On Truth and Rule-Following

by joshuahodges

Hello everyone,

Here is the link to my presentation from Tuesday:  On Truth and Rule-Following

I will provide a rough summary of my own talking points, followed by a recap of our discussion.

  • My Primary Goal:  to show that Haugeland can still preserve a sense of epistemic privilege for scientific knowledge.
  • The material on Brandom and Rorty is brief in order to keep my analysis manageable.  I am doing some research on Rorty now for my essay and would be willing to discuss this further in the event that anyone’s interested.
  • I argue that the phrase “Letting Be” (at least with regard to Heidegger and Haugeland) is unclear and potentially misleading unless you already know what he’s talking about.  I prefer to view it as a slogan and encourage the reader to tread carefully around this idea.  Although it wasn’t mentioned on Tuesday, I should note here that I have found some of the arguments in Brian Cantwell Smith’s “On the Origin of Objects” to work harmoniously with Haugeland’s conclusions on the nature of objects, and I think the two texts pair admirably (just as Dennett says in his review).
  • I emphasize the importance of finding “basic norms” which ground the possibility for objective truth-telling.

The initial class discussion focused on the relation between constitutive regulations and constitutive standards.  Can they be viewed as a hierarchy?  In the text, Haugeland separates the two after dealing with similar concepts in Rawls and Searle–I suggested that Haugeland’s conception is superior because it doesn’t lead us into making “Cartesian” errors about the nature of cognition and therefore keeps the pitcher from throwing us back to first base (see footnote 1 below).

Our talk then shifted into a critique of Haugeland’s stance on animals.  The question of existential commitment being essentially human (as per page 2) was raised by Alfredo.  I argued that this illustrates a potential problem for Haugeland; as best I can see, this is something he inherits needlessly from the existentialist tradition before him.  I discussed research from cognitive ethology and I brought up a few findings on insect minds, and concluded that Haugeland would be far better off to remain silent on this issue.  He notes at several points that his comments on animal minds are tangential to his main points and have little bearing on his core claims.  For what it’s worth, I actually think that some of the animal minds research plays very well with Haugeland’s ideas (see footnote 2 below).

The conversation shifted to an interrogation of Haugeland’s statement that “we should, however, be very hesitant to identify chess pieces with things.” (280)  As I recall, the following question was posed:  why doesn’t Haugeland just divide up his ontology into objects (such as water bottles) and their functions?  I want to (disclaimer) speculate that he avoids this because of his undergraduate degree in physics (see footnote 3).  Frank mentioned that a more explicit analysis of functions would strengthen Haugeland’s argument.

If I left out anything important, please feel free to post it as a comment below.  My footnotes follow.


1:  With regard to my March 5th post on Carruthers and the later comments made on the Pattern and Being post, I would like to now suggest that this is the key reason Carruthers’ arguments on internal norms fail to strike home against Haugeland.  I can explain my reasoning if anyone is curious or disagrees.

2:  A similar sentiment is put forth by Stephen Stich in The Fragmentation of Reason.  He makes the following comment while chastising some of the earlier work done by Davidson and Dennett.  “Philosophy has a long history of trying to issue a priori ultimatums to science, decreeing what must be the case or what could not possibly be the case.  Pace Kant, space is not Euclidean, nor are the laws of physics Newtonian.  Pace Hegel, there are nine planets, not seven.  But underlying my distrust of the a priori arguments against the possibility of systematically defective reasoning, there was more than a general skepticism about philosophy’s attempts to constrain science.”  (page 11)  This is just a nicer version of the same argument that Carruthers uses later on in his defense of belief/desire architectures in animal minds.

3:  On page 106, he openly mentions some of his uncertainty (haha) on how to integrate particle physics.  He explains that there is a tension between his understanding of micr0-physics and his desire to avoid eliminative materialism about wave-hits (mentioned in footnote 3).   And of course there is the 1997 appendix in which he says another argument is needed against the causal principle.  This is why I suspect that he avoids a simple account of object + function, because functions require an account of causality that he hasn’t yet formulated.  I’ll end my disclaimer here and invite someone with a better understanding of Haugeland’s ontology to weigh in.  Maybe if I get the chance to reread “Weak Supervenience” in a few days I can say more.  For what’s it worth, I’ve included a bit of analysis on this issue in my revised version of the handout:  keeping in mind that Haugeland is focused on the norms which are most “basic” for objective truth-telling, it seems to me that it won’t be enough for him to endorse just object + function.

Handout (Revised)

     “Even the most rudimentary (perhaps prelinguistic) truth-telling presupposes an integrated structure of rules of several different and interdependent sorts.” (305)

          This chapter of Having Thought is focused on the sorts of rule-following that are “most fundamental to our telling what’s what about worldly entities.” (305)  In the handout, I included a table matching Haugeland’s exhibited rules/governing rules distinction with Searle’s terminology and the direction of fit.

Failure #1:  Biologically Evolved Normativity

  • Biologically evolved normativity answers to natural selection.  Normal operations enable the organism to function successfully in reproductive or evolutionary terms.
  • Key Point:  This basic-rule following can also be reductively understood in biological terms.  This is OKAY for Haugeland. (see page 309, paragraph three)
  • Biological systems can also carry information.
  • But biologically evolved normativity fails because it cannot distinguish between functioning properly and getting things right.
  • So long as it’s ensuring reproductive success, it’s as right as it can be.
  • See the bird and butterfly example on page 310.

Failure #2:  Socially Instituted Normativity

  • Steer between the Scylla of infinite regress (all things are normative) and the Charybdis of non-normative basic levels (which would probably fall victim to the same critiques leveled at Davidson in the “Weak Supervenience” chapter).
  • Social norms are distinguished by the way in which they create compliance.
  • Social norms can also carry information, but they still cannot distinguish between functioning properly and getting things right.
  • So long as it conforms, it’s as right as it can be.  (This is another reason why a simplistic account of object + function probably isn’t enough — the function can vary dramatically!  Example:  sometimes I use a button to stand in for a chess piece that I’ve lost, but sometimes I also use it to fasten clothing.  In this regard, the button answers in many different ways to normativity, and offers remarkably different affordances.  See the first bullet point on what happens if you try to dodge this by saying the basic level is non-normative.  But perhaps I am attacking a straw man and have misunderstood the critique of Haugeland.)


  • Constitution is neither creating nor counting as, but rather letting be.  (352)
  • Actual phenomena are *known* when they are recognized via mundane skills.
  • Phenomena are *understood* when they are recognized as possible via constitutive skills.