Do We Have Free Will?

How Language May Hold the Key to Freedom

Few problems in philosophy have spawned as much debate as the concept of free will. Do we really have control over our actions, or are we carried along by causal necessity? And what does it mean to have control, to have choice?

It seems to me that if we do have the power of choice, if multiple possibilities exist, this cannot be because of the introduction of random noise into an otherwise deterministic universe. As William James said,

if a ‘free’ act be a sheer novelty, that comes not from me, the previous me, but ex nihilo, and simply tacks itself on to me, how can I, the previous I, be responsible? How can I have any permanent character that will stand still long enough for praise or blame to be awarded? — William James, Pragmatism

I think that any concept of freedom consistent with the common understanding of that word would be a freedom to act in accordance with reasons of an intellectual and moral nature. But we are evolved creatures made of matter and situated within a universe of physical laws. Therefore, for any of our reasons to make sense, and not simply be random events that pop into existence, they must be an organic outgrowth of deeper reasons, reasons of a more fundamental nature: biological, chemical, physical. In other words, free will must be compatible with determinism.

This position is known as compatibilism. Most philosophers who accept the idea of free will would probably describe themselves as compatibilists. But if this is the version of free will you prefer, then you have to be able to answer the most powerful argument against it. Known as the Consequence Argument, it can be stated thus:

M: We have no control over the facts of the past or the laws of nature.

m: The facts of the past and the laws of nature determine the facts of the future.

C: Therefore, we have no control over the facts of the future.

I have no quarrel with the minor premise. That the facts of the past and the laws of nature determine the facts of the future is simply another way of saying that determinism is true. Any compatibilist will accept this.

Nor do I disagree with the first part of the major premise, that we have no control over the facts of the past. In a previous article I explained why I think that time travel is impossible, both as a practical matter and even in theory. But even if it were possible, what good is free will if being able to change the past is a pre-requisite to exercising it? That’s hardly what most people expect when they imagine they are free to choose.

That leaves the second half of the major premise, that we have no control over the laws of nature. Now this would seem to be even more obviously true than the idea that we have no control over the past. But as I hope to show, it’s really not.

Essentially, my argument is that we, including our actions, are themselves manifestations of nature. To believe otherwise is to assume a kind of metaphysical dualism, a breach between ourselves on the one hand and the natural world that gave rise to us on the other. When we act, we are not simply being carried along by laws of nature that are external to us. Instead, we are expressing the laws of nature. Or rather, we are nature expressing itself. When matter and energy are organized in certain, sufficiently complex, ways, they can become loci of causal origination, and these islands of teleology in a sea of mechanism are just another way that the laws of nature manifest themselves.

Consider the following thought experiment, taken from my book, Truth Evolves:

Imagine a very simple aquatic organism that has only two internal states. When the ambient light increases over a certain threshold, its tiny flipper-like appendages propel it downward. When the light decreases below a certain threshold, it swims up. Like a thermostat, it responds unthinkingly and immediately.

Let us construct an imaginary space and superimpose it over the universe containing our hypothetical organism, and let us surround the organism itself with a two-dimensional brane (think of it like a force field encasing the organism). This imaginary brane is meant to visually represent the degree to which the behavioral operations of the organism are restricted to reacting to stimuli available only within a certain spatiotemporal range.

Put another way, the extent of the brane, both in space and time, establishes a kind of cognitive horizon, such that the expansion of the brane beyond the spatiotemporal extension of the organism itself represents a spatiotemporal extension of the organism’s environment, and thus enlarges the set of possibly relevant stimuli that may affect the organism’s behavior.

Clearly in the case of our hypothetical organism, its cognitive horizon is spatiotemporally co-extensive with its physical self, since the organism can only react to those stimuli that are in immediate contact with it right now.

Now let’s imagine we make our organism more and more complex, by adding more variables from its environment that it will have to pay attention to, and more pre-programmed behavioral responses to the values held by those variables (temperature, current strength, presence of various chemicals, etc.). And let’s also imagine we give it a limited memory and the ability to adjust how it resolves conflicting behavioral impulses from different stimuli, giving more weight to more recent past experience.

Now our organism’s cognitive horizon will have to be expanded outward — in both space and time — to reflect the increase, both in the information available to the organism, and in the range of its possible behavioral responses.

We can continue on this way, adding complexity, improving memory, and so on, until we have arrived at an organism like a dog or an ape that evinces recognizably intelligent behavior. Such organisms, especially if they are social animals like primates or cetaceans, will have the benefit not only of their own intellect, but also of their instinctive ability to learn from one another.

This corresponds to a substantial expansion of their cognitive horizons, and thus to a much greater degree of freedom than the kind of mindless organism with which we began this experiment. These creatures have the ability to really look ahead, think back, make plans, and consider a variety of courses of conduct and their likely consequences.

The point of this thought experiment is to get you to consider how we might break down that wall between ourselves on the one hand, and the laws of nature on the other. The larger our cognitive horizons get, the more all the rest of nature is encompassed within our own.

Now, as I made clear in Truth Evolves, even very smart creatures, with quite extensive cognitive horizons, would still be limited in the sense that they would have to work with the intuitive ontology provided by evolution. They have to take the world as they find it.

But thanks to our capacity to learn language, we can actually question those intuitions, and we frequently do. We can build alternate models of the universe and the causal mechanisms that govern it. Many of our best scientific ideas — evolution, inertia, quantum mechanics — are deeply counter-intuitive. This makes a huge difference to our capacity for free will, since it effectively makes our cognitive horizons limitless.

That’s the essence of freedom, to my way of thinking. The more our cognitive horizon transcends the entire universe, the more the “laws of nature” and our own choices become, at least to some extent, one and the same. Therefore, the major premise of the Consequence Argument is shown to be flawed, and its conclusion no longer follows.

Of course, many of us will never question the basic nature of our reality. But thanks to language we at least have the possibility of expanding our cognitive horizons if we are willing to put in the effort. That’s why I concluded Truth Evolves with the observation that freedom, properly understood, entails intellectual responsibility. When we are responsible to the truth, we put ourselves on the path to freedom.