We theists recognize two general categories of causation: mechanistic (i.e., “cause-and-effect”) and agency (“ground-and-consequent”). Most people, including most God-deniers, will initially agree that these two categories are real, and distinct, and unbridgeable … until they see where the argument is going.From recognition of the unbridgeable distinction between mechanism and agency, I argue that agency cannot “arise” from mechanism – this is what the God-deniers who haven’t denied agency from the start will then deny and this denial can then be shown absurd and thus false – and thus that agency is, and must be, fundamental to [the] nature of reality.
The important step is the proof that agency cannot arise from mechanism (as he puts it); it is not obvious that this is true, and the fact that people might 'initially agree' to it doesn't establish it as more than an unchallenged assumption.
(By the way this frame is older than monotheism in the West: Aristotle explains causality in just this way in the second book of the Physics.)
Consider that, as far as we can tell, atoms have no agency. An atom of carbon or of hydrogen or oxygen seems to decide on nothing; it joins into bonds, such as hydrogen and oxygen forming water, for purely chemical and physical reasons. This is 'mechanistic' determination on the Orthosphere's model.
Yet water has properties that its components, hydrogen and oxygen, did not. Both of these are gaseous at room temperature, for example; water is liquid at the same range of temperatures. Water has the property of 'wetness,' then, which has somehow arisen from the bond between the things that both lack that property. We can say some things about how and why this happens, but that it happens is clear enough. New properties emerge from combinations that happen mechanistically.
Why, then, should not agency be a property that emerges from things that happen mechanistically? Other properties, even complex ones, seem to do this. The carbon joins into long protein chains, the water is joined with it, and (skipping a long discussion) eventually you have DNA. This has a new property -- the capacity to order things it encounters mechanistically into a design that is not random but follows a kind of 'intention.' This ability to take from the world and put things into the order that is also 'you' is called life (as explained by philosopher Hans Jonas).
If this kind of proto-intention can arise from what appear to be mechanistic actions, why not a real intention? Why shouldn't it be true that living beings of certain kinds have the property of agency, even though none of their components had it before they were joined and ordered into that form?
This is, by the way, a good reason to reject materialism: it is not merely the material that matters. All the same material -- all the same atoms of oxygen and hydrogen and carbon, etc -- if not ordered in this way lack the properties of life and agency. These only seem to arise when the right order is brought to them. Thus, the form -- which is not material, but the way in which the material is ordered -- exists and is causally important, and not only the material. Reality is not materialistic but hylomorphic as the ancients said.
This is not an anti-theistic argument or a theistic one; you can make both arguments from this ground. Perhaps a God is then unnecessary, and being unnecessary should be excluded according to Occam's Razor. Yet what explanation is there for reality having this strange quality, such that thinking agents can and do apparently automatically arise from deterministic material processes? Why should reason and decision be inherent in a material that does not need them, existing whether or not agents do? Occam's Razor is only a tool for gamblers, not a proof; and here it seems clear that unnecessary things do exist, because we experience being one of those things all the time.
Perhaps, then, reality has this order because the order was wanted; and if it was wanted, there must have been someone who wanted it. Someone who had the power to set this basic structure of reality, either through design or through will, or possibly merely through longing.
As to physics, I am a determinist, (yes, yes, quantum supposed indeterminacy, though schrodinger's equation is quite deterministic also. That's a deep rabbithole that doesn't branch in the ways popular depictions indicate.)
ReplyDeleteThere is a reason for this: If you are trying to model how the universe works, and you hold that the universe follows a set of laws that refer only to the state of the universe, then determinism must follow or you reach a contradiction. It's for the same reason that no one can write a true algorithmic random-number generator.
For whatever reason, people seem to get crosseyed about their will when it comes to a deterministic view of the universe. At any given time, you exist. When you make a decision, there is a thing that makes that decision, by some process, according to whatever mental states you have at the time. Your internal states are part of the state of the universe. The universe includes you, by definition! You can analyze different outcomes of decisions by "cutting yourself out" of the picture and asking what happens with each of several choices (such a process is probably part of how we evaluate and make choices.)
Control theory is like this: The system will behave one of several different ways depending on what sort of thing you plug in to turn the sensory signals into a control signal. At the end of the day though, there is a thing that does that according to some process. (A dirt simple one in the case of thermostats and PID controllers, but it illustrates the principle.)
But you eventually have to put yourself back into the picture. There is a thing you do to make the choice you do. You work somehow.
I don't believe "intrinsic randomness" if such a thing really exists, alters the picture much. The important parts that make you who you are (as opposed to someone/something else) are the consistent, persistent (and therefore "mechanical") parts, not some arbitrary agitation that can only add noise to the system.
Forgive my putting it this way, because you are clearly an intelligent person and I don't mean to imply that your thinking isn't original. However, I would say that your description of the issue is probably the usual way that intelligent people approach the problem. I understand why it makes sense. What we can see clearly at the lower-middle level of physics looks deterministic (as indeed it does); what we can see at the grand high level looks deterministic (indeed even relativity can be graphed, as I was taught to do at one time); therefore, why shouldn't every level be so?
ReplyDeleteYet the lower-middle level and the grand high level share a characteristic, which is a relative lack of complexity. What is happening with the grand objects of gravity is easy to calculate because the grand objects exert such overwhelming force. And what is happening at the lower-middle level -- atoms and molecules and such -- is similarly simple. What is going on at the level of life, and even more so at the level of complexity involved in a brain, is quite a different sort of complexity. The simple may be determined because few things are at issue; and the grand, similarly, reduce the number of things you really have to think about. Supermassive objects overdetermine.
But it is our experience that life exists, and seems to have an intention of a sort; and it is our experience that we do, and clearly seem to. (That is not new; Kant, very familiar with the issue of determinism, points out that we can not act except under the idea of freedom in his Groundwork. If you think you'd like to argue about this, well, what sense does that make except if you could possibly want something, and choose to do it? What sense would it make to argue with me, if I were determined to believe X instead?)
I think it is likely that at this middle scale, actual agency is an emergent property. How and why it emerges from lower levels is a black box; but so too would be the deterministic argument that somehow a 'sense of agency' emerges that has no function whatsoever, since everything is in fact determined. That too should be barred by Occam's razor; yet either agency or the illusion must be true, since we do experience the sense of it.
Oh, a minor point of housekeeping: anonymous comments are welcome, but you have to sign them with some persistent name. Otherwise we can't tell anonymous commenters apart, and we have several interesting ones.
ReplyDeleteHello again. I'll go by "MadRocketSci". I sent the anonymous comment above.
ReplyDeleteI'm not sure yet if I even disagree with you. I personally don't see a conflict between determinism and agency. (Also, I sort of agree about the level of abstraction being important - whatever it is, agency is something that we care about on the level of people, and our attitudes about it probably shouldn't depend on the details of however microphysics in the universe operates.) "Atoms follow mechanical laws, therefore you don't get to run your own life" seems non-sequiturish to me.
Also, determinism doesn't mean that things can be predicted: Even certain very simple systems like double-pendulums, or the Lorenz butterfly equation, exhibit "chaotic behavior" (meaning that the state of the system for two very close initial conditions diverges exponentially with time.) An exponential increase in the precision of knowledge of the state only yields a linear increase in the length of time that the system remains close to a prediction. If we turned the moon into a sci-fi supercomputer, it couldn't predict the weather in Florida next Tuesday. And these are inanimate processes governed by laws that are "simple" enough that we know them.
To me it's a matter of identity. Even if we can't necessarily follow the evolution of the world, or even a very small part of the world (like a double pendulum), that doesn't mean that there isn't a way that it is, and a way that it evolves. And as far as agency goes, an agent is something that makes decisions given internal state and input. There is a way that any particular agent is, and how it will act. (The thermostat example, like any other mechanical metaphor isn't intended to belittle mankind or minimize complexity - it's just the simplest possible example of an agent, and one where you can easily "get inside it's head" and know what it does and why.)
To me, saying Bob has agency, means specifically that Bob has Bob's agency. One possible (far too crude still) definition of who Bob is is the person who chooses x in situation y. (for some set X and Y). (with internal state z element of Z). It's probably a better definition of Bob than "the set of atoms that were within these two cubic meters of space at 8AM on Monday" (ship of Theseus, respiration and eating and all that.) Closer to what we care about anyway, though subject to being hacked by Chinese Room arguments and the somewhat unsolved mystery of subjective experience.
The free will types seem to me to be trying, with their objections, to deny that Bob is necessarily who he is. They seem to want him to be "free" to be someone else at random.
MadRocketSci
Agency being emergent behavior: I don't disagree.
ReplyDeleteConsciousness being emergent behavior: Very probably the case, IMO, though Thomas Nagel would object that things aren't that simple.
MadRocketSci
Welcome.
ReplyDeleteI'm not sure how much we disagree either; however much it is, though, is all right. I'm comfortable disagreeing with people.
So here's something interesting you said:
"To me, saying Bob has agency, means specifically that Bob has Bob's agency. One possible (far too crude still) definition of who Bob is is the person who chooses x in situation y. (for some set X and Y). (with internal state z element of Z). It's probably a better definition of Bob than "the set of atoms that [etc]...."
The last sentence, which I excerpted, seems right to me. It's a better definition. I also agree that it is too crude to work as a full definition.
However, it does put me in mind of another thing Kant said about this. I don't want to seem too inclined to Kant, with whom I mostly disagree; but he was interested in this problem (via Hume) and had a lot to say about it.
Kant's idea about freedom (which is similar to 'agency' in the sense we are using it) is that it was exactly equivalent to one's capacity for rationality. He proposes a test of freedom in which you can show yourself that you are free in this true sense of being able to make a decision in a way that (he thought) an animal guided by instincts could not.
What he proposes is essentially that your Bob should refuse to choose x in situation y if and only if his reason told him that another course was better. Even though he might prefer x, might be habituated to x, might have done x his whole life, as a rational being he could reason to a decision that not-x was in fact the more appropriate choice; and, if he could then act on that decision, he was free.
People do this all the time. I have a friend who's a former Green Beret who, like many in the service, developed an intense alcoholism. (He was fun, too; we used to get together once in a while and drink like lords.) You could readily have said of him, as he was in those days, that he could almost be defined as 'the guy who does x given y and z,' so to speak. Yet he decided to get clean about ten years ago, and he did.
That's not dispositive; one could come up with a story to tell about it that was compatible with determinism. It does look like an instance of Kant's proof, though. Here was a 'Bob' who decided to stop being Bob, and to be somebody else instead. He had instincts, pleasures, deeply ingrained habits, and all of that; and he walked away from them and became someone else, at least insofar as we are using that definition, because he chose to.
These kinds of phenomenological evidences are of course not logical proofs, and our stories about our experiences are constructed by the same mind that has the experiences. Like I said, we can tell another kind of story. But it does seem plausible, at least, that he really was free and he really did something that defied all the things we'd normally think of as determinative.