Free will
How can some notion of “free will” exist in a deterministic (or randomly noisy) universe?
The unidirectional arrow of time creates a chain of events, like a chain of falling dominoes, that does not seem to admit to any gaps where something like free will might emerge.
An escape route out of this chain-like cage of thinking involves two key steps:
1. Our brains are capable of representing abstract, atemporal conceptions of the world, that are detached from any specific causal chain, and allow us to consider many different alternative courses of future actions and events, and their potentially distant future outcomes. We can engage in sufficiently extended and abstracted thought processes, such that the outcome of the process becomes conditioned by the content of the thoughts themselves rather than the underlying physical processes supporting these thoughts. In this sense, we are focused more specifically on the concept of self-determination: the behavior of the system is determined by something that we would recognize as the self.
2. This “level break” of causal influence across levels of physical organization is an example of emergence where the behavior of a system is determined by larger-scale organizational properties, rather than by the lower-level molecular constituents. A simple canonical example is that of two interacting gears, which exhibit emergent phenomena in terms of relative rates of rotation and torque, that depend only on the relative sizes of the two gears, and not on their molecular composition: the same dynamics can emerge across a wide range of constituent materials (wood, plastic, metal). These phenomena are thus fundamentally independent of the underlying physical substrate.
In the same way as the gears made of different materials, two different people can come to behave in similar ways to each other, by developing similar sets of internal mental representations (“values” and “morals”) that guide their behavioral choices, despite having entirely different causal life histories and molecular constituents. Conversely, the same person across very different moments in time and across many different physical contexts, can choose to behave in a similar and consistent manner, according to their own morals and values.
Furthermore, the molecular constituents of any given individual are constantly undergoing rapid turnover, constantly being broken down and regenerated anew (Rolfs et al., 2021), and yet we clearly experience a continuity of thought and memory over time that represents a strong emergent level of organization that persists independent of the lower physical components.
There is strong evidence that the synaptic weights among neurons play a dominant role in shaping the nature of our thoughts and memories, and these synapses are shaped by our experiences through learning mechanisms. Artificial systems composed of entirely different substrates can simulate these same processes and replicate a remarkable amount of the same cognitive and behavioral functions as the human brain, providing yet another demonstration of the reality of this emergent level of organization.
As in the examples above, two simulated networks with entirely different random initial weights will develop similar behavior when exposed to the same set of experiences, demonstrating that the system comes to be governed by what it learns, rather than by the randomness in the physical substrate. This includes considerable noise tolerance due in part to the threshold nature of neural spiking.
In summary, the human brain can exhibit “free will” because the it develops emergent levels of organization where the state evolution of the system is governed by “thoughts” that incorporate abstract, atemporal concepts such as morals and values and distant episodic memories, that then drive the future behavior of the system in accordance with these factors, effectively independent of the underlying molecular substrates of the brain itself. The cage of the causal chain is thereby broken, and we are free of its confines, as long as we continue to actually think about what we’re doing, and not just react reflexively.
See the page on emergence for a more thorough discussion of the properties that allow an emergent system to effectively become decoupled from its underlying physical substrate, and therefore exhibit an effective causal break from this substrate, to be governed instead by higher-level emergent properties: the “software” is in charge of the “hardware”. In effect, the brain is like a self-programming computer, where learning and our capacity for mental reasoning can establish new software programs that then drive our behavior, rather than being subject to any fixed deterministic behavior driven directly by the underlying hardware of the brain.
Cultural and life history determinism
Even if one allows that emergence does free the human mind from immediate physical determinism from the lower-level substrate, one could still object that the “true” causes can be traced back to the processes that shaped the “software”. Clearly, an individual’s thoughts, values etc are determined by their life history in the context of their culture, and therefore perhaps free will still does not exist? This then becomes more of a terminological question, which ends up exposing the fundamental inadequacy of the concept of “free will” in the first place.
It is difficult to conceive of any such notion where someone would be “truly free” of any past influences, or why anyone would would ever really want such a thing in the first place. Thus, attempts to establish some kind of strongly principled “pure” form of free will (i.e., “libertarian” free will) are likely to only be satisfied by some kind of magical, dualistic, metaphysical notion of “downward causation” or something like that.
Down here in the “real” world, of course we are all influenced by our life experience and culture, and from these experiences we develop our own internal mental world, with all of its own values and abstractions that together constitute the “self”. This is presumably what constitutes the “will” part of “free will”, and I think we want that part of the story too.
Thus, perhaps we could be satisfied by saying that people have the capacity to behave in a way that is not directly causally determined by lower-level physics, nor by higher-level cultural or life history factors, but rather is directly determined by a conscious, deliberative thought process that considers the various possible courses of action, and reasonably attempts to anticipate the outcomes and implications of those actions (i.e., self-determinism).
This is basically the legal definition of free will that entails a legal responsibility for one’s actions, and it is good enough for me. Even if much of the time we fail to consider our decisions very thoroughly, and thus end up not being very “free” most of the time, the legal burden, and the logical one here, is merely to establish the capacity, not the preponderance.
As a specific example, I know a lot of people who have become vegetarians as a result of engaging in “rational thought” about their values and the alignment of those values with their behavioral choices. Nobody forced them to do so. They just became aware in any of a number of different ways about the impacts of their choices, and decided to change their behavior accordingly. Two people receiving the same information may come to entirely different conclusions. Some of the critical inputs to their thoughts may include very remote events like the development of the ancient Buddhist philosophy thousands of years ago. These kind of very remote influences are inconsistent with simplistic local causal determinism.
For what it’s worth, I have not become a vegetarian (again – I already was one from 3rd grade until college), despite seeing the movie Okja, because I feel that death is a fundamental part of life (see meaning of life), and the “food chain” is unavoidable. Nevertheless, inspired by my colleague Seth Herd, I try my best to only eat “happy meat”, and I also do try to in general reduce my meat consumption due to the overall environmental impacts.
Metaphysical determinism
Even if one agrees that the emergent “level break” frees the human mind from immediate determinism driven by the lower level physical substrate, there is still a very large-scale metaphysical determinism that could possibly be at work, such that the exact same course of events would transpire if the universe were rebooted and run again with the exact same initial starting conditions.
We do not know enough about the actual physical reality at either the smallest or largest scales to know whether this kind of metaphysical determinism holds. Nevertheless it would in principle be possible to recreate a microcosmic example by running a massive simulation of a human brain on a digital computer with perfect error correction, so that it is entirely deterministic. In this case, you could end up creating a situation where the simulated brain is programming its own thoughts according to everything it has learned, while nevertheless existing within a perfectly deterministic universe.
Would that situation be any different than an alternate simulation where the computer was not fully deterministic, and it had genuine sources of non-pseudorandom noise? I’m not sure that it would. Therefore, my main conclusion is that such metaphysical level questions are not of much relevance, to beings that have a clear subjective experience of being able to think about their own actions (self-determinism) on a daily basis.
Compatibalism
The wikipedia entry on compatibalism captures many of the relevant issues with this interpretation, from my perspective, and I agree with the critiques provided by William James and Immanuel Kant:
James accused the soft determinists of creating a “quagmire of evasion” by stealing the name of freedom to mask their underlying determinism.
Kant called it a “wretched subterfuge” and “word jugglery”.
Specifically, if you posit that the human mind is determined directly by physical causality, such that there is no sense in which it could be considered to act in a manner that is free from the direct causal chain, and at the same time claim that this is nevertheless consistent with free will, then that does seem like a contradictory argument to me.
Interestingly, Kant’s view is similar to emergence-based approach, where conscious reasoning provides a clear break from ordinary physical determinism:
Kant’s argument turns on the view that, while all empirical phenomena must result from determining causes, human thought introduces something seemingly not found elsewhere in nature—the ability to conceive of the world in terms of how it ought to be, or how it might otherwise be. For Kant, subjective reasoning is necessarily distinct from how the world is empirically. Because of its capacity to distinguish is from ought, reasoning can “spontaneously” originate new events without being itself determined by what already exists.
I guess I need to go back and read Kant and see what kind of metaphysical story he had about the nature of human reasoning (no memories of such remain from my college philosophy courses).