Throughout Sean Carroll's best-selling book, The Big Picture: On the Origins of Life, Meaning, and the Universe Itself (Dutton, 2016), Carroll seems comfortable holding two apparently contradictory views. This has been show throughout our review series. For example, he's fine both believing that causality is illusory (at the fundamental level of reality) and true (at the macroscopic level.)

We see this again in the chapter he dedicates to free will, which begins with this assessment (emphasis mine):

"There's a sense in which you do have free will. There's also a sense in which you don't. Which sense is the 'right' one is an issue you're welcome to decide for yourself (if you think you have the ability to make decisions.)" (378)

Carroll lets us know which view he holds: he thinks free will is fundamentally an illusion, and the only reason we use "free will" language is because it's useful. And why do we find it useful? Later he writes, "The unavoidable reality of our incomplete knowledge is responsible for why we find it useful to talk about the future using a language of choice and causation" (380). In other words, free will is false at the fundamental level of reality and we only use "free will" language at higher levels because we lack a complete knowledge of the current state of the universe. If, like Laplace's Demon, we knew the positions and velocities of all the particles in the universe, understood all the forces they are subject to, and had sufficient computational power to apply the laws of motion, then we would not use "free will" language—we would agree that everything is determined.

(NOTE: In a previous post, I showed why the Laplace's Demon idea comes up short.)

We find the same tension elsewhere in the chapter:

"A poetic naturalist says that we can have two very different-sounding ways of describing the world, a physics-level story and a human-level story, which invoke separate sets of concepts and yet end up being compatible in their predictions concerning what happens in the world." (381)

Setting aside the strange description of "physics-level" stories contra "human-level" stories—doesn't physics include human-level phenomena? Carroll probably means "quantum-level" stories—there's still a big problem. Notice that Carroll only defends these two descriptions as being predictively compatible. In other words, both descriptions are acceptable since both make accurate and/or useful predictions about the world.

What he didn't say is whether it's fine to hold these two views even if they contradict. It's not difficult to find contradictory views of the world that nevertheless make similarly accurate predictions. For example, both Newtonian and quantum theories can accurately predict the motions of human-sized objects. So on Carroll's view, they would both be acceptable. However, at the quantum level, Newtonian physics breaks down. It's simply no longer accurate. At best, it offers a good approximation of macroscopic phenomena, but it's fundamentally inaccurate when you consider reality as a whole.

But that doesn't bother Carroll so much. His poetic naturalism permits him to embrace inaccurate accounts of reality so long as they prove useful in daily life. Carroll writes (emphasis mine):

"There is no such notion as free will when we are choosing to describe human behavior as collections of atoms or as a quantum wave function. But that says nothing about whether the concept nevertheless plays a useful role when we choose to describe human beings as people. Indeed, it pretty clearly does play a useful role." (379)

In essence, Carroll's position is that at the quantum level, everything is determined. But at the level of everyday life, the concept of free will is useful. So poetic naturalists hold both views—both "stories"—in tension.

In fairness to Carroll, he doesn't say that the concept of free will is true on a macroscopic level, only that it's useful. But the implication is that he's fine holding erroneous views so long as they're useful—another example of his instrumentalism, which was examined in a previous post.

There are several problems with Carroll's rejection of free will. First is that it's clearly self-contradictory. Look at the above quote. Carroll twice talks about choosing a description of reality. But if we legitimately choose something, free will must exist. If we aren't able to choose something, then its outcome is determined. Thus we can't choose to deny the reality of free will without falling into contradiction.

A second problem, one Carroll admits, is the haunting fact that it seems as if we have free choice. Day to day, it seems as if we freely choose when to get out of bed, what to have for breakfast, how to start our day, what tasks to engage in, who to talk to, when to do home, etc. The common sense view is that each of us make millions of free decisions every day, some conscious and many unconscious.

Carroll actually agrees (emphasis mine): "The concept of choice does exist, and it would be difficult indeed to describe human beings without it" (379). In fact, in the very next chapter, which concerns meaning, Carroll notes several times how we choose what kind of life to live and how we choose to "expand our horizons, to find meaning in something larger than ourselves" (393). Without free will, it's hard to see how anyone could choose their own meaning or purpose.

A third problem is that if determinism is true, and none of our thoughts, conclusions, or actions are freely derived, then there's no reason to believe our views actually correspond to reality. On determinism, a set of elementary particles in my brain interacted to produce a thought such as, "Free will is false." But if the origin of that thought was determined and involved no free thinking on my end, then I can't trust that thought is true! I was determined to arrive at that conclusion, regardless of whether it accurately describes reality. I may believe that "free will is false," but I have very little confidence that's true.

A fourth problem is that if determinism was true, Carroll would not be writing books attempting to persuade people of that fact. If reality is fundamentally determined, why would he spend time trying to convince readers to freely change their minds, to freely adjust their understanding of the world to align with poetic naturalism? Even if I, a theist, read Carroll's book and become convinced that poetic naturalism was true, I couldn't freely reject my theism, no matter what I chose or how hard I tried—I'm simply determined to believe what I believe.

A fifth and very significant problem is human responsibility. If free will is fundamentally an illusion, then what do we do with praise and blame? Do criminals really shoulder moral blame for their actions? Do heroes really deserve praise? In both cases, the actors were just doing what their elementary particles determined them to do. We should neither praise or blame them any more than we would a tree for growing or the rain for falling.

To his credit, Carroll recognizes this final problem as a serious challenge for determinists. He writes:

"At extreme levels of free-will denial, the idea of 'responsibility' is as problematic as that of human choice. How can we assign credit or blame if people don't choose their own actions? And if we can't do that, what is the role of punishment or reward?" (383)

How does Carroll answer this challenge? He writes:

"Poetic naturalists...don't need to face up to these questions, since they accept the reality of human volition, and therefore have no difficulty in attributing responsibility or blame." (383)

Remember the passage quoted earlier where Carroll affirmed that, "There is no such notion as free will when we are choosing to describe human behavior as collections of atoms or as a quantum wave function". In other words, at the fundamental level of reality, free will is an illusion—everything is determined. But in the passage above, Carroll also affirms "the reality of human volition". Once again, the reason Carroll dismisses the challenge of human responsibility is because he has no problem holding two contradictory views.

Carroll closes his chapter on free will with a chilling look at what the future holds if determinism does prove true:

"To the extent that neuroscience becomes better and better at predicting what we will do without reference to our personal volition, it will be less and less appropriate to treat people as freely acting agents. Predestination will become part of our real world."

Most people will find this vision frightening. It seems Carroll is advocating something like The Minority Report, where citizens are punished for what they appear determined to do in the future.

Thankfully, Carroll doesn't think this will ever actually happen, not because it's a bad idea but because the "complexity of cognitive functioning [makes] predicting future actions infeasible in practice" (384). But such a society nevertheless follows in principle from Carroll's other commitments. There's nothing in principle stopping Carroll and others from punishing thought crimes, or even physical crimes years before they occur. It's not a huge leap to envision killing young children who seem determined to make bad choices in the future. For what on Carroll's view would prevent this?

We have just two more posts left in this series. Next time we'll look at Carroll's chapter on morality, then finally his “Ten Considerations” for naturalists. Stay tuned!

Today we continue our series looking at physicist Sean Carroll's new book, The Big Picture: On the Origins of Life, Meaning, and the Universe Itself (Dutton, 2016).

After exploring whether the universe needed a cause to get started, Carroll next turns to the topic of determinism: is reality determined or free?

Laplace's Demon

Carroll's answer relies on a famous thought-experiment involving "Laplace's Demon". Pierre-Simon Laplace was an accomplished French physicist and mathematician. He's also, according to Carroll, one of the fathers of determinism, which holds that all future states are inevitable consequences of past events and causes. Laplace proposed a hypothetical vast intellect (aka "Laplace's Demon") who has an omniscient grasp of reality. In today's scientific language, the Demon would know the positions and velocities of all the particles in the universe, understand all the forces they are subject to, and have sufficient computational power to apply the laws of motion. The Demon would then be able to determine any future event by simply analyzing the state of the universe right now. Or as Carroll says, "Laplace was pointing out that the universe is something like a computer"—current state in, future state out.

Now, Carroll admits that for all practical purposes, Laplace's Demon is not real. He's just a useful thought experiment. Such a Demon will never exist in the real world since there's simply too much information to account for. We will never invent a computer that can ascertain the state of every particle and force at every moment in time. To do so, Carroll admits, would require God-like programming (which he immediately dismisses, uncomfortable with any conclusion that may result in God.) So we will never be able to tell the future with complete certainty, ala Laplace's Demon.

But that's not what Carroll's primarily interested in. He doesn't aim to show that physics can ascertain the future; he's just interested showing that physics can determine the future.

That leads to a natural question: why do we think this is true? Even if a Demon or a computer had complete knowledge of the state of the universe, why should we assume such prior states determine what happens next? Carroll never explains. He simply presumes causal determinism without proving it.

(Careful readers will note that in the previous chapter, Carroll aims to show that causality is not a feature of fundamental reality. But here he aims to prove causal determinism, that prior states cause future states. Carroll either doesn't notice or doesn't worry about the apparent contradiction.)

Later in the book, in a different chapter, Carroll gestures toward a supporting argument for determinism. He writes:

"The Laplacian view...is based on patterns, not on natures and purposes. If this certain thing happens, we know this other thing will necessarily follow thereafter, with the sequence described by the laws of physics. Why is it this way? Because that's the pattern we observe."

This again raises many questions. For instance, why think patterns are mutually exclusive of natures and purposes? Again, Carroll just asserts this without evidence. Why can't things follow patterns given by their nature?

Also, why do things in the world follow these specific patterns and laws, and not others? It's not enough to say, "Because those are the patterns we observe." That just avoids the question. It's equivalent to saying, "Nature follows the patterns we observe because those are the patterns we observe," or to put it more simply, "That's just the way it is," an answer that may satisfy Bruce Hornsby but not the truly curious skeptic.

Still, the biggest problem with Carroll's Laplacian defense of determinism was already preempted by David Hume. The Scottish skeptic affirmed what stock brokers remind us of today, that past performance is no guarantee of future results. The fact that past events usually or even always occur in some pattern doesn't mean future events have to occur that way.

For example, suppose that from birth to age 30, I ate a vanilla ice cream cone at exactly 2:00pm, every single day. So when my 31st birthday party rolled around, you glance at the clock and see it's 1:59pm, and you're nearly certain what will happen next. After all, that's the pattern you've always observed, me eating ice cream at 2:00pm, not just once or twice, but repeatedly and without exception for thirty years. As the hour chime hits, you see me scoop ice cream into a cone, lift the cone to my mouth....and then I stop. I strangely put the cone down and decide not to have any today.

Now, if you suggested I was determined to eat ice cream on my 31st birthday since "that's the pattern [you] observed" (to use Carroll's language), you would have been wrong. Thus determinism can't be solely grounded in the knowledge of past patterns.

But suppose the determinist replied, "Ah! But maybe you were determined to eat ice cream every day for thirty years and then determined to stop eating it on your 31st birthday!" Perhaps. But how would we know it? What arguments or evidence could we offer to support that proposal? Carroll says we know events are determined "because that's the pattern we observed." But in this case, it's precisely not the pattern we observed. Assuming it would be a new determined pattern that was previously unobservable would only beg the question in favor of determinism and make it unfalsifiable.

Determinism, Destiny, and Fate

Bad arguments aside, Carroll's not wholly comfortable with the implications of determinism. Therefore, he tries to soften its blow by distinguishing it from destiny and fate (emphasis mine):

"The physical notion of determinism is different from destiny or fate in a subtle but crucial way: because Laplace's Demon doesn't actually exist, the future may be determined by the present, but literally nobody knows what it will be." (36)

This reveals a confusion. Carroll implies that since we can't know the future, determinism doesn't imply fate or destiny. But our knowledge of the future is irrelevant to the question of whether we're fated or free. Determinism indeed implies fate since perfect knowledge of the current moment yields perfect knowledge of the next, and so on and so forth, ad infinitum. That means every future moment is already set based on the state of the world right now (which is, in turn, based on the states before it). On determinism, your entire future is unavoidably fated.

Carroll tries another route to sidestep determinism's dreary implications, this time returning to his poetic naturalism:

"There is one way of talking about the universe that describes it as elementary particles or quantum states, in which Laplace holds sway and what happens next depends only on the state of the system right now. There is also another way of talking about it, where we zoom out a bit and introduce categories like 'people' and 'choices'...Our best theories of human behavior are not deterministic. We don't know any way to predict what a person will do based on what we can readily observe about their current state. Whether we think of human behavior as determined depends on what we know."

Here, again, we spot the confusion. Carroll seems to think determinism means you can predict future states of events—to the degree you can predict them, they're determined. But determinism is independent of what we know or predict. If reality was completely determined, that would remain true whether we predicted some, all, or no future events.

Also, Carroll affirms that while fundamental reality (i.e, level one reality, the deepest, quantum level of the universe) is deterministic, the higher levels (i.e., "emergent" levels) are not necessarily deterministic. In fact, they seem to be quite the opposite. Yet how can this be so? If elementary particles and quantum states are determined, then categories like "choice" would be, at best, useful fictions. They may help us get along in the everyday world, but they're ultimately out of step with fundamental reality. Once again, we see another example of Carroll's instrumentalism, preferring "stories" that work rather than explanations that are true.

The Final Blow Against Determinism

In the end, Carroll offers no convincing reasons to think determinism is true. Yet even if Carroll thought it was true, why would he try to persuade us of it? If determinism was true, then we've been pre-determined to either accept or reject it—we have no choice in the matter! And so we arrive at the final knockout blow against determinism: anyone trying to convince people determinism is true, to convince them to freely change their mind about determinism, implicitly undermines it.

In the next post, we'll looks at Carroll's thoughts about the philosophy of time. Stay tuned!

In his Philosophical Investigations, Ludwig Wittgenstein complained that “in psychology there are experimental methods and conceptual confusion.” What he meant is that academic psychologists too often interpret empirical evidence in light of unexamined and dubious metaphysical assumptions. What is presented as good science is really just bad philosophy.

The recent spate of neuroscientific and psychological literature claiming to show that free will is an illusion provides a case in point. Philosopher Alfred Mele’s new book, Free: Why Science Hasn’t Disproved Free Will (Oxford, 2015), is a brief, lucid, and decisive refutation of these arguments. Mele demonstrates that scientific evidence comes nowhere close to undermining free will, and that the reasoning leading some scientists to claim otherwise is amazingly sloppy.

Perhaps the best known alleged evidence against free will comes from the work of neurobiologist Benjamin Libet. In Libet’s experiments, subjects were asked to flex a wrist whenever they felt like doing so, and then to report on when they had become consciously aware of the urge to flex it. Their brains were wired so that the activity in the motor cortex responsible for causing their wrists to flex could be detected. While an average of 200 milliseconds passed between the conscious sense of willing and the flexing of the wrist, the activity in the motor cortex would begin an average of over 500 milliseconds before the flexing. Hence the conscious urge to flex seems to follow the neural activity which initiates the flexing, rather than causing that neural activity. If free will requires that consciously willing to do something is the cause of doing it, then it follows (so the argument goes) that we don’t really act freely.

As Mele shows, the significance of Libet’s results has been vastly oversold. One problem is that Libet did not demonstrate that the specific kind of neural activity he measured is invariably followed by a flexing of the wrist. Given his experimental setup, only cases where the neural activity was actually followed by flexing were detected. Also, Libet did not check for cases where the neural activity occurred but was not followed by flexing. Hence we have no evidence that that specific kind of neural activity really is sufficient for the flexing. For all Libet has shown, it may be that the neural activity leads to flexing (or doesn’t) depending on whether it is conjoined with a conscious free choice to flex.

There’s a second problem. The sorts of actions Libet studied are highly idiosyncratic. The experimental setup required subjects to wait passively until they were struck by an urge to flex their wrists. But many of our actions don’t work like that—especially those we attribute to free choice. Instead, they involve active deliberation, the weighing of considerations for and against different possible courses of action. It’s hardly surprising that conscious deliberation has little influence on what we do in an experimental situation in which deliberation has been explicitly excluded. And it’s wrong to extend conclusions derived from these artificial situations to all human action, including cases which do involve active deliberation.

Even if the neural activity Libet identifies (contrary to what he actually shows) invariably preceded a flexing of the wrist, it still wouldn’t follow that the flexing wasn’t the product of free choice. Why should we assume that a choice is not free if it registers in consciousness a few hundred milliseconds after it is made? Think of making a cup of coffee. You don’t explicitly think, “Now I will pick up the kettle; now I will pour hot water through the coffee grounds; now I will put the kettle down; now I will pick up a spoon.” You simply do it. You may, after the fact, bring to consciousness the various steps you just carried out; or you may not. We take the action to be free either way. The notion that a free action essentially involves a series of conscious acts of willing, each followed by a discrete bodily movement, is a straw man, and doesn’t correspond to what common sense (or, for that matter, philosophers like Wittgenstein or Aquinas) have in mind when they talk about free action.

Other arguments against free will are no better. For example, in psychologist Stanley Milgram’s famous obedience experiments, participants were instructed to administer what they falsely supposed were genuine electric shocks to people who gave incorrect answers to questions put to them. Many participants reluctantly obeyed these commands even when they seemed to be causing severe pain. As with the neuroscientific evidence, some have argued that such data casts doubt on free will. But as Mele says, it’s difficult to see “exactly what the argument is supposed to be.” Is the claim that Milgram’s experimental setup made it inevitable that participants would obey? That can’t be it, because not every participant obeyed the commands. Is the idea merely that situations exist in which people find it difficult to disobey authority figures? If so, what defender of free will ever denied it?

Mele’s book shows that, if anyone has been too quick to follow authority, it’s those who swallow dubious philosophical claims merely because they are peddled by scientists.

Originally posted in the City Journal. Used with permission.
(Image credit: Templeton)