable to have an account of how this applies to the case of social cognition.
Regarding the second possibility:
This was not discussed in class, but I think it would be fair to argue that the abilities observed in 1 year olds may actually be just a reflection of their more general ability to take the intentional stance already at that age. The paper I presented was of 2003. In 2005 Onishi and Baillargeon showed that children in a similiar age range were able to pass implicit (non-verbal) false-belief tasks. This is exactily what one would expect if one views the attribution of intentionality and rationality as inextricably linked.
All of this aside, I wanted to share another challenge to the viewpoint of Haugenett with you. In his The Architecture of the Mind, Peter Carruthers ( 2006, see p. 262/263) claims that it is a mistake to view norms of rationality as cosntitutive for thinking and for the attribution of thought. On Carruthers view, it is beliefs about norms rather than the norms themselves that do the work. Viewing norms themselves as constitutive for thought-ascripition and thinking makes it impossible, Carruthers claims, to naturalise thought in the first place. That is why Dennett can’t help himselv but to turn away from full-blown realism about intentional states (or so Carruthers claims). Do you agree that viewing norms of reasoning as constitutive for thinking/thought ascripition makes a naturalistic account of these thoughts impossible? I find myself agreeing with Carruthers that it seems strange to think of norms of reasoning as somehow constituting thought-processes rather than being constituted by them. Discuss!