No Fixed Past
– A Compatibilist Reply to the Consequence Argument
Inget fixerat förflutet
– Ett kompatibilistiskt svar på konsekvensargumentet
Author: Elin Wengelin Grantén
Supervisor: Christian Löw
Department of Historical, Philosophical and Religious studies Master Thesis Philosophy, 30 hp
Abstract: Carl Hoeferʼs “freedom from the inside out” is a compatibilist account of freedom that claims that we can have a robust kind of freedom even if the world is deterministic. The key insight is that determinism belongs in tenseless B-series time in the block universe, not in our everyday notions of the world where time is experienced in a tensed way with a past, a present, and a future, where the past is viewed as fixed and the future as open. The block universe is time symmetric, i.e., determination can go in both temporal direction, and no particular part of the block universe is determined or fixed prior to any other part. Therefore, we do not have to think of deterministic logical relations as placing constraints on our actions in advance. Instead, we are free to view our actions as primary explainers that very partially determine both the future and the past. Since our actions are thought to place constraints on the microscopic past, it will be argued that Hoeferʼs account can question the Consequence Argument, the well-known incompatibilist argument that claims that if the past and the laws are not up to us, the consequences of the past and the laws cannot be up to us either, which includes our present actions. Hence, it will be argued that by rejecting that the past is not up to us, Hoeferʼs account can refute the Consequence Argument, which removes one possible threat to our freedom.
Abstract: Carl Hoefers “frihet från insidan ut” är en kompatibilistisk teori om frihet som hävdar att vi kan ha en robust sorts frihet även om världen är deterministisk. Den centrala idén är att determinism hör hemma i den statiska B-seriens tid i blockuniversumet, inte i världen såsom vi upplever den med ett förflutet, ett nu, och en framtid, där det förflutna upplevs som fixerat och framtiden som öppen. Blockuniversumet är tidssymmetriskt, vilket innebär att determinationen kan gå både framåt och bakåt, och ingen specifik del av
blockuniversumet är determinerad eller fixerad före någon annan del. Därför behöver vi inte tänka på de deterministiska logiska relationerna som någonting som i förväg begränsar hur vi kan agera. Vi är istället fria att betrakta våra handlingar som primära förklarare som är delaktiga i att determinera både det förflutna och framtiden. Eftersom Hoefer menar att våra handlingar begränsar hur det mikroskopiska förflutna kan vara, så kommer det att
Table of contents
Introduction...1
Hoeferʼs account...2
Time and the block universe...2
Bi-directional determinism...3
The constraints...5
Harmony...7
Downward causation...7
The Consequence Argument...8
Critique of the Consequence Argument...8
Using Hoeferʼs account to refute PAST...12
Some additional ideas...16
Possible objections...18
Merely an epistemic kind of freedom?...18
Only a conditional understanding of PAP?...20
Harmony and the micro-level...24
Could we make deterministic inferences?...25
Does using Hoeferʼs account to refute the Consequence Argument raise the price for compatibilists?...28
Final thoughts ...29
Introduction
In philosophical debates about free will, determinism is one of the central threats.
Determinism is often understood as the claim that the fundamental laws and a complete state of the universe at a time logically entail the complete state of the universe at all other times. Those claiming that we can be free even if the world is deterministic are referred to as compatibilists, since they argue that freedom and determinism are compatible, while those denying that freedom and determinism are compatible are called incompatibilists. Carl Hoeferʼs “freedom from the inside out”, is primarily a compatibilist account.
Why would determinism be a threat to our freedom? Deterministic fundamental laws and a past state of the world, usually the initial state of the universe, are often thought to determine everything that follows, including our choices and actions. According to this way of
reasoning, since neither the initial conditions nor the fundamental laws are up to us, the consequences of them cannot be up to us, therefore our choices and actions are not up to us, and we are not free. This idea lies behind the famous incompatibilist argument called the Consequence Argument. I believe that most people find the conclusion that our choices and actions are not up to us genuinely disturbing, so a way to dispel such a threat should be very welcome. The Consequence Argument has defined a lot of the philosophical debate in the past decades, and in some sense compatibilists seem to be obliged to show what is wrong with it. I believe that Hoeferʼs account of free will can offer a reply to it, because Hoeferʼs ideas seem capable of questioning that the past is not up to us. My purpose is therefore to analyse
whether Carl Hoeferʼs “freedom from the inside out” can contribute to the free will debate in a novel way. My main question is whether the objections to Hoeferʼs account and its response to the Consequence Argument can be addressed. My thesis is that Hoeferʼs account is a plausible compatibilist account of free will that succeeds in rejecting the Consequence Argument. If that is the case, it could be seen as an important advancement of the free will debate.
objections to Hoeferʼs account will be presented: Does Hoefer merely offer an illusionary epistemic kind of freedom where the ability to do otherwise is understood in conditional terms? How should the ideas about harmony be understood, and could deterministic inferences be available to us? Some worries about using Hoeferʼs account to refute the Consequence Argument will also be adressed. This will be followed by some conclusive remarks.
Hoeferʼs account
Our everyday experiences of the world tell us that past events have existed, that present events exist right now, and that future events do not yet exist. We also tend to believe that past events, unlike future events, can explain what happens here and now, assuming that the past is fixed, while the future is open. Hoefer aims to show that conceptions such as these have little to do with determinism, because determinism is a thesis about the world as it is described by fundamental physics, not the world as we experience it. To Hoefer, once determinism is properly understood, we can see that it is not a threat to our freedom.
Time and the block universe
The distinction between the so called A-series and B-series of time provides two different ways of ordering events (McTaggart 1908). A-series time captures time as we experience it, Hoefer claims, where there is a past, a present, and a future, and where the present constantly moves into the future. B-series time presents time as a static, linear ordering, where there is no past, present or future. Objects change in B-series time, since they can have one set of properties at time t, and another set of properties at another time, but time does not move or change in itself. (2002, p. 203) Since A-series time divides the world into a past, a present, and a future, it presents a tensed view on time, while B-series time is tenseless and orders events by their relative temporal relations to other events; there is no privileged “now”, instead events are ordered by dates or times that can be simultaneous, earlier or later than other events (Ney 2014, p. 148). According to Hoefer, in A-series time the future is
experienced as open, and we believe that it will be partly determined by our choices, while the past is viewed as fixed or already determined (2002, p. 205f). Even though such pre-theoretic notions are closely related to A-series time, they are not necessarily included in a tensed view of time without further arguments.1
In physics, Hoefer argues, the world is fully described in B-series terms (quantum mechanics might possibly be an exception), seemingly leaving A-series concepts of time, such as
“present” superfluous (2002, p. 203). Physics aims at describing the world in itself, and requires only B-series time, which Hoefer claims is partially isomorphic to, and underlies A-series time (2002, p. 220). So on Hoeferʼs view, A-A-series time merely reflects our way of thinking about time, but does not capture what time really is from the perspective of fundamental physics. This is not a wholly settled matter though, and there is no general agreement on whether tensed facts can be grounded in or be fully accounted for by tenseless facts in B-series time (Ney 2014, p. 148).
Combining space with B-series time, as in Minkowski space-time, gives us the so called block universe. This is a world with four dimensions – three spatial dimensions and one B-series temporal dimension. The block universe does not imply an ontological separation of past, present, or future parts of the world, instead past, present and future events are equally as real and exist equally as much. (Hoefer 2002, p. 204) This means that timeʼs passage has no ontological consequences, and is captured by a view called eternalism (Ney 2014, p. 143).
Bi-directional determinism
fundamental laws are deterministic and bi-directional, where the laws themselves do not pick out any specific boundary conditions (Thyssen & Wenmackers 2020, p. 11). Since no time slice has any privileged ontological status, neither any particular part of the block universe, nor any event or a certain set of events, is more responsible for determining our actions than any other part (Hoefer 2016a, ch. 6). This insight is thought to remove the deterministic threat to our freedom, since physics gives us no reason to pinpoint a specific part or state as
responsible for determining the rest of the universe.
Combining the thesis of determinism with A-series time and the idea that the past is fixed while the future is not, seems to entail that a pre-existing past has the power to bring about the not yet existing future. This is a conception that Hoefer rejects, claiming that it has no ground in fundamental physics (2002, p. 206). On such understanding, determinism seems to be about more than just logical entailment, suggesting that a fixed past somehow necessitates the future. Hoefer defines causal determinism as “the idea that every event is necessitated by antecedent events and conditions together with the laws of nature” (2016a). This is an understanding of determinism that Hoefer rejects, claiming that in recent debates, it is common practice to separate matters of causation from questions about the truth or falsity of determinism (2016a, ch. 1). Deterministic relations are about entailment, while causation is often seen as a metaphysical relationship that is not logical (Hoefer 2004, p. 101).
Determinism also seems to be silent on questions about causation – I could be the cause of future events, but still be fully constrained by deterministic relations (Hoefer 2016a, ch. 2.1). Ismael even argues that fundamental physics does not recognize any “fundamental,
If the direction of determination can go from the future, from the past, or from somewhere in the middle, oneʼs present action can be a part of a “now slice” that together with the laws entail the rest of the universe. Not just the choices and actions in one specific time slice can take part in determining the past and the future, though. Our choices can instead be seen as scattered over the parts of the block universe that contain agents, like a “starry sky of free choices”. (Thyssen & Wenmackers 2020, p. 15) Our actions “qua physical events” can then be seen as primary explainers that in a very partial way determine both the past and the future, where the relevant explanations are from our perspectives inside the block universe out, not from the outside – such as something in the distant past – in. (Hoefer 2002, p. 207) To say that our actions are qua physical events means that we focus on the particular role our actions have insofar as they are understood as physical events (Kutach 2014, p. 151). Not only our actions take part in creating the world; any event places constraints on how the rest of the world can be, Hoefer states (2002, p. 221).
The constraints
our actions have few, or even no necessary consequences towards the past. The constraints are merely that earlier microphysical states must be logically consistent with the microphysical state-type that a specific action supervenes on. (2002, p. 210f)
If a microstate is specified over a region of space and a time slice, its logical determination over the past and the future microstates is limited to so called light cones that symmetrically point to the past and to the future. Assuming that there is a restriction on the velocity of physical things, which follows from Special Relativity, these light cones have an extremely short lifespan. Hoefer makes an example where some specific microstates cover a spatial area in the block universe (a hypersurface) of 10 metres radius. Any of the microstates that can be found in this area can then logically determine the preceding and following microstates only over two shrinking spatial region shaped like away pointing light cones. It takes the same amount of time to reach the “tips” of the light cones as it takes light to travel 10 metres, and beyond these light cones, the area that can be logically determined by any of the microstates on this hypersurface ends. (2002, p. 211)
Hoefer claims that a typical materialist world view tends to be in A-series time and start at the Big Bang and then treat the world history as something that flows from the past to the future. This does not give us any answer to why the Big Bang turned out just like this. (2002, p. 221) Hoefer seems to say that since my actions, like all other physical events, take part in
determining the rest of the universe, they also take part in determining the Big Bang, i.e. the initial conditions, although in a very indirect way. Consequently, Hoefer appears to assume that our actions place logical constraints only on neighbouring microstates over extremely limited spatiotemporal regions, while our actions strangely enough also can take part in logically determining remote time slices such as the initial conditions. However, when discussing what an action in itself determines, it is about how it makes a direct difference to the world in its close vicinity. On Hoeferʼs account, it is a consequence of Special Relativity that local events have instant influence over finite domains, not over remote events (Thyssen & Wenmackers 2020, p. 15). I take it that this direct or instant influence differs from the indirect influence an action has over the deterministic logical relations that applies to the entire block universe. So to say that oneʼs actions take part in determining the initial
determine (almost) nothing about complete time slices such as the initial conditions.
Harmony
Hoefer claims that what matters is whether our freely chosen actions can fit together with each other and with the past as we know it macroscopically. Everyoneʼs free actions have to harmonise not only with their immediate pasts, but also with each other, forming a consistent world history. All actual choices must also be able to harmonise with all the alternative choices that we do not choose but believe that we could have chosen instead. All actual choices do in fact harmonise, because on the micro-level, determinism “holds sway over” every event that actually takes place. The worry is rather that our choices are more limited than we think – perhaps only one world history is possible. (2002, 213f)
According to Hoefer, our free actions do not causally bring about or explain macro-level features of the past. The world has temporally asymmetric features that make it very unlikely that our actions would leave traces on the macroscopic past, even though they can have a significant effect on the future. If each action logically entails merely that one of a possibly infinite set of microstates obtains for an extremely short time, and because of macro-level temporal asymmetries, our actions normally have no past consequences that are noticeable on the macro-level, then since each action hardly entails anything about the past, it seems
plausible that all our actual and alternative actions should be able to harmonise with each other, Hoefer concedes. (2002, p. 213ff)
Downward causation
from higher levels to lower levels, that is, things like intentions and choices can primarily explain our actions, instead of the explanations being “usurped” by what happens at the micro-level (Hoefer 2002, p. 207).
According to Pieter Thyssen and Sylvia Wenmackers, that Hoefer views our intentions, thoughts and choices as the primary explainers of our actions, suggests that Hoefer endorses some kind of non-reductive physicalism (2020, p. 25, n. 30). Hoefer does indeed seem to presuppose some kind of emergentism, where the correct explanations of our actions are in terms of our selves, our deliberations and decisions, not in terms of what goes on in our cells or among the fundamental constituents of the world (Hoefer 2016b). The thought might be that things such as intentions and choices are found on higher emergent levels, and even though they supervene on the micro-level, they cannot be fully reduced to or replaced by micro-level explanations.
The Consequence Argument
The Consequence argument is developed by Peter van Inwagen, and reaches the
incompatibilist conclusion that free will is not possible if the world is deterministic. The basic idea is:
If determinism is true, then our acts are the consequence of laws of nature and events in the remote past. But it’s not up to us what went on before we were born, and neither is it up to us what the laws of nature are. Therefore, the consequences of these things (including our present acts) are not up to us. (Vihvelin 2018, ch. 5)
This could be stated like this:
1) Determinism: Our acts are the consequence of laws of nature and events in the remote past. 2) PAST: It is not up to us what went on before we were born.
3) LAWS: It is not up to us what the laws of nature are. Therefore,
4) The consequences of these things, including our present acts, are not up to us.
Critique of the Consequence Argument
accept the conclusion. Lewis states the Consequence Argument in counterfactual terms, where the central idea is that, assuming determinism, there can be an act that I did not perform even though I have the ability to perform such an act. If I had performed that act, the remote past or the laws would have differed, and this appears to be absurd. Therefore, it seems that I lack the ability to perform the act that I did not perform. (Vihvelin 2018, ch. 5) Lewis distinguishes a weak thesis: “I am able to do something such that, if I did it, a law would be broken”, from a strong thesis: “I am able to break a law”. The weak thesis is supposed to be plausible, unlike the strong thesis. (1981, p. 115) To be able to act in a way that is not predetermined does not entail an ability to break a law, but merely that a law would have been broken if I had acted that way. The course of events just before the act would diverge slightly from the actual course of events, so a “divergence miracle” would occur, but my act would not have caused it, so the miracle would not be made up by my act, hence, I would not have the ability to break any law. (Lewis 1981, p. 116f)
Although Lewis relies on his own theory of counterfactuals, his critique against the Consequence Argument is not dependent on this particular counterfactual account, Kadri Vihvelin argues. The counterfactual version of the Consequence Argument shows that it relies on a counterfactual claim, because if determinism holds, then there either would have been a different past or different laws. Lewis suggests that if I had done otherwise, the past would have been the same until right before the alternative action, and the laws would have been slightly different. If counterfactuals instead should be evaluated by holding the laws fixed, the past would have been different back to the Big Bang, Vihvelin claims. (2013, p. 164) Barry Loewer favours the second alternative, arguing that the past (and the future) in our most similar worlds differ from the actual past (and the actual future), claiming that it is PAST that should be rejected (forthcoming, p. 12). Consequently, Loewer holds that in a possible world where an alternative action takes place, the world history all the way back to the Big Bang would differ (forthcoming, p. 16).
According to Vihvelin, an alternative possible world is not wholly identical to ours, so evaluating counterfactual worlds with false antecedents must either violate LAWS or PAST. Loewer claims that this observation alone leaves the Consequence Argument unsound. (Loewer forthcoming p. 10, n. 22) What seems to motivate this kind of reasoning is the idea that in a deterministic world, there is only one possible course of events, so an alternative world must differ from our actual world; if a counterfactual world had laws and a past micro-history that were identical to the laws and the past micro-micro-history in the actual world, then these worlds could not have different futures. I would not say that this necessarily means that the Consequence Argument is unsound, because if the ability to do otherwise would be exercised, then the laws or the past would differ from the actual laws or the actual past. However, if determinism has the consequence that only one actual course of events is possible, this ability could not be exercised in the actual world. Then it can be questioned whether this ability to do otherwise is enough to refute the Consequence Argument.
Loewer turns to the not entirely uncontroversial “Mentaculus”, a reductionist approach to the arrow of time, which tries to ground several temporal macro-level asymmetries in the
fine-grained complete description of a systemʼs physical state at time t, while a coarse-fine-grained description of the microstate of a system is a macrostate. Each microstate can be found in some particular macrostate, and a macrostate includes many microstates. (Ismael 2016, p. 141) The complete microstate of the universe at time t describes the fundamental state, i.e. the positions and the movements of all the fundamental particles in this time slice (Loewer forthcoming, p. 14). Entropy belongs on a macro-scale, where a high-entropy state is microscopically more coarse-grained than a low-entropy state. Starting from the
law-governed micro-level and the PH, we can get macro-level probabilities, because in a universe that starts in a low-entropy state, it is extremely likely that the entropy will steadily increase until it reaches a state of maximal entropy, where the universe will more or less remain. (Ismael 2016, p. 142)
Microstates evolve deterministically while macrostates evolve indeterministically due to statistical mechanical probabilities, so there is a “branching forks structure” on the macro-level, Loewer claims. Since macrostates can be realized by many different constellations of microstates, within an isolated system, if the macrostate is not at maximum entropy, then most microstates compatible with this macrostate will evolve into macrostates with higher entropy. There is an infinity of macroscopically indiscernible initial conditions, and at any time they can diverge into different macroscopic branches. Microstates that have been macroscopically distinct for a short time can also converge again. Two macroscopically indiscernible micro-histories can thus diverge into two different branches when an agent makes a choice, where one branch realizes the actual choice. Hence, that there is a certain macroscopic state at time t cannot determine oneʼs choice the next moment, because even though it is determined at the micro-level which branch realizes the action in question, on the macro-level, alternative decisions seem possible. If an alternative decision had been made, the worldʼs micro-history would have been different, leaving one with the ability to falsify the micro-history, Loewer claims. (forthcoming, p. 14ff)
The Mentaculus supports an account of influence counterfactuals, since it can account for the asymmetry of influence – why we regard the future as open to our influence, while the past seems closed, Loewer states (forthcoming, p. 18). In the Consequence Argument, the meaning of the phrase “up to us” is not entirely clear, but Loewer takes it to mean that we have
counterfactually depends on my decision. Unlike causal relations, where a cause precedes its effect, a decision can influence a state of affairs that occurs prior to the decision itself, Loewer claims. (forthcoming, p. 4f) PAST could then be stated like: “We cannot influence what went on before we were born”, where the conclusion of the Consequence Argument could be that we cannot influence the consequences of PAST and LAWS, so we cannot influence our present acts.
Loewer argues that decision counterfactuals show the probabilities of different outcomes of a certain decision. Alternative decisions take place in worlds with different micro-histories, thereby influencing the micro-history, making it possible to refute PAST. But our influence over the future is much greater that our influence towards the past. (Loewer forthcoming, p. 16) Suppose I decide to eat some of the ice cream I have at home. Then the probability that there is less ice cream in my freezer 30 minutes later is very high. If I had chosen not to have ice cream, the probability that there is less ice cream in my freezer 30 minutes later is instead very low, all other things equal. Due to the temporal asymmetry of these conditional
probabilities, my influence over the amount of ice cream in my freezer runs towards the future, but in a counterfactual scenario where I decide not to have ice cream, the micro-history of the world would differ from the actual micro-micro-history, so I would have influenced the past, even if this influence is negligible to me.
Since the influence oneʼs decisions have over the past is not noticeable to the agent, it does not amount to control. Immediate control is instead taken to be the ability to freely choose among alternative possibilities, independently of the rest of the world, Loewer claims. If the world is deterministic, the idea that we are agents with such immediate control seems to be mistaken. However, by replacing the idea of immediate control with the notion of influence, one can be said to have influence over events that take place at other times. Oneʼs decisions influence the movements of oneʼs body, and these movements will influence physical events in the world, while oneʼs alternative decisions take place in worlds where the microscopic future and past differ from the actual world. (Loewer 2012, p. 127)
Using Hoeferʼs account to refute PAST
On Lewisʼ account, if I had exercised my ability to do otherwise, the laws or the past would have been different, but this merely follows from Lewisʼ attempt to refute the Consequence Argument, and Lewis does not seem to offer any independent reasons for why we should accept this.
Neither Loewer provides positive reasons for why we can influence the past. Loewer does indeed rely on the Mentaculus, and explains this influence in terms of decision
counterfactuals, but the idea that alternative actions can influence the past does not appear to be a central part of the Mentaculus, and referring to decision counterfactuals does not offer many details about our influence over the past and how exactly it is supposed to manifest. Furthermore, since alternative decisions take place in alternative worlds where the
microscopic past differs from the actual microscopic past, is our assumed influence over the past really an interesting kind of influence? Just like on Lewisʼ account, the ability to
influence the past seems to be open to the charge that if this ability can never be exercised in the actual world, is it then enough to ground our freedom?
Since Hoefer argues that our actions can determine the past (as well at the future), it appears that he could question PAST. This is exactly what Thyssen and Wenmackers claim that Hoefer does, since Hoefer seems to say that a present condition together with the laws can put
constraints on the past and the future, so our actions can indirectly and partially determine the initial conditions (2020, p. 14). Hoefer does not explicitly discuss the Consequence
Argument. However, since the Consequence Argument concludes that freedom is not
compatible with determinism, it seems reasonable that Hoeferʼs compatibilist account should be able to respond to it. Unlike Lewisʼ and Loewerʼs accounts, Hoeferʼs account provides positive reasons for why determination can go from the future to the past, independent of any attempt to refute the Consequence Argument. These positive reasons come from the
Since Hoefer, just like Loewer, argues that the determination towards the past concerns the micro-level, and is not noticed by us, Loewerʼs thoughts about how influence differs from control or causation seem relevant. Hoefer clearly states that the logical constraints our actions place on how the past can be does not amount to causation. To refute PAST, it seems “up to us” must be understood in a way that does not require that the agent consciously and intentionally can control or cause past events. The limited constraints our actions place on the most recent microscopic past seem possible to understand as some kind of influence. By itself, this does not seem to be enough to refute PAST, if it is the distant past that is under consideration. However, if indirectly all our actions take part in logically determining the rest of the universe, it seems that they can be said to have some (almost) negligible influence over the distant past as well, thus refuting PAST.
Hoeferʼs and Loewerʼs accounts have some additional similarities. Loewerʼs ideas about macro-level freedom resemble Hoeferʼs, because Hoeferʼs understanding of the relation between the micro-level and the macro-level also seems to be inspired by statistical
mechanics (Thyssen & Wenmackers 2020, p. 12). That Loewer states that a macrostate can be realized by many different constellations of microstates, appears to merely be a slightly different way of expressing Hoeferʼs idea that there is a large number of microphysical state-types capable of serving as a supervenience base for an action. Hence, both Loewer and Hoefer view freedom as a macro-level phenomenon, because choices and actions are macroscopic events, and the same macroscopic history can encompass several different present choices, but our free choices are still grounded in the micro-level, where different choices entail different microscopic histories. The Mentaculus might also have the potential to help Hoefer by further removing the potential worry that his account proposes an
objectionable kind of backward causation, because the temporal asymmetry of causation can be grounded in the Mentaculusʼ probabilistic correlations (Loewer 2012, p. 132). This could explain why we cannot cause events in the macroscopic past, even though our actions place constraints on the past micro-history.
actual world, where oneʼs counterpart can act differently (2002, p. 214f). However, Hoeferʼs account does not primarily seem to be a counterfactual account, and it seems to be capable of refuting PAST without relying on any counterfactual reasoning. That Hoefer uses some counterfactual examples might simply be a way to illustrate his point. After all, Lewis developed a counterfactual version of the Consequence Argument, making way for
understanding the ability to do otherwise in a possible world setting. And generally, stating alternative possibilities in counterfactual terms seems to be a common philosophical practice among compatibilists. One might of course say that the alternatives that never become actualized in our world take place in similar possible worlds with false antecedents, but this does not seem to be the main idea that Hoeferʼs account rests on. Instead, what makes it possible for Hoeferʼs account to question PAST is the idea that our actions, seen as physical events that are primary explainers, determine both the future and the past. By acting, and thereby actualizing one of several possible actions, the agent very partially participates in creating the actual world history, and all the alternatives turn into unactualized possibilities. Also, Loewer claims that immediate control is the ability to freely choose among alternative possibilities, independently of the rest of the world, and that this is not possible if the world is deterministic (2012, p. 127). On Hoeferʼs account, not only can we influence the macroscopic future, it seems that it can be “up to us” in an even stronger sense, where we actually can cause or control future events, even though we only have a very limited influence over the past.
potential threat to our freedom, might at least partially be motivated by the fact that we cannot cause past events. Bi-directional determinism on the micro-level in the block universe does not concern a level of the world that we experience, in fact, it completely goes against our pre-theoretic intuitions about time, which seems to challenge whether it is appropriate for the Consequence Argument to appeal to our intuitions about what is absurd.
Some additional ideas
I will use some thoughts by Jenann Ismael. Even though they do not explicitly take part of Hoeferʼs account, I find that they can help illustrate and further motivate some of Hoeferʼs points. That some of Ismaelʼs ideas can be relevant seems to be further supported by the fact that Hoefer himself finds Ismaelʼs attempt to reconcile determinism with free will “broadly similar” to the ideas underlying “freedom from the inside out” (2016b). According to Ismael, Minkowski space-time, which is associated with Special Relativity, is hospitable for
determinism. The spatiotemporal structure in Minkowski space-time is a light cone structure, defined by the speed of light, because nothing can travel faster. (See picture 1.) Points located in the absolute elsewhere cannot affect what happens at p, or be affected by it. A temporal cross-section somewhere in an eventʼs past light cone, or the past light cone of this event, would show a set of events sufficient to determine this event. However, an event can be seen as fixed by its causal past only from a retrospective view in Minkowski space-time. From a prospective view, the laws instead allow the causal past to extend in a large number of ways.2
(2019, p. 484f)
Picture 1: “Light cones in Minkowski space-time” (Ismael 2019, p. 484).
It is tempting to think that a retrospective view of an eventʼs causal past reveals how everything already was, Ismael claims. Just like it is easy to think that there can be future information “already there” just waiting to intersect a future light cone; that there are events that have already occurred even though information about them is not yet here. These are conceptions that are not supported by Minkowski space-time. What takes place within an eventʼs past or future light cone has a causal order, but there is no such order in the absolute elsewhere, relative to the specific event in question. (2019, p. 485ff)
We tend to think of space as containing fixed objects that were already there before we experienced them, while time is unfolding as it is experienced. Relativistic physics removes the pre-theoretic difference between space and time; space is not a substance while time is a dimension of becoming, instead the spatial past has a dimension of becoming just like the temporal future. (Ismael 2019, p. 487) To say that a future action exists already implies that it exists anyway, independently of me, but the future is dependent on what I do, so my future actions cannot be said to be fixed in advance (Ismael 2016, p. 149f). Ismael admits that if we assume that the universe is deterministic as a whole, all events in the absolute elsewhere are also determined, but this makes it difficult to see the causal structure. What is important is whether there are deterministic facts that put constraints on our actions, but from our
perspective, as beings embedded in the world, events in the absolute elsewhere are nowhere. To use temporal language and claim that things already are fixed, makes sense only from the perspectives of beings embedded in time, but in Minkowski space-time there is no such point of view. That the whole universe is determined should instead be seen as a fact that emerge from the sum of everything that happens “on the ground”. (2019, p. 488)
are therefore not part of the ground level facts, so they are not localized in space or time, and cannot bring about events or keep events from occurring. (2016, p. 176f)
Hoefer does not specify whether he has a Humean or a non-Humean understanding of laws, which makes me inclined to think that “freedom from the inside out” is supposed to be an interesting kind of freedom regardless of what view on laws one endorses. Hoeferʼs point instead seems to be that the deterministic threat that Ismael describes presupposes that the past is fixed and exists before the future, which is a picture that lacks support if B-series time and eternalism are assumed. According to Thyssen and Wenmackers, the lesson from Hoefer is that even if the laws determine the course of events in the world, they do not by themselves constrain any given instant, since they neither determine their own boundary conditions, nor determine where a boundary condition should be drawn, so the same laws can be compatible with different courses of actions. The freedom on Hoeferʼs account thus lies in the fact that no specific boundary conditions are privileged. (2020, p. 14) So while Ismael argues that the laws are not fixed in advance, and cannot keep events from happening since they are not localized in space or time, Hoefer holds that no boundary conditions are fixed beforehand, because physics does not single out any specific time slice in the block universe that is immanently privileged in determining the rest of the world. Hoefer seems to assume that the fact that no part of the block universe is fixed in advance is enough to dispel the deterministic threat to our freedom, so there is no need to also question the fixity of the laws (2016b).
Possible objections
Merely an epistemic kind of freedom?
Hoeferʼs account can be interpreted in different ways, and I will outline both an epistemic and an ontic reading. The epistemic reading has much in common with more traditional
understandings of determinism, and Thyssen and Wenmackersʼ interpretation of “freedom from the inside out” can be seen as an example of such a reading. I will understand an epistemic reading as an interpretation that ultimately claims that Hoeferʼs freedom is no genuine freedom, it only appears to be due to our epistemic limitations. Consequently, I would say that on an epistemic reading, Hoeferʼs freedom is quite uninteresting. I will instead try to show that Hoeferʼs account can be interpreted in a stronger, more original and
offer an interpretation where Hoeferʼs freedom is real, and objectively exists in the world.
According to an epistemic reading, a past time slice, such as the initial conditions, together with the laws, predetermines the course of events in the world. If determinism is
bi-directional, a complete future time slice plus the laws, or a complete “now slice”, would also determine all the other time slices, but then the notion that everything is predetermined would become difficult to uphold. Although, since any complete time slice and the laws are thought to place constraints on how we can act, a past slice, such as the initial conditions, and the laws actually do determine all the other time slices, which could be said to show that the future nevertheless is predetermined. One example that could be said to capture such epistemic interpretation is:
[T]here will always be a set of alternative courses of events (histories and futures) consistent with the currently perceived state of the universe. […] this follows simply from our subjective
perspective and limited cognitive capacities, and in no way challenges the fact that the given initial conditions and deterministic laws of nature entail only one possible (actual), real course of events. (Pernu 2017, p. 5)
Only a complete time slice of the entire universe, plus the laws, can logically entail all the other time slices, so when it comes to our actions on the macro-level, determinism might be seen as a weak or irrelevant thesis that we can simply disregard. An epistemic reading of Hoefer would disagree. When thinking about determinism we should accept Hoeferʼs statement that determinism belongs in the block universe in B-series time, because then the fact that local matters are most relevant to us is merely about our limited perspectives. This does not make us entitled to ignore that determinism does constrain our actions, which challenges the idea that we are free agents. Hoefer does privilege causal macro-level
explanations, so in some sense, “freedom from the inside out” does speak to our interests and purposes, since we are interested in what can cause or explain our actions on the macro-level that matters to us. This leave us free to ignore the deterministic micro-level in the block universe, Hoefer claims. (2002, p. 207f) Hoefer also admits that his account has an idealistic tendency, since it involves prioritising agents over microstates covering spatiotemporally extended regions (2002, p. 221). Hence, Hoefer appears to neglect to properly take the deterministic constraints into consideration, only offering a shallow, epistemic kind of freedom.
On an ontic reading of Hoefer, the claim that determinism belongs in the block universe instead is a way to show that notions such as “in advance” do not make sense when B-series time and eternalism are assumed. If my actions, like all other events in the universe, are not fixed in advance, we might actually be entitled to ignore deterministic logical relations on the micro-level, since they do not place constraints on our actions beforehand, and instead turn to causal macro-level explanations of our actions that seem more relevant for our interests and purposes. An ontic reading of Hoefer therefore succeeds better than an epistemic reading in fully capturing one main idea on Hoeferʼs account – that our actions are primary explainers from which the direction of determination can originate. An epistemic reading seems to have difficulties fully giving up an A-series conception of time and closely related ideas about certain states of the world being fixed prior to other states.
Only a conditional understanding of PAP?
captured by PAP: The principle of alternative possibilities (Thyssen & Wenmackers 2020, p. 2). Christian List claims that a standard incompatibilist argument goes like this:
Premise 1: A necessary condition for someoneʼs action to count as free is that the agent can do otherwise.
Premise 2: Determinism implies that the agent cannot do otherwise.
Conclusion: Either there are no free actions, or determinism is false (or both). (2014, p. 156)
There are several understandings of what it means to be able to do otherwise, but I will focus on conditional and modal interpretations. A conditional interpretation can say: “If the agent had tried to do otherwise, then the agent would have succeeded”. A modal interpretation can be that: “It is possible for the agent to do otherwise (there are forks in the road)”. (Thyssen & Wenmackers 2020, p. 15) Incompatibilists tend to think that only a modal interpretation fully captures a relevant form of freedom. List, for instance, claims that the ability to do otherwise “comes too cheap” if this ability is understood in conditional terms (2014, p. 159).
By only holding the macroscopic history fixed, the freedom we usually believe that we have is possible, Hoefer claims (2002, p. 214f). This focus on the macroscopic history seems to be what motivates Thyssen and Wenmackers to assume that the kind of alternative possibilities discussed by Hoefer are conditional (2020, p. 13). Thyssen and Wenmackers state that Hoefer implies that we can have counterfactual beliefs; we believe that we could have done otherwise while the macroscopic history is the same, even though determinism entails that we actually could not. (2020, p. 26) Thyssen and Wenmackers thus exemplify an epistemic reading of Hoefer, where deterministic constraints rule out all alternative actions. The weak micro-level constraints that Hoefer describes, that a specific action could supervene on a possibly infinite number of microphysical state-types, then turn into something purely hypothetical, reflecting the belief that if the agent was in a different microstate, then she could have acted otherwise, even though the agent could not have been in a different microstate. These counterfactual beliefs reflect our lack of knowledge about which microstate a specific action supervenes on, Thyssen and Wenmackers claim (2020, p. 12). This appears to go against Hoeferʼs idea that his account captures the freedom we usually think we have, because even though our
actually could not, then I would not be free.
If the belief that at time t one could have acted differently is illusionary because we only have knowledge of the coarse-grained macro-level, then it seems that our actions are primary explainers only in an illusionary sense as well, because if our actions are predetermined, they would not genuinely be able to partially determine the past and the future. Moreover, if we only believe that we are free because we lack knowledge of the deterministic constraints, then it would not seem to matter whether determinism belongs in A-series time or B-series time, or whether there is some part of the universe that is fixed in advance, because these believes merely reflect our limited epistemic faculties, not the world itself. On an epistemic reading, Hoeferʼs ideas about time and the block universe therefore appear to lack relevance.
Like Thyssen and Wenmackers, Loewer also seems to think that determinism places
constraints on our actions, claiming that a modal interpretation of PAP is incompatible with determinism, because it states that the only important understanding of possibility is physical possibility, but if determinism holds, there is only one physically possible choice at any given time (forthcoming, p. 4). Loewer instead favours a conditional understanding of PAP, where if I had chosen differently, then the worldʼs micro-history would have been different.
determined, which an epistemic reading of Hoefer implies, knowledge of all the microscopic details would render the macroscopic freedom merely apparent, and our alternative actions would be possible in similar alternative worlds, but not in our world.
If Hoefer merely offered a conditional understanding of being able to do otherwise, the freedom he posits would not be that interesting. However, the epistemic reading can be challenged. On an ontic reading of Hoefer, our alternative choices are hypothetical, in the sense that they do not take place, but the only thing that keeps them from taking place is that the agent does not choose them. Alternative actions are instead physically possible in our world at the time when the agent acts, and are not merely illusionary beliefs due to our epistemic limitations. The idea of being able to do otherwise only makes sense in retrospect; yesterday I could have acted otherwise, and then have taken part in creating a slightly
different world than the one I experience today. At the specific point when I act, there are only different possibilities; prospectively there are just different possible actions where no action carries any special status of being the determined actual action. From an embedded
perspective inside the universe, there is no predetermined course of events that constrains our actions, nor is there a fully fixed microscopic past. An ontic reading thus states that Hoeferʼs account is compatible with a modal interpretation of PAP, because even on the micro-level there are no deterministic constraints, only the weak almost trivial local constraints Hoefer discusses, so the direction of determination can originate in our actionsʼ supervenience bases. Then it seems that we could have the kind of freedom we usually think we have, and this could indeed be the kind of particularly robust compatibilism that Hoefer claims that his account entails (2002, p. 203). This is also what makes Hoeferʼs macro-level freedom more robust than Loewerʼs, since it can resist the charge that the macro-level freedom in question is merely apparent.
Second, the very fact that the microscopic past is not constant or fixed in advance is what makes it possible to have a modal understanding of PAP, even though determinism logically entails our actions. Since physics gives us no reason to hold the microscopic past constant, we are not forced to assume that alternative actions could never have been actualized in our world. Unlike the branching forks structure that Loewer describes, where the actual micro-level is fixed, an ontic reading of Hoefer instead suggests that at the time when I face a choice, the macroscopic future in the actual world could branch into several different
alternatives, where all the alternatives supervene on different microstates, and where different choices also logically entail slightly different microscopic pasts.
Harmony and the micro-level
If an ontic reading of Hoefer claims that one can do otherwise in a modal sense, then why does Hoefer focus on macro-level freedom instead of some kind of fine-grained freedom that holds on the micro-level as well? Assuming determinism, the microphysical state-types that an action can supervene on have to be logically consistent with the past microstates in an extremely limited spatiotemporal region. Suppose that being able to act freely required that the immediate microscopic past had to be logically consistent with, i.e. must be able to harmonise with, all the particular microstates that each of my possible actions at time t supervene on. A requirement of such harmony on the micro-level seems strange and
misplaced. On an ontic reading of Hoefer, if I had acted differently at time t, the microscopic past would be slightly different too, so a different action would not need to harmonise with the same microscopic past as the original action. This is not how we experience the world though, because on the macro-level there are temporal asymmetries that prevent our actions from having any significant consequences towards the past, which might be associated with problematic backward causation. The need for harmony is something that arises on the macro-level that we experience, because in order to conceive of ourselves as free we want the same macroscopic past to be able to contain different future alternatives.
Freedom then seems to be a macro-level phenomenon, independent of details on the micro-level. When we make choices, we do not choose to actualize specific microstates, Hoefer points out. (2002, p. 211). The many-one relationship between the many microscopic state-types that can realize one macroscopic action is what can make it possible for us to
experience a consistent macro-history: Since there is a possibly infinite number of
macro-level, without there being any logical inconsistencies on the micro-level. So even though we do not know exactly which microphysical state a specific action supervenes on, due to our epistemic limitations, this does not necessarily mean that the macro-level freedom is
illusionary. If my action is not fixed in advance, then even though determinism entails that my action supervenes on only one microstate, this microstate can become fixed by me at the time
when I act.
Could we make deterministic inferences?
Hoefer claims that the focus on macro-level freedom removes the traditional threat that the actual microscopic history of the world determines the present (2002, p. 216). Still, Hoefer concedes that the worldʼs actual micro-history does determine oneʼs present act (2002, p. 216). This might contradict the ontic reading where the supervenience base for a specific action is not fixed beforehand, possibly leaving Hoeferʼs macro-level freedom merely epistemic after all.
Thyssen and Wenmackers seem to rely on the possibility that we could in principle hold a time slice fixed, claiming that if we are free to choose any time slice as a boundary condition, we could choose the initial conditions. Together with the laws, the traditional deterministic threat would then be intact, since everything that follows would be predetermined. (2020, p. 15) Following Ismael, the view that the laws and the initial conditions, or the actual micro-history of the world, determine everything else comes from a view sub specie aeternitatis – an expression that comes from Spinoza, literally saying “under the aspect of eternity”, which is about the world being represented from a temporally transcendent perspective (2016, p. 183). This kind of perspective presents the entire universe as fixed. From an embedded perspective inside the block universe, the deterministic world instead emerges from all the events that actually take place in it. The future is then open, since it is not fixed in advance, and an action can objectively take part in creating the deterministic world. An embedded perspective and a view sub specie aeternitatis thus reveal the world from different viewpoints. In some sense, this seems to resemble Hoeferʼs point that determinism belongs in the block universe in B-series time, where the entire universe indeed is determined, not in the tensed A-series time where we experience the future as open, because from the perspective of a being existing in time, future events have not yet happened, and do not yet exist, so the
deterministic relations that emerge from events inside the universe are not yet “completed”.
So, sub specie aeternitatis, my present action is always determined, to the extent that “always” is applicable to a perspective outside of time and space. From an embedded perspective inside the universe, my present action becomes determined because I act this way; my action exists at a particular point in space-time, and I can bring it about at that specific point. Consequently, logical deterministic relations are not accessible to someone inside the universe, since she takes part in forming the very events that these logical relations emerge from. Ismael also notes that information about the worldʼs total state at a time in the past contains information about the future (Ismael 2019 p. 481). I take this to mean that a complete past state of the world is logically entailed by future events, such as our future actions. This is something that also seems to follow from Hoeferʼs account, because if the direction of determination can go both towards the past and towards the future, then events in the past seem to partially be logically entailed by future events. So, information about a complete past state of the universe is not even in principle available to an embedded
sub specie aeternitatis everything just is. If a complete past time slice is supposed to
determine the future, it has to be specified that it constitutes the total state of the world, and such information is not available to an embedded subsystem, Ismael claims (2019, p. 483). A stipulation that there will be no external influences from something outside the universe would have to be added, and a stipulation like that is not immanent in the world itself (Ismael 2016, p. 190, n. 4).
Hence, Thyssen and Wenmackers are right, because the initial conditions and the laws do entail the rest of the universe, just like Hoefer is right that the actual micro-history of the world at time t, together with the laws, determines oneʼs subsequent act(s). By applying Ismaelʼs ideas, it becomes visible that they are right only from a view sub specie aeternitatis, but this can be consistent with the idea that our actions are genuine primary explainers from an embedded point of view. However, Thyssen and Wenmackers are not justified in claiming that we could single out the initial conditions, and everything that follows would be
predetermined. Does this show that the definition of determinism is misleading or problematic somehow? No, since it appears to be silent on questions about perspectives or epistemic issues about what kind of information we can have access to as beings embedded in the universe, and only really reveals logical relations.
Ismael has pointed out that if the entire universe is determined, then everything in an eventʼs absolute elsewhere is determined as well. So, even if we cannot make deterministic inferences as beings embedded in the world, why could determinism in itself not just as well go “from the outside in”, where the entire universe is fixed and our embedded perspectives merely unfold according to an always existing “manuscript”? Where there is an immanent logical relation between all the time slices that somehow precedes the unfolding of the universe. I suppose that might be possible, but if the laws themselves do not determine anything without a specific boundary condition and they are compatible with other courses of events as well, and no particular state of the universe is fixed prior to any other state, then the motivation for thinking that the entire universe is fixed beforehand does not seem to come primarily from the thesis of determinism. Instead, it might be about some kind of philosophical fatalism,
Does using Hoeferʼs account to refute the Consequence Argument
raise the price for compatibilists?
The Consequence Argument states that our acts are consequences of “events in the remote past” that are not “up to us”, but does not specify whether it is micro-level events or macro-level events that is in question. I find that the most straightforward reading is to assume that what is in question are events on the macro-level, that is, events that we can perceive. Since Hoeferʼs account rejects that our actions influence the macro-history, it would not be able to refute PAST understood in terms of macro-level events. Would this mean that using Hoeferʼs account to refute PAST presupposes an implausible understanding of the Consequence Argument? On the contrary, because only a microscopically complete time slice of the block universe and the laws can logically entail the future. The macroscopic past and the laws do not determine the future. Consequently, if PAST is concerned with events on the macro-level, the Consequence Argument is invalid. If it instead is the microscopic past that is in question, the Consequence Arguments is valid, but then one could use Hoeferʼs account to refute PAST, since the temporal symmetry on the micro-level makes it possible for influence to go in both directions. Hence, the Consequence Argument might be an equivocation in another sense than the one Lewis suggests, and only seems to be valid if PAST somehow concerns the
microscopic past, but then it can be refuted.
If the influence one of my actions has over the distant past is almost negligible, using Hoeferʼs account to refute PAST might force the compatibilist to rely on a watered down notion of influence. The extremely limited influence we have towards the distant microscopic past seems to be something accidental, a mere by-product of an act. However, normally we do not demand that acting freely must entail that we are aware of all the fine-grained
but it seems that it could also refer to a strictly logical influence.
If we instead had more substantial influence over the past, the threat from some objectionable kind of backward causation might arise. Hoeferʼs idea that the same macroscopic history can lead to different choices could then turn out to be problematic, because if our actions could make substantial changes to the past micro-history, then these might at least sometimes be noticeable on higher levels, and we could possibly detect problematic cases of backward causation. That our actions take part in forming deterministic logical relations that emerge from everything that happens inside the universe does not mean that they cause or change anything in the past – that would seem to imply that the past first was in a certain way, then some later action could change this fact, which indeed would be problematic. These logical relations instead concern how everything fits together in the entire universe, viewed sub
specie aeternitatis. From our embedded perspective, it is not about our actions being able to
change the past, rather, when we act, our actions take part in forming logical relations that span over the entire universe, which includes the parts that have not yet happened to us, and the parts that we already have lived.
Final thoughts
What is novel about Hoeferʼs account is the explicit claim that bi-directional determinism belongs in B-series time in the block universe, where no part of the universe is fixed prior to any other part, i.e., there is no particular boundary condition. This does not seem to be emphasized very often in the free will debate, and allows us to view our actions as primary explainers that affect the future and place constraints on the microscopic past. Consequently, the microscopic past is not fixed prior to our actions and it does not have any special role in determining our actions. If determinism does not place constraints on our actions, we can instead focus on the temporally asymmetric macro-level that is relevant to us, and assume that our actions are caused by our intentions and choices, instead of being determined by what goes on at the micro-level. This does presuppose some kind of downward causation, though.
freedom can be understood as the ability to do otherwise in a modal sense, where I can make different choices in the actual world, but each action entails a slightly different microscopic past. PAST, that it is not up to us what went on before we were born, can thus be refuted. One reason that seems to have motivated PAST in the first place is that determinism is so easily conflated with causation. Even though it might seem a bit like exploiting a loophole to claim that one can influence the past when it is only about oneʼs actions placing logical constraints on the microlevel, I have shown that if PAST is supposed to concern the macro-level, the Consequence Argument is not valid. It could also be objected that the influence we can have over the past on Hoeferʼs account is a watered down kind of influence that is not significant. I have responded that even though the influence we have over the past is very limited, what matters might be that it stems from our free actions, and if our influence over the past was more substantial, and not just about logical entailment, then we might instead face the problem with backward causation.
An ontic reading of Hoefer appears to be able to survive the charges that the freedom in question is merely epistemic, and that the freedom we have to do otherwise is only conditional. One thing that speaks in favour of an ontic reading is that Hoeferʼs account becomes more coherent. On an epistemic reading it is difficult to see how Hoeferʼs rejection of A-series time together with a fixed complete (past) state of the universe is fully taken seriously, because without a fixed boundary condition, the laws by themselves do not seem to be enough to determine anything, which seemingly undermines the central idea that motivates an epistemic reading of Hoefer – the idea that there are deterministic constraints that leave only one course of events possible. It could also be questioned how Hoeferʼs central idea that our actions are primary explainers would be compatible with the existence of such pre-existing deterministic constraints, because then we would be able to do otherwise merely in a conditional sense, and the freedom that Hoefer posits would not be as robust as Hoefer argues that it is.
laws, seem capable of further strengthening Hoeferʼs idea that our actions are primary explainers. In addition to Hoeferʼs idea that no part of the block universe has any privileged role in determining the rest of the universe, Ismaelʼs ideas can help to further rule out the threat that we nonetheless potentially could hold a past time slice of the block universe fixed, and then everything that follows would be predetermined. Ismaelʼs ideas that the
deterministic universe emerges from what happens on the ground indeed seem to point in the same direction as Hoeferʼs idea that our actions, and all other events, partially determine the rest of the block universe “from the inside out”, which raises the question how we could gain knowledge about a complete time slice if it is partially entailed by events that have not yet happened. However, Ismael also makes the distinction between an embedded perspective and a perspective sub specie aeternitatis, which further makes it clear why we, as being embedded in time, could never gain access to such knowledge.
I have tried to show that Hoeferʼs account could refute the specific incompatibilist threat that is expressed by the Consequence Argument. Merely refuting the Consequence Argument does not imply that all potential threats from determinism are necessarily disarmed, and Hoeferʼs account does not eliminate all possible challenges to the idea that we can act freely. Hoefer is committed to several theoretical standpoints about such things as the block universe, bi-directional determinism, emergence, etc., which could turn out to be mistaken, and there are rival views, so Hoeferʼs account of freedom will not persuade everyone. I have merely tried to show that Hoeferʼs account can be seen as plausible, and Hoeferʼs main standpoints do not seem to be highly controversial or incompatible with current science in any obvious ways.
References
Elzein, Nadine and Tuomas Pernu (2017): “Supervenient Freedom and the Free Will Deadlock”, Disputatio, Vol. IX, No. 45.
Hoefer, Carl (2002): “Freedom from the Inside Out”, in Callender, C. (ed.), Time, Reality and
Experience, Cambridge University Press.
Hoefer, Carl (2004): “Causality and Determinism: Tension, or Outright Conflict?”, Revista de
Filosofía, 29:2, 99-115.
Hoefer, Carl (2016a): “Causal Determinism”, The Stanford Encyclopedia of Philosophy , Edward N. Zalta (ed.),
https://plato.stanford.edu/archives/spr2016/entries/determinism-causal/
Hoefer, Carl (2016b): “Review: How Physics Makes Us Free”, Notre Dame philosophical
reviews, https://ndpr.nd.edu/reviews/how-physics-makes-us-free/
Ismael, J.T (2016): How Physics Makes Us Free, Oxford University Press, New York.
Ismael, Jenann (2019): “Determinism, Counterpredictive Devices, and the Impossibility of Laplacean Intelligences”, The Monist, 102, 478–498.
Kutach, Douglas (2014): Causation, Polity Press, Cambridge.
Lewis, David (1981): “Are we free to break the laws?”, Theoria vol. 47:3 113-121.
List, Christian (2014): “Free Will, Determinism, and the Possibility of Doing Otherwise”,
NOUS 48:1, 156–178.
List, Christian (2019): Why Free Will Is Real, Harvard University Press, Cambridge.
Loewer, Barry (2012): “Two accounts of law and time”, Philos Stud, 160:115–137.
Loewer, Barry (forthcoming): Abstract for “The Consequence Argument Meets the Mentaculus”, in Time’s Arrow and the Probability Structure of the World, ed. Loewer, Weslake, Winsberg, Harvard University Press.
McTaggart, J.E. (1908): “The Unreality of Time”, Mind, 17(68): 457-474.
Ney, Alyssa (2014): Metaphysics, an introduction, Routledge, New York.
Pernu, Tuomas (2017): “Can Physics Make Us Free?”, Frontiers in Physics 5:54 https://doi.org/10.3389/fphy.2017.00064
Thyssen, Pieter and Sylvia Wenmackers (2020): “Degrees of freedom”, Synthese https://doi.org/10.1007/s11229-020-02714-z
Oxford University Press, New York.
Vihvelin, Kadri (2018): “Arguments for Incompatibilism”, The Stanford Encyclopedia of
Philosophy , Edward N. Zalta (ed.),
