Determinism and Freedom
Determinism and free will are often thought to be in deep conflict. Whether or not this is true has a lot to do with what is meant by determinism and an account of what free will requires.
First of all, determinism is not the view that free actions are impossible. Rather, determinism is the view that at any one time, only one future is physically possible. To be a little more specific, determinism is the view that a complete description of the past along with a complete account of the relevant laws of nature logically entails all future events.1
Indeterminism is simply the denial of determinism. If determinism is incompatible with free will, it will be because free actions are only possible in worlds in which more than one future is physically possible at any one moment in time. While it might be true that free will requires indeterminism, it’s not true merely by definition. A further argument is needed and this suggests that it is at least possible that people could sometimes exercise the control necessary for morally responsible action, even if we live in a deterministic world.
It is worth saying something about fatalism before we move on. It is really easy to mistake determinism for fatalism, and fatalism does seem to be in straightforward conflict with free will. Fatalism is the view that we are powerless to do anything other than what we actually do. If fatalism is true, then nothing that we try or think or intend or believe or decide has any causal effect or relevance as to what we actually end up doing.
But note that determinism need not entail fatalism. Determinism is a claim about what is logically entailed by the rules/laws governing a world and the past of said world. It is not the claim that we lack the power to do other than what we actually were already going to do. Nor is it the view that we fail to be an important part of the causal story for why we do what we do. And this distinction may allow some room for freedom, even in deterministic worlds.
An example will be helpful here. We know that the boiling point for water is 100°C. Suppose we know in both a deterministic world and a fatalistic world that my pot of water will be boiling at 11:22am today. Determinism makes the claim that if I take a pot of water and I put it on my stove, and heat it to 100°C, it will boil. This is because the laws of nature (in this case, water that is heated to 100°C will boil) and the events of the past (I put a pot of water on a hot stove) bring about the boiling water. But fatalism makes a different claim. If my pot of water is fated to boil at 11:22am today, then no matter what I or anyone does, my pot of water will boil at exactly 11:22am today. I could try to empty the pot of water out at 11:21. I could try to take the pot as far away from a heating source as possible. Nonetheless, my pot of water will be boiling at 11:22 precisely because it was fated that this would happen. Under fatalism, the future is fixed or preordained, but this need not be the case in a deterministic world. Under determinism, the future is a certain way because of the past and the rules governing said world. If we know that a pot of water will boil at 11:22am in a deterministic world, it’s because we know that the various causal conditions will hold in our world such that at 11:22 my pot of water will have been put on a heat source and brought to 100°C. Our deliberations, our choices, and our free actions may very well be part of the process that brings a pot of water to the boiling point in a deterministic world, whereas these are clearly irrelevant in fatalistic ones.
Three Views of Freedom
Most accounts of freedom fall into one of three camps. Some people take freedom to require merely the ability to “do what you want to do.” For example, if you wanted to walk across the room, right now, and you also had the ability, right now, to walk across the room, you would be free as you could do exactly what you want to do. We will call this easy freedom.
Others view freedom on the infamous “Garden of Forking Paths” model. For these people, free action requires more than merely the ability to do what you want to do. It also requires that you have the ability to do otherwise than what you actually did. So, If Anya is free when she decides to take a sip from her coffee, on this view, it must be the case that Anya could have refrained from sipping her coffee. The key to freedom, then, is alternative possibilities and we will call this the alternative possibilities view of free action.
Finally, some people envision freedom as requiring, not alternative possibilities but the right kind of relationship between the antecedent sources of our actions and the actions that we actually perform. Sometimes this view is explained by saying that the free agent is the source, perhaps even the ultimate source of her action. We will call this kind of view a source view of freedom.
Now, the key question we want to focus on is whether or not any of these three models of freedom are compatible with determinism. It could turn out that all three kinds of freedom are ruled out by determinism, so that the only way freedom is possible is if determinism is false. If you believe that determinism rules out free action, you endorse a view called incompatibilism. But it could turn out that one or all three of these models of freedom are compatible with determinism. If you believe that free action is compatible with determinism, you are a compatibilist.
Let us consider compatibilist views of freedom and two of the most formidable challenges that compatibilists face: the consequence argument and the ultimacy argument.
Begin with easy freedom. Is easy freedom compatible with determinism? A group of philosophers called classic compatibilists certainly thought so.2 They argued that free will requires merely the ability for an agent to act without external hindrance. Suppose, right now, you want to put your textbook down and grab a cup of coffee. Even if determinism is true, you probably, right now, can do exactly that. You can put your textbook down, walk to the nearest Starbucks, and buy an overpriced cup of coffee. Nothing is stopping you from doing what you want to do. Determinism does not seem to be posing any threat to your ability to do what you want to do right now. If you want to stop reading and grab a coffee, you can. But, by contrast, if someone had chained you to the chair you are sitting in, things would be a bit different. Even if you wanted to grab a cup of coffee, you would not be able to. You would lack the ability to do so. You would not be free to do what you want to do. This has nothing to do with determinism, of course. It is not the fact that you might be living in a deterministic world that is threatening your free will. It is that an external hindrance (the chains holding you to your chair) is stopping from you doing what you want to do. So, if what we mean by freedom is easy freedom, it looks like freedom really is compatible with determinism.
Easy freedom has run into some rather compelling opposition, and most philosophers today agree that a plausible account of easy freedom is not likely. But, by far, the most compelling challenge the view faces can be seen in the consequence argument.3 The consequence argument is as follows:
- If determinism is true, then all human actions are consequences of past events and the laws of nature.
- No human can do other than they actually do except by changing the laws of nature or changing the past.
- No human can change the laws of nature or the past.
- If determinism is true, no human has free will.
This is a powerful argument. It is very difficult to see where this argument goes wrong, if it goes wrong. The first premise is merely a restatement of determinism. The second premise ties the ability to do otherwise to the ability to change the past or the laws of nature, and the third premise points out the very reasonable assumption that humans are unable to modify the laws of nature or the past.
This argument effectively devastates easy freedom by proposing that we never act without external hindrances precisely because our actions are caused by past events and the laws of nature in such a way that we not able to contribute anything to the causal production of our actions. This argument also seems to pose a deeper problem for freedom in deterministic worlds. If this argument works, it establishes that, given determinism, we are powerless to do otherwise, and to the extent that freedom requires the ability to do otherwise, this argument seems to rule out free action. Note that if this argument works, it poses a challenge for both the easy and alternative possibilities view of free will.
How might someone respond to this argument? First, suppose you adopt an alternative possibilities view of freedom and believe that the ability to do otherwise is what is needed for genuine free will. What you would need to show is that alternative possibilities, properly understood, are not incompatible with determinism. Perhaps you might argue that if we understand the ability to do otherwise properly we will see that we actually do have the ability to change the laws of nature or the past.
That might sound counterintuitive. How could it possibly be the case that a mere mortal could change the laws of nature or the past? Think back to Quinn’s decision to spend the night before her exam out with friends instead of studying. When she shows up to her exam exhausted, and she starts blaming herself, she might say, “Why did I go out? That was dumb! I could have stayed home and studied.” And she is sort of right that she could have stayed home. She had the general ability to stay home and study. It is just that if she had stayed home and studied the past would be slightly different or the laws of nature would be slightly different. What this points to is that there might be a way of cashing out the ability to do otherwise that is compatible with determinism and does allow for an agent to kind of change the past or even the laws of nature.4
But suppose we grant that the consequence argument demonstrates that determinism really does rule out alternative possibilities. Does that mean we must abandon the alternative possibilities view of freedom? Well, not necessarily. You could instead argue that free will is possible, provided determinism is false.5 That is a big if, of course, but maybe determinism will turn out to be false.
What if determinism turns out to be true? Should we give up, then, and concede that there is no free will? Well, that might be too quick. A second response to the consequence argument is available. All you need to do is deny that freedom requires the ability to do otherwise.
In 1969, Harry Frankfurt proposed an influential thought experiment that demonstrated that free will might not require alternative possibilities at all (Frankfurt  1988). If he’s right about this, then the consequence argument, while compelling, does not demonstrate that no one lacks free will in deterministic worlds, because free will does not require the ability to do otherwise. It merely requires that agents be the source of their actions in the right kind of way. But we’re getting ahead of ourselves. Here is a simplified paraphrase of Frankfurt’s case:
Black wants Jones to perform a certain action. Black is prepared to go to considerable lengths to get his way, but he prefers to avoid unnecessary work. So he waits until Jones is about to make up his mind what to do, and he does nothing unless it is clear to him (Black is an excellent judge of such things) that Jones is going to decide not to do what Black wants him to do. If it does become clear that Jones is going to decide to do something other than what Black wanted him to do, Black will intervene, and ensure that Jones decides to do, and does do, exactly what Black wanted him to do. Whatever Jones’ initial preferences and inclinations, then, Black will have his way. As it turns out, Jones decides, on his own, to do the action that Black wanted him to perform. So, even though Black was entirely prepared to intervene, and could have intervened, to guarantee that Jones would perform the action, Black never actually has to intervene because Jones decided, for reasons of his own, to perform the exact action that Black wanted him to perform. (Frankfurt  1988, 6-7)
Now, what is going on here? Jones is overdetermined to perform a specific act. No matter what happens, no matter what Jones initially decides or wants to do, he is going to perform the action Black wants him to perform. He absolutely cannot do otherwise. But note that there seems to be a crucial difference between the case in which Jones decides on his own and for his own reasons to perform the action Black wanted him to perform and the case in which Jones would have refrained from performing the action were it not for Black intervening to force him to perform the action. In the first case, Jones is the source of his action. It the thing he decided to do and he does it for his own reasons. But in the second case, Jones is not the source of his actions. Black is. This distinction, thought Frankfurt, should be at the heart of discussions of free will and moral responsibility. The control required for moral responsibility is not the ability to do otherwise (Frankfurt  1988, 9-10).
If alternative possibilities are not what free will requires, what kind of control is needed for free action? Here we have the third view of freedom we started with: free will as the ability to be the source of your actions in the right kind of way. Source compatibilists argue that this ability is not threatened by determinism, and building off of Frankfurt’s insight, have gone on to develop nuanced, often radically divergent source accounts of freedom.6 Should we conclude, then, that provided freedom does not require alternative possibilities that it is compatible with determinism?7 Again, that would be too quick. Source compatibilists have reason to be particularly worried about an argument developed by Galen Strawson called the ultimacy argument (Strawson  2003, 212-228).
Rather than trying to establish that determinism rules out alternative possibilities, Strawson tried to show that determinism rules out the possibility of being the ultimate source of your actions. While this is a problem for anyone who tries to establish that free will is compatible with determinism, it is particularly worrying for source compatibilists as they’ve tied freedom to an agent’s ability to be source of its actions. Here is the argument:
- A person acts of her own free will only if she is the act’s ultimate source.
- If determinism is true, no one is the ultimate source of her actions.
- Therefore, if determinism is true, no one acts of her own free will. (McKenna and Pereboom 2016, 148)8
This argument requires some unpacking. First of all, Strawson argues that for any given situation, we do what we do because of the way we are ( 2003, 219). When Quinn decides to go out with her friends rather than study, she does so because of the way she is. She prioritizes a night with her friends over studying, at least on that fateful night before her exam. If Quinn had stayed in and studied, it would be because she was slightly different, at least that night. She would be such that she prioritized studying for her exam over a night out. But this applies to any decision we make in our lives. We decide to do what we do because of how we already are.
But if what we do is because of the way we are, then in order to be responsible for our actions, we need to be the source of how we are, at least in the relevant mental respects (Strawson  2003, 219). There is the first premise. But here comes the rub: the way we are is a product of factors beyond our control such as the past and the laws of nature ( 2003, 219; 222-223). The fact that Quinn is such that she prioritizes a night with friends over studying is due to her past and the relevant laws of nature. It is not up to her that she is the way she is. It is ultimately factors extending well beyond her, possibly all the way back to the initial conditions of the universe that account for why she is the way she is that night. And to the extent that this is compelling, the ultimate source of Quinn’s decision to go out is not her. Rather, it is some condition of the universe external to her. And therefore, Quinn is not free.
Once again, this is a difficult argument to respond to. You might note that “ultimate source” is ambiguous and needing further clarification. Some compatibilists have pointed this out and argued that once we start developing careful accounts of what it means to be the source of our actions, we will see that the relevant notion of source-hood is compatible with determinism.
For example, while it may be true that no one is the ultimate cause of their actions in deterministic worlds precisely because the ultimate source of all actions will extend back to the initial conditions of the universe, we can still be a mediated source of our actions in the sense required for moral responsibility. Provided the actual source of our action involves a sophisticated enough set of capacities for it to make sense to view us as the source of our actions, we could still be the source of our actions, in the relevant sense (McKenna and Pereboom 2016, 154). After all, even if determinism is true, we still act for reasons. We still contemplate what to do and weigh reasons for and against various actions, and we still are concerned with whether or not the actions we are considering reflect our desires, our goals, our projects, and our plans. And you might think that if our actions stem from a history that includes us bringing all the features of our agency to bear upon the decision that is the proximal cause of our action, that this causal history is one in which we are the source of our actions in the way that is really relevant to identifying whether or not we are acting freely.
Others have noted that even if it is true that Quinn is not directly free in regard to the beliefs and desires that suggest she should go out with her friends rather than study (they are the product of factors beyond her control such as her upbringing, her environment, her genetics, or maybe even random luck), this need not imply that she lacks control as to whether or not she chooses to act upon them.9 Perhaps it is the case that even though how we are may be due to factors beyond our control, nonetheless, we are still the source of what we do because it is still, even under determinism, up to us as to whether we choose to exercise control over our conduct.
Free Will and the Sciences
Many challenges to free will come, not from philosophy, but from the sciences. There are two main scientific arguments against free will, one coming from neuroscience and one coming from the social sciences. The concern coming from research in the neurosciences is that some empirical results suggest that all our choices are the result of unconscious brain processes, and to the extent choices must be consciously made to be free choices, it seems that we never make a conscious free choice.
The classic studies motivating a picture of human action in which unconscious brain processes are doing the bulk of the causal work for action were conducted by Benjamin Libet. Libet’s experiments involved subjects being asked to flex their wrists whenever they felt the urge to do so. Subjects were asked to note the location of a clock hand on a modified clock when they became aware of the urge to act. While doing this their brain activity was being scanned using EEG technology. What Libet noted is that around 550 milliseconds before a subject acted, a readiness potential (increased brain activity) would be measured by the EEG technology. But subjects were reporting awareness of an urge to flex their wrist around 200 milliseconds before they acted (Libet 1985).
This painted a strange picture of human action. If conscious intentions were the cause of our actions, you may expect to see a causal story in which the conscious awareness of an urge to flex your wrist shows up first, then a ramping up of brain activity, and finally an action. But Libet’s studies showed a causal story in which an action starts with unconscious brain activity, the subject later becomes consciously aware that they are about to act, and then the action happens. The conscious awareness of action seemed to be a byproduct of the actual unconscious process that was causing the action. It was not the cause of the action itself. And this result suggests that unconscious brain processes, not conscious ones, are the real causes of our actions. To the extent that free action requires our conscious decisions to be the initiating causes of our actions, it looks like we may never act freely.
While this research is intriguing, it probably does not establish that we are not free. Alfred Mele is a philosopher who has been heavily critical of these studies. He raises three main objections to the conclusions drawn from these arguments.
First, Mele points out that self-reports are notoriously unreliable (2009, 60-64). Conscious perception takes time, and we are talking about milliseconds. The actual location of the clock hand is probably much closer to 550 milliseconds when the agent “intends” or has the “urge” to act than it is to 200 milliseconds. So, there’s some concerns about experimental design here.
Second, an assumption behind these experiments is that what is going on at 550 milliseconds is that a decision is being made to flex the wrist (Mele 2014, 11). We might challenge this assumption. Libet ran some variants of his experiment in which he asked subjects to prepare to flex their wrist but to stop themselves from doing so. So, basically, subjects simply sat there in the chair and did nothing. Libet interpreted the results of these experiments as showing that we might not have a free will, but we certainly have a “free won’t” because we seem capable of consciously vetoing or stopping an action, even if that action might be initiated by unconscious processes (2014, 12-13). Mele points out that what might be going on in these scenarios is that the real intention to act or not act is what happens consciously at 200 milliseconds, and if so, there is little reason to think these experiments are demonstrating that we lack free will (2014, 13).
Finally, Mele notes that while it may be the case that some of our decisions and actions look like the wrist-flicking actions Libet was studying, it is doubtful that all or even most of our decisions are like this (2014, 15). When we think about free will, we rarely think of actions like wrist-flicking. Free actions are typically much more complex and they are often the kind of thing where the decision to do something extends across time. For example, your decision about what to major in at college or even where to study was probably made over a period of months, even years. And that decision probably involved periods of both conscious and unconscious cognition. Why think that a free choice cannot involve some components that are unconscious?
A separate line of attack on free will comes from the situationist literature in the social sciences (particularly social psychology). There is a growing body of research suggesting that situational and environmental factors profoundly influence human behavior, perhaps in ways that undermine free will (Mele 2014, 72).
Many of the experiments in the situationist literature are among the most vivid and disturbing in all of social psychology. Stanley Milgram, for example, conducted a series of experiments on obedience in which ordinary people were asked to administer potentially lethal voltages of electricity to an innocent subject in order to advance scientific research, and the vast majority of people did so!10 And in Milgram’s experiments, what affected whether or not subjects were willing to administer the shocks were minor, seemingly insignificant environmental factors such as whether the person running the experiment looked professional or not (Milgram 1963).
What experiments like Milgram’s obedience experiments might show is that it is our situations, our environments that are the real causes of our actions, not our conscious, reflective choices. And this may pose a threat to free will. Should we take this kind of research as threatening freedom?
Many philosophers would resist concluding that free will does not exist on the basis of these kinds of experiments. Typically, not everyone who takes part in situationist studies is unable to resist the situational influences they are subject to. And it appears to be the case that when we are aware of situational influences, we are more likely to resist them. Perhaps the right way to think about this research is that there all sorts of situations that can influence us in ways that we may not consciously endorse, but that nonetheless, we are still capable of avoiding these effects when we are actively trying to do so. For example, the brain sciences have made many of us vividly aware of a whole host of cognitive biases and situational influences that humans are typically subject to and yet, when we are aware of these influences, we are less susceptible to them. The more modest conclusion to draw here is not that we lack free will, but that exercising control over our actions is much more difficult than many of us believe it to be. We are certainly influenced by the world we are a part of, but to be influenced by the world is different from being determined by it, and this may allow us to, at least sometimes, exercise some control over the actions we perform.
No one knows yet whether or not humans sometimes exercise the control over their actions required for moral responsibility. And so I leave it to you, dear reader: Are you free?
- I have hidden some complexity here. I have defined determinism in terms of logical entailment. Sometimes people talk about determinism as a causal relationship. For our purposes, this distinction is not relevant, and if it is easier for you to make sense of determinism by thinking of the past and the laws of nature causing all future events, that is perfectly acceptable to do.
- Two of the more well-known classic compatibilists include Thomas Hobbes and David Hume. See: Hobbes, Thomas, (1651) 1994, Leviathan, ed. Edwin Curley, Canada: Hackett Publishing Company; and Hume, David, (1739) 1978, A Treatise of Human Nature, Oxford: Oxford University Press.
- For an earlier version of this argument see: Ginet, Carl, 1966, “Might We Have No Choice?” in Freedom and Determinism, ed. Keith Lehrer, 87-104, Random House.
- For two notable attempts to respond to the consequence argument by claiming that humans can change the past or the laws of nature see: Fischer, John Martin, 1994, The Metaphysics of Free Will, Oxford: Blackwell Publishers; and Lewis, David, 1981, “Are We Free to Break the Laws?” Theoria 47: 113-21.
- Many philosophers try to develop views of freedom on the assumption that determinism is incompatible with free action. The view that freedom is possible, provided determinism is false is called Libertarianism. For more on Libertarian views of freedom, see: Clarke, Randolph and Justin Capes, 2017, “Incompatibilist (Nondeterministic) Theories of Free Will,” Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/entries/incompatibilism-theories/.
- For elaboration on recent compatibilist views of freedom, see McKenna, Michael and D. Justin Coates, 2015, “Compatibilism,” Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/entries/compatibilism/.
- You might be unimpressed by the way source compatibilists understand the ability to be the source of your actions. For example, you might that what it means to be the source of your actions is to be the ultimate cause of your actions. Or maybe you think that to genuinely be the source of your actions you need to be the agent-cause of your actions. Those are both reasonable positions to adopt. Typically, people who understand free will as requiring either of these abilities believe that free will is incompatible with determinism. That said, there are many Libertarian views of free will that try to develop a plausible account of agent causation. These views are called Agent-Causal Libertarianism. See: Clarke, Randolph and Justin Capes, 2017, “Incompatibilist (Nondeterministic) Theories of Free Will,” Stanford Encyclopedia of Philosophy, https://plato.stanford.edu/entries/incompatibilism-theories/.
- As with most philosophical arguments, the ultimacy argument has been formulated in a number of different ways. In Galen Strawson’s original paper he gives three different versions of the argument, one of which has eight premises and one that has ten premises. A full treatment of either of those versions of this argument would require more time and space than we have available here. I have chosen to use the McKenna/Pereboom formulation of the argument due its simplicity and their clear presentation of the central issues raised by the argument.
- For two attempts to respond to the ultimacy argument in this way, see: Mele, Alfred, 1995, Autonomous Agents, New York: Oxford University Press; and McKenna, Michael, 2008, “Ultimacy & Sweet Jane” in Nick Trakakis and Daniel Cohen, eds, Essays on Free Will and Moral Responsibility, Newcastle: Cambridge Scholars Publishing: 186-208.
- Fortunately, no real shocks were administered. The subjects merely believed they were doing so.
Frankfurt, Harry. (1969) 1988. “Alternative Possibilities and moral responsibility.” In The Importance of What We Care About: Philosophical Essays, 10th ed. New York: Cambridge University Press.
Libet, Benjamin. 1985. “Unconscious Cerebral Initiative and the Role of Conscious Will in Voluntary Action.” Behavioral and Brain Sciences 8: 529-566.
McKenna, Michael and Derk Pereboom. 2016. Free Will: A Contemporary Introduction. New York: Routledge.
Mele, Alfred. 2014. Free: Why Science Hasn’t Disproved Free Will. Oxford: Oxford University Press.
Mele, Alfred. 2009. Effective Intentions: The Power of Conscious Will. Oxford: Oxford University Press.
Milgram, Stanley. 1963. “Behavioral Study of Obedience.” The Journal of Abnormal and Social Psychology 67: 371-378.
Strawson, Galen. (1994) 2003. “The Impossibility of Moral Responsibility.” In Free Will, 2nd ed. Edited by Gary Watson, 212-228. Oxford: Oxford University Press.
van Inwagen, Peter. 1983. An Essay on Free Will. Oxford: Clarendon Press.
Deery, Oisin and Paul Russell, eds. 2013. The Philosophy of Free Will: Essential Readings from the Contemporary Debates. New York: Oxford University Press.
Mele, Alfred. 2006. Free Will and Luck. New York: Oxford University Press.