Yes, and it is this particular philosophical definition that I'm challenging, since the sort of determinism entailed by the idea that we make decisions based on a mixture of internal and external factors is very different than the type that places conscious choice outside of the causal chain and reduces all behavior to unconscious, physical processes. When you conflate these two types of "determinism," one of which is required for free will and the other of which is incompatible with it, you basically define the libertarian view out of existence. Which is very convenient for the determinist, I'm sure, but not so great for discourse!
I'm not really interested in defining this or that out of existence, but just using simple definitions to produce a coherent description of the world.
It seems to me that the definition of determinism I described (i.e. events have prior causes) says nothing about consciousness or free will beyond the implication that they will be causal (or, at least,
effectively causal). It also seems to me that not all behaviour is unconscious, in as much as we are conscious of some of our behaviour; also to the extent that behaviour consists of actions, it is physical.
I don't see how conscious choice can be 'outside the causal chain'. The choices we make, whether we are conscious of them or not, are not, in general, random, so are, in general, causal.
This seems to me quite consistent with free will, if free will is the ability to make choices according to our preferences, inclinations, desires, needs, wants, etc., without coercion or constraint. It only suggests that those preferences, inclinations, desires, needs, wants, etc., have causal origin.
... just as the brain is configured by events outside of our conscious direction, neuroplasticity means that we also have the power to reconfigure it. It is at the higher operating levels that we decide who we are to be, and this is something that we decide every moment of every day. I take the "will" part of free will very seriously, since I think freedom in this sense is more a matter of discipline than something we have absolute access to at every moment.
Sure, we're complex goal-seeking, learning, self-modifying systems. We can model outcomes, plan ahead, and defer gratification. We make decisions for reasons - e.g. pleasure, pain, reward, etc., and our innate and learned values and past and ongoing life experiences determine our goals and motivations in respect of those reasons.
I don't really care about quantum indeterminancy, except insofar as I'm intrigued by the Aristotelian interpretation whereby potentiality and actuality are genuine features of reality.
Yes; modern physics has some interesting ideas relating to potentiality and actuality involving certain versions of the multiverse, where anything that can happen (potentiality) does happen (actuality). This is true of both the cosmological multiverse (i.e. if our universe is spatially infinite), and the quantum multiverse (Everettian 'Many Worlds'), where quantum indeterminacy is the result of each possible outcome having a 'version' of the observer.
Oh, the two-stage model of free will is not specifically my position. On the days I'm not inclined to take a Mysterian approach and declare free will a genuine paradox, I prefer an agent-causal model. We are agents in our own right and more than just the epiphenomenal accumulation of more basic physical processes. I don't know how well this works on a materialistic metaphysics--the closest I ever get to that is a slightly naturalized Aristotelian hylomorphism, but you could probably invoke emergence to account for the sort of top-down causality necessary to view the person as a causal agent.
I would still be interested in an example of what you see as an exercise of free will, and where you see causality ans indeterminism being involved.
Aristotlian metaphysics is interesting background material, but given the advances in physical knowledge, I'm not sure how useful it is to modern metaphysics.
I think you're right to introduce emergence here. There are levels of description or abstraction, each with its own ontology and descriptive language, and it's a mistake to mix levels of description. Just as, on one level, a gas is a collection of freely moving atoms or molecules with varying velocities, and, on another level, it is a compressible substance with a temperature and pressure. I suggest that top-down causality is a high level of description of certain emergent behaviours of a system composed of elements that have their own behaviours at a lower level (the interacting patterns of
Conway's Game of Life, or flocking and shoaling behaviours come to mind - you can describe the movements of individual birds as being caused by the movement of the flock, or the movement of the flock as caused by the individual movements of many birds). It also seems to me that agency is part of a high-level behavioural language, where the agent is a particular nexus that (crudely) accumulates environmental influences and processes them to produce behavioural outputs.
As for the two-stage model, there's a fair amount of information on that here:
http://www.informationphilosopher.com/books/scandal/Two-Stage_Models.pdf
Thanks, an interesting paper. There does seem to be a consistent use of 'random' and 'chance' to mean unexpected or unpredictable due to underlying complexity (e.g. the interaction of untraced or untraceable causal chains that are not directly related). This kind of randomness or chance is subjective, a result of lack of knowledge of the deterministic processes involved - it's pseudo-random.
Also, I think the suggestion that, "...in a deterministic universe, there are no genuinely new creative acts. Determinism is “information preserving.”", is mistaken (although it depends on precisely what is meant by the terms they use). Creativity is the combination and/or manipulation of existing concepts and ideas to produce novel concepts and ideas, which does not exclude determinism; and if we take a simple definition of information to be a particular arrangement of matter, new arrangements are not beyond determinism either (this also applies for arrangements with semantic content).
