Why emergent levels will not save free will (& 2)

Photo: Akshay Nanavati / Unsplash

In the first entry of this series, I briefly explained Christian List’s attempt 1 to vindicate the ontological and scientific reality of intentional action as a real emergent phenomenon. In a nutshell, intentional, deliberate and often ‘rational’ action is not a fiction (as some skeptics like Alex Rosenberg have defended 2), but a totally legitimate inhabitant of the natural world, a type of process that emerges out of its microphysical basis, but that is ‘autonomous’ from them, in the literal (or better, etymological) sense of having its own (autos) natural laws (nomoi), laws that are as causally efficient as those of any other complex natural cause (like chemical or biological ones), and are as apt to be scientifically studied as them.

My goal in this new entry is to show that having established the reality of the phenomenon of intentional action as an emergent natural process is far to be sufficient for providing an argument in defence of the reality of free will, in the classical sense in which defenders of free will as a philosophical (or ‘metaphysical’) concept maintain that we are free, i.e., the sense in which our intentional actions are ‘first causes’ (i.e., not mere effects of other previous facts on which we have no control) and the result of a ‘real choice’ (i.e., one in which we had the real option of having decided otherwise). I described here these two conditions under the labels of ‘ultimate control’ and ‘possibilism’, respectively. After all, what we need is not merely to show that intentional, voluntary decisions are real, non-reducible causes, as flowering or digesting are, say, for obviously the flowering of a rose is not ‘free’, no matter how ‘emergent’ it is, and how robust our scientific knowledge of the process is (on the contrary, we tend to think that the more we could ‘scientifically’ know about a person’s decisions, the less freedom we could attribute her). The most important piece of the conundrum lies in the question of what List calls ‘alternative possibilities’: when I take a decision, might I have really chosen otherwise? (and, in case I might, would the choice have been really ‘mine’?).

According to List, the worst enemy of the philosophical concept of free will is determinism: if the course of events were totally determined by the past state of the world, then this rose would have had no other option but to flower in exactly the same way and at the same pace it has done, and exactly the same would happen with all our decisions. List recognises that the laws of quantum mechanics are to some extent indeterministic, but also that they is not science’s ‘last word’: perhaps the really ultimate laws of nature at the most fundamental physical level are deterministic; but this is something we cannot know, at least by now. Instead of deferring to what fundamental physics ends saying, List’s strategy consists in showing that even if nature is deterministic at the most basic level, this does not imply that all natural phenomena (in particular the ‘emergent’ ones) are condemned to be deterministic as well. I.e., determinism at one level (say, the micro level) wouldn’t imply determinism at higher levels (say, the macro levels). In his own words:

It is easy to see that, unlike the micro-histories, the macro-histories are not deterministic. Regardless of the system’s macro-state at time t = 1, several sequences of subsequent macro-states are possible: the macro-histories exhibit branching. This illustrates that macro-level indeterminism, such as the indeterminism we find in the human and social sciences, can be an emergent byproduct of micro-level determinism. More technically, the property of determinism is not preserved under changes in the level of description, such as when we move from a lower, more fine-grained level to a higher, more coarse-grained one. Crucially, all of this is entirely consistent with the higher level supervening on the lower one.

This simply means that, if a natural process or entity (for example, a rose blossom, together with all its relevant environment) is in a particular macro-state (one that is consistent with many micro-states; i.e., that can be ‘multiply realized’ by these), the natural laws of botany do not entail that the blossom will flower towards exactly another determinate macro-state and only to that one macro-state, but the leave ‘open’ several possibilities. Of course, if the laws of the micro level are deterministic, the micro-state in which the blossom actually is will evolve in exactly one possible way and only in that way, with no ‘options’ at all. But, according to List, since the macro causal processes are scientifically legitimate realities, indispensable for any explanation of the macro facts (as we saw in the preceding entry), it would follow from this ‘indispensability’ that determinism at the micro level does not overcome the macro level, which, as its own laws assert, is still indeterministic.

The problem is that this is a non-sequitur. List claims that, since the macro laws are ‘ontic’, and not only ‘epistemic’ (i.e., they are not a mere ‘perspectival description’ of what happens, but correspond to the real, substantial truth of the macro causal processes, as real and substantial as the micro causal processes themselves are), the indeterminism we find in the macro laws is not a mere epistemic limitation of our own, but is something as real and substantial as the determinism we find (if we do) in the micro laws. But, pace List, there is absolutely no reason to assume that accepting the ‘substantiality’ of macro causes entails anything about the ‘substantiality’ of the possibilities or ‘options’, which is what is required to sustain a substantial indeterminism. At moment t0 the blossom is both in a particular macro state (that we observe) and in a particular micro state (one amongst a plethora of micro-states that are both compatible with that macro-state, and impossible for us to discern, and, more importantly, irrelevant for us to predict and understand what will and may happen to the blossom); in which macro-state will it be at moment t1? The laws of botany, of course, do not totally determine which one, and say that the blossom may flower in different ways. However, if the fundamental laws of physics (or the physical laws at some micro-level) are deterministic, they do determine in which micro-state the rose will be at t1, and it is this future micro event what determines in which macro-event the rose will be at t1 (for we suppose that the macro-events supervene to the micro-events). List’s argument forgets that determinism at the micro-level completely eliminates all ‘possibilities’ at the macro levels: there is, in a deterministic nature at the micro level, only one ‘possible’ chain of events, the one that actually happens at that level, and correspondingly, there is only one possible chain of events at every macro level: the chain of macro-events that the actually occurring micro-events determine according to the supervenience relation. This is compatible with macro causes being real and with macro laws being indispensable; it is only incompatible with interpreting these real and indispensable laws and causes as referring to substantially existing alternative possibilities. The real and indispensable macro causal laws are, sadly, just incomplete laws: they fail to help us identify the-one-and-only-true-course-of-events that deterministic physical laws assure (if they are right) that happens in the world at both the micro and the macro levels.

Lastly, even if the indeterminism apparently entailed by the incompleteness of the psychological laws were really substantial (i.e., even if List were right in saying that we really have alternative options to choose amongst when we take a decision), this would be an insufficient argument in order to show that our decisions are really free, in the sense of our having ‘ultimate control’ on them. For, obviously, the flowering of the rose blossom might be equally ‘indeterministic’ in that very same sense, and the rose, however, wouldn’t be ‘free’ at all. The rose ‘choosing’ one way of flowering instead of another, between the ‘open alternatives’, would be a matter of mere chance: the botanic causal laws just leave ‘open’ which state will be realized, and the selection of one state amongst all the equally possible ones will be like drawing dice. In the same way, our state of making an intentional decision instead of another one (amongst the alternative options the psychological laws leave open) would be just the result of a ‘lottery’ if indeterminism is substantially true at the psychological level.

In conclusion, when examined with sufficient detail, List’s ingenious efforts to save free will with the help of emergentism are foundering.


  1. List, Ch., Why Free Will is Real, Harvard University Press, 2019.
  2. Rosenberg, A., How History Gets Things Wrong. The Neuroscience of Our Adiction to Stories. The MIT Press. 2018.


Leave a comment


When examined with sufficient detail, List’s ingenious efforts to save indeterminism as an emergent phenomenon are foundering (libertarians are right. If supervenience is accepted, free will is incompatible with determinism at any level). But examined in a little more detail, his efforts to save free will with the help of indeterminism and ontological emergentism (intentional, goal-directed agency, as a kind of downward causation), are not.


In other words, randomness is only one kind of indeterminacy. It has a constructive role in nature, as all biologists know. However, the most relevant kind of indeterminacy for emergentism is not randomness, but causal openness. New kinds of causal relations emerge as a result of new dynamics at macro levels, as Terrence Deacon wrote:

“Intentional, deliberate and often rational action is not a fiction, but a totally legitimate inhabitant of the natural world”. If we agree on that, it can’t be random.


New kinds of causal relations emerge as a result of new dynamics at macro levels

It is true. The human living machine is free because in its interior there are emergent causal relationships that do not occur in any living vegetable machine and that surely cannot emerge in any artificial machine either.

Gonzalo GénovaGonzalo Génova

Jesús Zamora is right in his criticism to Christian List.

However, if Zamora is right in his denial of free will, then his whole discourse becomes nonsense. “My goal in this new entry…” he writes. My goal??? Stones have no goals. Only free beings can have self-proposed goals.

Diogenes Laërtius already realized this 2.200 years ago.


To be fair, Zamora’s goals are his even in a deterministic universe, because he accepts (at least in this entry) List’s strong emergentism. He just says he couldn’t have others. If we accept (we dont have to) the compatibilist definition of emergence and free will, emergent causal relationships can be real without genuine novelty and choices can be own without alternative options.

We agree on his criticism of List’s argument about supervenience.

But Zamora is wrong when he says that, in an indeterministic universe, our actions can be deliberate, often rational and random at the same time. He contradicts himself because he has already accepted intentional, goal-directed agency and its causal role in nature. Indeterminacy AND downward causation make free will possible.

Gonzalo GénovaGonzalo Génova

To say that a deterministic being has goals is to pervert the original meaning of the word ‘goal’.

Having a goal is not the same as having a ‘tendency’. A stone has the tendency to fall, and it is truly ‘its’ tendency.

Having a goal implies the real possibility of an alternative option. Otherwise, it is void wordiness, nonsense. That’s what I meant.


I know and I agree. Compatibilist definitions pervert the meaning of the words. And it gets worse. Dennett definition of conscience is anything but conscience. I was just for a moment on his side of the playing field.

Leave a reply

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>