Ontology, Justification, Direct Access, and Drano

by Asher Kay

We had a rousing discussion the other day here at Vole Central about emergence and reduction. The perennial topic of the justification of ontological theories came up, clothed in the distinction between strong and weak theories of emergence.

For the purposes of our discussion, strong emergence was defined as an “ontological” stance in which emergent properties are seen as a part of the actual, really-real world. Weak emergence, on the other hand, was defined as a more “epistemological” stance in which emergent properties are seen as necessary to our explanations of the real world, but not necessarily existent in the real world.

A lot of people think that we are pretty much doomed to a weak stance, essentially because we have no way of grounding ontology in the real, “noumenal” world. But even if we accept this rather Kantian limitation, the questions don’t go away.

Even if we can’t know the world as it really is, can we know something about the world? Our theories about how things are appear to have some relation to things rather than none at all. And if they do, we’re still stuck with the problem of justifying how we know those things. And it’s probably just as complicated as justifying a “strong” theory, because we still have to explain what our relationship is to things such that we can say stuff about them.

This justification can be called many things. I’ve been calling it a “theory of theories”, because it attempts to explain what makes our theories work. But it could also be called an epistemological theory, since it tries to explain how we know things. And it could be called an ontological theory, because it tries to explain how things are such that we know them. If you’re ever accused of conflating epistemology with ontology (or are accusing someone else of it), this is a good thing to keep in mind. When it comes to justifying theories, the two are conflated, at least to some degree. And I think we need to conflate them some more.

The burning question is: How are things such that our knowledge of them comes about? What access do we have to reality? Is it direct or indirect?

As a thought experiment, pour some Drano on your hand. The Drano is external reality, and your now smoking and sizzling hand is knowing about it. When I say “knowing” here, I do not mean you feeling the pain of being seriously burned by Drano. I mean that your hand is affected by the Drano in the exact manner of human flesh when it comes into contact with Sodium Hydroxide. Your hand does not have to know anything in particular about causticity, molecular structure, or anything else to be so affected. Your hand does not have to have a model of the hand-Drano interaction in order to be severely damaged. The “knowing” of the Drano is not “direct” in an absolute sense (your hand doesn’t, for example, “become” the Drano or “inhabit” the Drano, or even occupy a sensuous bubble of intentionality with the Drano — its “knowing” is just the damage and disfigurement), but we’re tempted to call this knowing “about as direct as it gets”.

Okay, you can wash your hand off now. Wasn’t that instructive?

It’s possible that we know the world in the same way as the hand knows the Drano. In the case of mental knowing, the process by which the world “damages” our conceptual apparatus is far less immediate and way more complicated, but ontologically, there may be no essential difference between the two processes.

Assuming that’s the case, what form does a theory of theories take?

[UPDATE: Apparently John at Ktismatics was already musing upon the selfsame subject of direct access to the world. It’s more evocatively put than this post *and* it doesn’t require a trip to the emergency room!]


16 Comments to “Ontology, Justification, Direct Access, and Drano”

  1. I think we’re barking up the same tree from different sides of the trunk, or some such analogy. Here’s a question: when the Drano eats into my hand, is it merging with the skin on my hand, becoming something else? The chemicals in the Drano interact at the molecular level with my hand, producing new compounds that are neither Drano nor human skin. Still, even with this chemical alteration the deformed and melting thing is still my hand at the emergent macro level. Of course if the burning continues eventually my hand melts away altogether and goes down the drain, clearing it of gunk as it goes…

  2. Regarding whether the hand “knows” the Drano, I think about the Biblical deployment of “to know” as a euphemism for “to have sex with.”

    It’s been posited that the representational theory of knowledge is an unwarranted extension of sensory encounters to cognitive ones, that the object being thought about makes a direct impression of itself in the mind. But the Drano doesn’t make a Drano-shaped hole in my hand; it interacts with my hand in an emergent way. So too with cognitive knowledge of Drano: it affects nodes and synapses in a way that doesn’t necessarily conform to the shape of the can or the molecular structure of the caustic substance.

  3. I also think we’re barking up the same tree. The conceit of saying that the hand “knows” something hopefully does less to anthropomorphize the hand and more to de-anthropomorphize knowing. There is a distinction between cognitive and non-cognitive, but it’s not as fundamental as some people have made it out to be.

    It’s really kind of Harmanian, now that I think about it. I’m saying that the basics of what we call cognition are already there with inanimate objects.

    And now that I think about it, it’s also a little Bryantian, in the sense that the hand’s knowing necessarily involves translation (in this case, damage).

    Your point about merging vs. becoming something else is one of the hinky things for me in this whole affair. I’m tempted toward a formulation that says that the hand (qua object) is just a conceptual structure, but I run into the same questions as I do in the post with the relation of thoughts to reality. We obviously don’t conceive of the hand as a semi-autonomous thing (or part of a thing) for totally arbitrary reasons. There has to be *something* about the hand in the first place that biases us toward seeing it as an object.

  4. Regarding whether the hand “knows” the Drano, I think about the Biblical deployment of “to know” as a euphemism for “to have sex with.”

    This takes the metaphor in a kind of scary direction, but it actually works very well. Who is the “active” agent in the biblical sort of knowing? It seems to be both parties. This active-ness on both sides gets at the crux of what separated Kant’s views from Hume’s in Kant’s “Copernican revolution”.

  5. But the Drano doesn’t make a Drano-shaped hole in my hand; it interacts with my hand in an emergent way.

    And it doesn’t interact (hopefully) with the other side of your hand. To me, the “formal” causality in the situation is the emergent constraints of the interaction.

  6. Maybe I’m just old fashioned and stodgy, but isn’t knowledge something like “justified true belief” (we can forget about Gettier problems for the moment)? So how can the hand know anything, since it can’t justify or hold a belief?

    I wonder if there’s some basic issue concerning the concept of ‘activity’ and ‘active’ in here somewhere. I mean, from a German perspective, ‘being active’ = ‘spontaneity,’ i.e. the ability to start or end causal interactions at will (definition of freedom, more or less in Kant). Within the purview of this definition, no causal interaction is ‘active.’ I might say that it is middle voiced (I have a paper on Husserl and the normative-logical character of ‘activity’ and the middle voice of intentionality, but it’s really boring — made Mikhail cry, if i remember correctly). does anyone else feel the tension between activity and causal chain/network of interactions (the latter being used for lack of a better word)?

  7. Asher, I agree that your idea of the hand knowing the Drano sounds Harmanian notes. Knowing has had such a long run in philosophy that it’s hard to talk about it applying to non-sentient things without evoking panpsychism. From my post, does the frog “know” that the flitting little object is an insect, even if that knowledge is instinctual and not propositional? You could say that the frog has justified true belief about the insect, and acts accordingly by nabbing it with his tongue. For that matter, do I “know” my neighborhood even if I can’t describe it adequately in words?

  8. Does the Drano have a justified true belief that it’s going to go medieval on your hand? It acts pretty confident about that. But isn’t there some value in having a vocabulary that allows us to distinguish the things we do as human-type objects from the things chemicals do as chemical-type objects? I mean, does the ontology have to be flat enough that my Carlness and the flitter’s bugness make no substantive difference that deserves a vocabulary shift? And therefore when we say the Drano knows the hand, aren’t we speaking metaphorically at best? What exactly is accomplished by blurring that distinction?

  9. Alexei – You are being old-fashioned and stodgy. Justified true belief is so… baggage-laden.

    But seriously. I think the exercise here is to kind of back up from ideas like “belief” and look at things through a complex dynamical systems metaphor. Causality and aggregation are truly bitches in this metaphor, but I’m trying to present “knowing” in a way that lets us see the kinship it has with other complex physical systems.

    Carl –

    But isn’t there some value in having a vocabulary that allows us to distinguish the things we do as human-type objects from the things chemicals do as chemical-type objects?

    Yes, absolutely. I think the way to go is to avoid being hegemonic or dogmatic about the description, or even to attempt to “reduce” the explanation of knowing to something that is “no more than” a chemical reaction. What I think a “chemical reaction” description does is give us a foothold on how integrated we are into the world. When we speak using a folk-psychological or (stodgy) analytic philosophical vocabulary, it creates the appearance of a large gap between the mental world and the physical one, which leads us to think of things like world/mind and physical/mental as dichotomies. I think that the more high-level vocabularies are needful in order to think through certain problems (especially when they involve things like truth and propositions), but we will always have to be careful about where the less stable ground is around the edges.

  10. John – I think what you’re pointing out is what I was calling the “less stable ground” above. A “belief” or a “piece of knowledge” is a formal construction that lives in a formal system – a fish that needs a particular temperature range and salinity level to thrive. If I define “belief” as something propositional in my formal system, then I can’t say that a frog believes anything. Within the metaphor I’m using, the instinctual reaction to a moving dark patch in a particular area of the visual field amounts to believing that it is food.

  11. I’m trying to present “knowing” in a way that lets us see the kinship it has with other complex physical systems.

    Aha. Carry on then.

  12. Right! Stop this thread! It’s silly, and I won’t have it.

  13. Re: direct access, definitely worth checking out the post “Philosophy Kills” at Overcoming Bias. The driving conceit is deathpoint cryogenics in hopes of eventual computer download of one’s neurons, and the question whether that’s really ‘us’ being downloaded; the question of access (via what we actually experience as ‘us’) figures prominently in the discussion.

  14. That was a really cool article. Everybody read it. Hanson is advocating the same idea as I am in this post, although admittedly, he’s expressing it better. Caplan’s part of the dialogue is also very interesting.

    I think Caplan is right to say that there’s a paucity of evidence for the physicality of mind. We have a lot to learn about how features like identity and awareness emerge from the brain. But Hanson would be right not to take his bet, because Caplan is looking for a piece of evidence that’s simply not there. He is wanting to point at the sense of identity in the pattern of physical interactions and say, “There it is” before he will believe. It’s impossible, in the same way that it’s impossible to point at the liquidity in water. To us, everyone else may well be a zombie.

    But it’s important, I think, to remember that a dualism like Caplan’s runs into its own problems. First, it has to explain how the non-physical mind is connected to the physical brain (which Caplan does try to do, in a relatively incoherent way). In essence, it ends up being the same problem.

    So in time, what you see is dualism’s solutions slowly converging with physicalism’s. Someone like John Searle’s philosophical movement over the years tracks this pretty well. Eventually, dualism will look down at the cups in its hands and see that everything is in the brain cup, while the mind cup is almost empty.

  15. I tend to agree with Alexei here. Only particular kinds (architectures) of physical systems are capable of knowing or believing anything. It seems to me that knowing at least requires the possibility of not knowing, but I don’t see how the hand or the Drano have the possibility of not “knowing” each other in the situation you describe.

    Perhaps a more productive formulation might be that information processing is happening between the hand and the Drano. Information processing might be necessary for knowing, but it needn’t be sufficient. Certainly the entropy of the system is increasing.

    and for baggage:

    Baggage means progress! And the Gettier problem is so fun! Justified true belief, let’s replace that with something Dretskean like: Does your hand believe that it is contact with Drano and is its belief caused by having the information that it is in contact with Drano?

    Anyway, what one knows one also believes, and having a belief means something like a willingness to act on that belief. Is the hand acting on anything? Does it believe anything? Does it know anything? Does it have the information that anything? I think I need to know more about hands. I’m more prepared to believe that the hand knows something than believe the same thing of the Drano. Why would that be?

    A lot of minimally cognitive systems are not representational systems (and hence perhaps not best described using words like belief and knowledge); they are cognitive-environmental couplings. Moving air vibrating the hairs on a fly’s legs trigger its leap into the air directly. It is acting, not on belief, but on minimal physical information processing of a stimulus. Does the fly’s leg know anything?

  16. Moving air vibrating the hairs on a fly’s legs trigger its leap into the air directly. It is acting, not on belief, but on minimal physical information processing of a stimulus.

    My question would be: what’s the difference in this scenario between the fly and the cognitive system of something that you *would* say knows something? The only real physical difference is that the information goes through more systems, all of the same basic sort that the fly has. Would the knowing be in those extra systems? If it is, then what makes them special that we’re inclined to say that they make knowing possible? If it’s in the particular structure of those extra pathways, then how can we express knowing as the constraints of a structure?

    It seems to me that knowing at least requires the possibility of not knowing

    Why does it seem that way to you? I think the answer to that will take us somewhere useful.

    P.S. Sorry I missed this comment.

Leave a Reply!

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: