ANOTHER SCIENTIFIC THREAT TO FREE WILL?

The claim that neuroscientists have proved thator at least produced

powerful evidence thatfree will is an illusion has received a lot of press.
Here are just two examples:
"Free will" is not the defining feature of humanness, modem neuroscience
implies, but is rather an illusion that endures only because biochemical complexity conceals the mechanisms of decision making. {Science News, Dec. 6,
2008; Siegfried [2008])
Researchers have found pattems of brain activity that predict people's
decisions up to 10 seconds before they're aware they've made a choice. . . .
The result was hard for some to stomach because it suggested that the unconscious brain calls the shots, making fi-ee will an illusory afterthought.
{ScienceNOWDaily News, April 14, 2008; Youngsteadt [2008])

In Effective Intentions: The Power of Conscious Will (Mele 2009), I argue

that scientistsneuroscientists and othershave not proved that fi-ee will
is an illusion and have not produced powerful evidence for that claim.
Manuel Vargas has suggested that in that book I ignore a serious scientific
threat to free will (2009). The alleged threat is identified in section 1. It is
the topic of this article.
Scientists who argue for the illusion thesis about fi-ee will arrive at
that conclusion by way of some pretty striking empirical propositions that
they infer from their data. For example, Benjamin Libet contends that the
brain produces unconscious decisions to act about a third of a second
before the person becomes aware of them (1985, 2004), and Daniel
Wegner argues that "conscious intentions" are never among the causes of
corresponding actions (2002,2004, 2008). In Effective Intentions I review
the experiments that are claimed to support striking empirical propositions such as these and I explain why the propositions are not justified by
the data. Because these empirical propositions play a central role in the
reasoning that leads such scientists as Libet and Wegner to their conclu"Another Scientific Threat to Free Will?" by Alfred Mele,
The Monist, vol. 95, no. 3, pp. 422^140. Copyright 2012, THE MONIST, Peru, Illinois 61354.

ANOTHER SCIENTIFIC THREAT TO FREE WILL?

423

sions about free will, if my arguments that the propositions are not
justified by the data hit their mark, they undermine the scientific
arguments at issue about free will. For example, it may be plausible that a
person whose conscious intentions never play a role in producing corresponding behavior never acts freely; but if we lack good reason to believe
that our conscious intentions never play a role of this kind, this particular
line of argument for the thesis that free will is an illusion is out of the
running (at least until powerful evidence for the claim about conscious intentions is produced).
1. What Threatens Free Will?
In a review o Effective Intentions., Manuel Vargas suggests that "the
core threat" to free will might arise not from the striking empirical propositions that I discuss there, but instead "from the bare fact that there are
neurological antecedents to conscious decisions. . . . What is really at
stake is whether our conscious intentions, even given some role in the production of action, can be picked out of the causal nexus and treated as
special or 'free'" (2009). He offers the following diagnosis of what scientists I disagree with might be thinking:
[They] are sometimes motivated by what the philosophical literature labels
as "source" intuitionsthe idea that for us to act with a free will we must be
the ultimate origins of strands of the causal nexus. On one way of putting
things, source theorists thitik that free acts catmot have causal antecedents
that extend back in time prior to the decisions of the agent or the agent's free
formulation of the relevant characterological inputs to that decision. . . . Acknowledging that our actions have causal roots in pre-conscious brain
activity, as Mele does, just highlights the fact of our causal embeddedness. It
does nothing to block the basic worry of how we could be the kinds of beings
that stop the buck enough to count as free, or as deserving of moral praise
and blame.

I said that my topic in this article is an alleged scientific threat to free

will. The alleged threat is the one just reported.
Some scientists do seem to understand free will in such a way that,
as Vargas puts it, "the bare fact that there are neurological antecedents to
conscious decisions" is incompatible with free will. After describing the
threat that he believes Libet's work poses to free will, CM. Fisher writes:
Somewhat the same conclusion may be reached on the basis of rather elementary observation. Every thought, feeling, inclination, intention, desire . . .

424

ALFRED MELE
must be created by nervous system activity. How else could they arise? Ideas
would have to arise without a physical basis. Nervous system activity must
always precede.... If one has the experience of 'willing' the nervous system
to do something, the impression of willing must have been preceded by
nervous system activity. Otherwise there would be no source and we are in
the realm of the supernatural. (2001, 364-65)

Fisher's claim seems to be that because all mental events are produced by
"nervous system activity," free will is an illusionthat the Libet-style

thesis that free will is an illusion is entailed by the fact that all mental
events are produced by brain events.' He seems to assume that free will
has to be "supernatural."
P. Read Montague takes a similar position:
Free will is the idea that we make choices and have thoughts independent of
anything remotely resembling a physical process. Free will is the close
cousin to the idea of the soulthe concept that 'you', your thoughts and
feelings, derive from an entity that is separate and distinct from the physical
mechanisms that make up your body. From this perspective, your choices are
not caused by physical events, but instead emerge wholly formed from
somewhere indescribable and outside the purview of physical descriptions.
This implies that free will cannot have evolved by natural selection, as that
would place it directly in a stream of causally connected events. (2008, 584)

Here Montague represents free will as something that depends for its
existence on the truth of substance dualisma view that includes a commitment to the idea that "associated with each human person, there is a
thinking thing . . . not composed of the same kinds of stuff as . . .
nonmental things" (Zimmerman 2006, 115; Zimmerman describes the
"thinking thing" as a soul, but some substance dualists prefer to use the
word "mind"). And, as I pointed out (Mele 2009, 67-68, 110-13), there
are similar dualistic elements in Libet's and Wegner's thinking aboutfreewill.
As Vargas mentions, in Free Will and Luck (Mele 2006) and
elsewhere I defend a position on what "free will" means. My position is
thoroughly naturalistic. I certainly do not view free will as something that
depends for its existence on the truth of substance dualism. Furthermore,
in my view, "causal embeddedness" and the fact that "our actions have
causal roots in pre-conscious brain activity" (Vargas 2009) pose no threat
at all to free will, and all of our actionsfree and unfreeare causally
influenced by things that happened long before we were bom. But in
Effective Intentions, I did not want to appeal to this philosophical view of

ANOTHER SCIENTIFIC THREAT TO FREE WILL?

425

mine in rebutting the scientific arguments I examined for the thesis that
free will is an illusion. Instead, I wanted to rebut those arguments by
showing that the striking empirical propositions I referred to earlier and
other propositions of this kindpropositions that play a pivotal role in
these argumentsare not justified by the data. This is something I could
do without arguing about what "free will" means. And that is a reason for
proceeding in the way I did: readers do not need to agree with me about
what "free will" means in order to see why the empirical propositions at
issue are not justified by the data.
2. Arguing with Scientists about What "Free Will" Means
The following argument is suggested by some remarks I quoted from
Vargas (2009) in the preceding section and by the passages I quoted from
Fisher and Montague:
1. All conscious decisions have neurological antecedents.
2. A conscious decision is freely made (or is an exercise of free will)
only if it has no neurological antecedents.
3. So no conscious decision is freely made (no conscious decision is
an exercise of free will).
I certainly accept premise I. I do not see how a real human being (or
nonhuman animal) can make conscious decisions in the absence of "neurological antecedents." And I certainly reject premise 2. Philosophical
grounds for rejecting premise 2 are implicit in, for example, my Free Will
and Luck (Mele 2006). Here I will take a more scientific approach to the
question how progress might be made in efforts to resolve disputes about
that premise.
As I mentioned. Fisher, Montague, and others understand free will
very differently than I do. For example, they seem to understand free will
in a way that makes premise 2 true, whereas I do not. If I were to try to
persuade them to understand free will differently, how would I try to
persuade them?
I take up this question shortly. Answering another question provides
some useful background. Here is the question: Why might I try to persuade
Fisher, Montague, and others to understand free will differently?

426

ALFRED MELE

There is evidence that lowering the subjective probability people

give to the proposition that free will exists increases bad behavior. In one
study (Vohs and Schooler 2008), people who read passages in which scientists deny that free will exists cheat significantly more often on a
subsequent task than others do. (People who read pro-free-will passages
do about the same as those who read neutral passages.) In another study
(Baumeister et al. 2009), people who read anti-free-will statements
behave more aggressively than a control group that reads neutral statements: they serve significantly larger amounts of spicy salsa to people
who say they dislike spicy food, despite being told that these people have
to eat everything on their plates! Now, suppose that what such scientists
as Fisher and Montague mean by "free will" is very different from what
most people mean by it. As they themselves understand free will. Fisher
and Montague may be entitled to be very confident that free will is an
illusion; and they may feel entirely justified in going public with that news
even if they leam that the news makes people's behavior worse. What
would they do if they were shown that most people understand free will
very differently? Perhaps they would try to convert people to their understanding of "free will." But they might instead try to look for evidence
about whether or not free will, as many people understand it, exists.
I have as much concem for tmth as most people, I hope; and I do not
advocate censorship about free will. But if, despite the evidence that there
is no free will as Fisher and Montague conceive of it, we lack good evidence
that there is no free will on a much more common conception of it, persuading scientists who share Fisher's and Montague's understanding of
free will that they are confused may prove useful. It may reduce the
amount of misleading and potentially harmfial free-will news; and a better
understanding of what "free will" means may foster experiments that have
more to do with free will than with substance dualism, for example.
How would I try to persuade scientists who take free will to depend
on substance dualism or on the existence of mental events that have no
brain events as causes that they are confused about the meaning of "free
will"? Well, scientists know that the simple fact that they are scientists
does not give them any special insight into what the expression "free will"
means. (A scientist might believe that philosophers do not have any
special insight into the meaning of "free will" either, and he or she might
offer as evidence the great amount of disagreement among philosophers
about what "free will" means.) They can be led to entertain the thought

ANOTHER SCIENTIFIC THREAT TO FREE WILL?

427

that their understanding ofthat expression might just be an artifact of their

own personal upbringing and to consider the hypothesis that they are out
of touch with ordinary usage of "firee will." In experiments with human
participants ("subjects," in slightly older terminology), scientists definitely
prefer to have a sample size larger than one person; and any scientist can
see that if the way he or she goes about determining what "free will"
means is simply to consult his or her own feeling or intuition about the
meaning, thento the extent to which it is important not to be mistaken
about the meaning of "fi-ee will"he or she should seek a better method.
There is an interesting body of work in psychology and experimental
philosophy on what nonspecialists mean by "free will." Attention to some
of it will prove useful. Eddy Nahmias, Stephen Morris, Thomas Nadelhoffer, and Jason Turner conducted survey studies about fi-ee will (2005,
2006). One of the studiesconducted with undergraduates who had not
studied free willfeatured the following story:
Imagine there is a universe that is re-created over and over again, starting
from the exact same initial conditions and with all the same laws of nature.
In this universe the same conditions and the same laws of nature produce the
exact same outcomes, so that every single time the universe is re-created,
everything must happen the exact same way. For instance, in this universe a
person named Jill decides to steal a necklace at a particular time, and every
time the universe is re-created, Jill decides to steal the necklace at that time.
(2006, 38)
In response to the question whether Jill decided to steal the necklace of
her own fi-ee will, 66% of the participants answered yes (p. 38). Very
similar results (rangingfi-om68% to 79%) were obtained for other stories
designed to portray actions in deterministic universeswhether those
actions were bad, good, or neutral (p. 39).
Eddy Nahmias, Justin Coates, and Trevor Kvaran conducted a study
designed to test, among other things, how the kind of causation featured
in a probe affects judgments about fi-ee will (Nahmias et al. 2007). In
some of the probes decisions are "completely caused by the specific
chemical reactions and neural processes" occurring in a person's "brain,"
whereas in other, parallel probes decisions are "completely caused by the
. . . thoughts desires, and plans" occurring in the person's "mind" (p. 223).
This difference had the biggest effect when the probe was "abstract" (that
is, did not mention specific actions) and featured scientific hypotheses
about the actual universe.

428

ALFRED MELE

Bracketed material in the text below appeared in a parallel probe:

Many respected neuroscientists [psychologists] are convinced that eventually we willfigureout exactly how all of our decisions and actions are entirely
caused. . . . If these neuroscientists [psychologists] are right, then once
specific earlier events have occurred in a person's life, these events will
definitely cause specific later events to occur. For instance, once specific
chemical reactions and neural processes [thoughts, desires, and plans] occur
in the person's brain [mind], they will definitely cause the person to make the
specific decision he or she makes. (Nahmias et al. 2007, 224)
Participants were asked to indicate their degree of agreement or disagreement with the following statement: "If the neuroscientists [psychologists]
are right, then people make decisions of their own free will" (225). When
the probe featured causation by "thoughts, desires, and plans," 82.9%
agreed with the statement; and when it featured causation by "chemical
reactions and neural processes," 38.3% agreed (227). This is, of course, a
huge difference. But it is interesting that 38% of the participants count
even decisions that are "entirely caused [by] specific chemical reactions
and neural processes" as being made of the person's own free will. Such
participants would seem to see no need for a nonphysical mind or soul to
be at work in the production of free decisions.
It also is interesting that when the setting is switched from the actual
universe to a possible universe and the probes involve an agent who
performs a good or a bad action, most participants regard the decisions as
being made of the person's own free will. The results were as follows (A^
indicates causation by "chemical reactions and neural processes" and P
indicates causation by "thoughts, desires, and plans"):
Of One's Own Free Will: Agree
N
P

Bad
59.8
66.1

Good
57.4
61.1

Apparently, when they entertain these # probes, a majority of people see

no need for a nonphysical soul or mind to play a role in the production of
decisions made of the person's own free will.
I conducted a simple survey study of my own that bears on these
issues. Participants in part 1 of the study were 69 students in a basic philosophy class at Florida State University. Free will and moral responsibility

ANOTHER SCIENTIFIC THREAT TO FREE WILL?

429

were not on their syllabus. About half of them (33) were presented with
the following text:
First answer the question on page # I. Then turn the sheet over and answer
the question on page # 2.
We're interested in how you understand free will. Please read the following
sentences and answer the question by circling your answer.
In 2019, scientists finally show exactly where decisions and intentions are
found in the brain. Our decisions are brain processes, and our intentions are
brain states.
In 2009, John Jones saw a 20 dollar bill fall from the pocket of the person
walking in front of him. He considered returning it to the person, who did not
notice the bill fall; but he decided to keep it. Of course, given what scientists
later discovered, John's decision was a brain process.
Question: Did John have free will when he made his decision?

On page 2, the participants read an expression of our interest in how they

understand "deserved moral blame," the same probe as on page 1, and the
following question: "Does John deserve to be blamed for what he does?"
The other participants (36) were presented with the same material in the
opposite order. (Order did not have a significant effect.)
The results were as follows. Nearly 90% (89.85%) of the participants
answered yes to the question about free will and about 87% (86.95%)
answered yes to the question about deserved blame. Apparently, for the
overwhehning majority, viewing a person's decisions as brain processes
was compatible with regarding the person as having free will (and as
deserving blame).
Part 2 of the study was conducted with another group of students in
basic philosophy courses at Florida State University. Again, free will and
moral responsibility were not on the course syllabi. Because the responses
to the questions about free will and moral responsibility in part 1 were so
similar, I decided to ask about just one this timefree will. I used a version
of the initial probe that was augmented to include the idea that "our
decisions and intentions are caused by other brain processes" (see note 2).
There were 86 participants; 74 of them (86%) said that John had free will
when he made his decision. For the overwhehning majority, viewing a
person's decision as "a brain process that was caused by other brain
processes" was compatible with regarding the person as having free will.^
Does my participant pool have extremely low, atypical standards for
ascriptions of free will in cases of human decision making? I designed

430

ALFRED MELE

part 3 of the study with that question in mind. Participants were another
group of Florida State Universify undergraduates (N = 90) taking a basic
philosophy course that did not deal with free will. This time I used two
stories. One is a version of the story used in part 2 in which I strengthened
the physicalist aspect by adding (at the beginning of the story) that in
2019, "scientists finally prove that everything in the universe is physical
and that what we refer to as 'minds' are actually brains at work." (The
remainder of the story is the same as the story in part 2.) The other is the
following "compliance drug" story:
In 2019, scientists who work for a secret military organization finally
develop a fool-proof compliance drug. The drug is used to make people
decide to do various things. Whenever they give a person the drug and then
suggest a course of action, that person is irresistibly caused to decide to take
that course of action. They make their suggestions through a tiny computer
chip that they implant in a person's brain.
These chemists gave the compliance drug to John Jones, a very honest man.
When John saw a 20 dollar bill fallfi^omthe pocket of the person walking in
fVont of him, they suggested keeping it. John considered returning it to the
person, who did not notice the bill fall; but, of course, he decided to keep it.
After all, the combination of the compliance drug and the suggestion forced
John to decide to keep it.

The order of presentation was counterbalanced, and participants were instructed not to change their answer to the first question in light of their
answer to the second. (They answered the question on the front of the
sheet, turned the sheet over, and answered the question on the back.)
Only 21.11% of the participants said that John had free will when he
made his decision in the compliance drug story, and 73.33% said this in
the physicalist scenario.^ The strong "no free will" response to the compliance drug story indicates that the great majority of participants do not
take a free-will-no-matter-what perspective on human decision making
(see Feltz et al. 2009, 16-19). And a scenario in which physicahsm is very
salient elicits a strong "free will" response.
Andrew Monroe and Bertram Malle conducted a study in which participants responded to the following request: "Please explain in a few lines
what you think it means to have free will" (2010, 214). The 180 participants were undergraduates at the Universify of Oregon. Monroe and Malle
report that "no assumptions of substance duahsm . . . were expressed"
(216). Nahmias reports related findings:

ANOTHER SCIENTIFIC THREAT TO FREE WILL?

431

a surprisingly low proportion of respondents: (1) agreed with the statement

"Humans have free will only because they have nonphysical souls" (1525%); (2) agreed with the statement "Our power of free will is something
that is not part of our brain" (18%); or disagreed with the statement "It is
because our minds are the products of our brains that we have free will"
(only 13% when the statement followed a description of our brains as
complex and unique, and still only 25% when the statement followed a description of the brain as mechanistic, govemed by physical laws, and soon to
be understood scientifically). (2011, n. 5)''
In a discussion of Daniel Wegner's work, Daniel Dennett writes:
If you are one of those who think that free will is only reallyfreewill if it
springs from an immaterial soul that hovers happily in your brain, shooting
arrows of decision into your motor cortex, then, given -whdXyou mean by free
will, my view is that there is no free will at all. If, on the other hand, you
think free will might be morally important without being supematural, then
my view is that free will is indeed real, but just not quite what you probably
thought it was. (2003, 222)
Dennett adds that, despite his admiration for Wegner's work, he sees
Wegner as "the killjoy scientist who shows that Cupid doesn't shoot
arrows and then insists on entitling his book The Illusion of Romantic
Lov" (224). One moral to take away from this, as I pointed out in
Effective Intentions, "is that if one sets the bar for free willthat is, for
the power or ability to act freelyridiculously high, the thesis that people
sometimes act freely should strike one as ridiculous" (2009, 110).
As I also observed in Effective Intentions, substance dualist views are
rarely advocated in contemporary philosophical publications on free will,
and, indeed, contemporary philosophers who argue for the existence of
free will typically shun substance dualism (2009, 67). Nahmias and his
coauthors have provided evidence that many nonspecialists are not
committed to a conception of free will according to which nonphysical
souls or minds need to be at work in the production of free decisions. The
study of mine that I reported provides additional evidence for this, as does
work reported by Monroe and Malle (2010). Now, even if philosophers
are not experts about the meaning of "free will," neither are scientists.
And when it comes to a disagreement between philosophers like me and
scientists like Fisher and Montague about whether free will must be supematural, the results of the survey studies would seem to suggest that to
the extent that we are disagreeing about the meaning of an everyday ex-

432

ALFRED MELE

pression, those scientists have more of a burden than I do. How would
they defend their position on what "free will" means? If and when such a
defense is forthcoming, it can be evaluated.
3. Another Experiment and General Observation
In Effective Intentions I examine a wide range of data that have been
used to support the claim thatfi-eewill is an illusion. The "ScienceNOW
Daily News" article from which I quoted at the beginning of this article
refers to a study (Soon et al. 2008) that was published after the book was
in press. I discuss the study shortly with the primary aim of illustrating a
general point.
Some philosophical background is in order first. The claim that we
act freely only if substance dualism is true should be distinguished fi-om
the claim that we act fi-eely only if our conscious decisions or conscious
intentions are at least sometimes among the causes of corresponding
actions.5 The existence of effective conscious decisions or intentions
seemingly does not depend on the truth of substance dualism. Conscious
decisions and intentions might, for example, be physical items or
supervene on physical items. In Effective Intentions, I examine alleged
evidence for the claim that conscious intentions and decisions are never
among the causes of corresponding actions (2009, chs. 5 and 6). I will not
repeat the examination here. However, I will repeat an observation that
helps forestall confusion.
For the most part, scientists are not metaphysicians; and they should
not be expected to take a stand on metaphysical connections between
mental items and physical itemsfor example, on whether conscious intentions supervene on physical states. (There is an enormous literature on
supervenience. For the uninitiated, I recommend Kim [2003].) From a
physicalist neuroscientific point of view, proof that the physical correlates
of intentions are among the causes of some corresponding actions counts
as proof that intentions are among the causes of some corresponding
actions, and evidence that the physical correlates of intentions are never
among the causes of corresponding actions counts as evidence that intentions are never among the causes of corresponding actions. As I observed
in Effective Intentions, it is primarily philosophers who would worry
about the metaphysical intricacies of the mind-body problem despite

ANOTHER SCIENTIFIC THREAT TO FREE WILL?

433

accepting the imagined proof about physical correlates, and the argumentation would be distinctly philosophical (2009, 146).^ (For a discussion of
evidence that conscious intentions or their physical correlates sometimes
are among the causes of corresponding actions, see Mele [2009], ch. 7.)
Chun Siong Soon, Marcel Brass, Hans-Jochen Heinze, and JohnDylan Hajoies conducted an experiment in which participants were asked
to perform a "motor-decision task while their brain activity was measured
using . . . fMRJ" (2008, 543). "When they felt the urge to do so, they were
to freely decide between one of two buttons, operated by the left and right
index fingers, and press it immediately." The study's "key question [is]
whether any brain region encoded the subject's motor decision ahead of
time" (544). Soon et al. write: "we found that two brain regions encoded
with high accuracy whether the subject was about to choose the left or
right response prior to the conscious decision." They report that "The predictive information in the fMRI signals" from a region of the frontopolar
cortex (BAIO) "was already present 7 s before the subject's motor decision.
Taking into account the sluggishness of the BOLD responses, the predictive neural information will have preceded the conscious motor decision
by up to 10 s." The second predictive region is in the parietal cortex.
When signals from the two regions just mentioned are combined, the
encoding accuracy is greatest; and even then the accuracy is only about
60%, with 50% being chance, of course (see Soon et al. [2008], Supplementary figure 6, Haynes [2011], 93). This study raises several questions.
I will pursue just two of them.
First, is it true that the brain regions at issue "encoded the subject's
motor decision ahead of time" (544)? Given that the encoding accuracy
was only 60%, it certainly is rash to conclude that a decision was actually
made at this early time (7 to 10 seconds before subjects were conscious of
a decision), and it is less rash to infer that, at this time, brain activity was
occurring that made it probable (though not highly probable) that the
agent would select a particular button. The brain activity may indicate that
the agent is, at that point, slightly more disposed to press that button than
the other one the next time he or she presses.
My second question is one we should ask about any alleged empirical
threat to the existence of free will. On what conceptions of free will, if
any, would the alleged threat actually be a threat? This is not the place for

434

ALFRED MELE

a thorough survey of the various philosophical positions on what "free

will" means, but I will comment briefly on two schools of thought about
this: compatibilism and incompatibilism.
Compatibilism and incompatibilism are theses about the conceptual
relationship between free will and determinism.' Determinism, according
to a standard definition, is the thesis that a complete statement of the laws
of nature together with a complete description of the entire universe at any
point in time logically entails a complete description of the entire imiverse
at any other point in time. (Human beings are parts of the universe, and a
description of what we were doing an hour ago is part of a complete description of the universe at that point in time.) Compatibilism is the thesis
that free will is compatible with the truth of determinism. Owing to their
acquaintance with contemporary physics, the great majority of contemporary compatibilists do not believe that determinism is true; but they do
believe that even if it were true, that would not preclude our having free
will. Incompatibilism is the thesis that free will is incompatible with the
truth of determinism. Most incompatibilists endorse Ubertarianismthe
conjunction of incompatibilism and the thesis that at least some people
sometimes have free will.
Compatibilism can sound very strange to nonspecialists. Elsewhere,
I have fried to explain why the strange sound might be misleading and
why compatibilism should be taken seriously (Mele 2009, 153-55); I will
not do so again here.^ Recall Vargas's report that "source theorists think
that free acts cannot have causal antecedents that extend back in time prior
to the decisions of the agent or the agent's free formulation of the relevant
characterological inputs to that decision" (2009). Compatibilists do not
see even causal antecedents that sfretch all the way back to the Big Bang
as being necessarily incompatible with free will: they contend that as long
as the causal chain goes through the agent in an appropriate way, the agent
can exercise free will.
It might be claimed that if, in situations of the sort that Soon and colleagues discuss, people make decisionsor have specific intentions
several seconds before they think they do, even compatibilists should
worry. How should compatibilists respond? As I explained, it has not
been shown that decisions about which button to press are made at this
early time nor that intentions to press a particular button are present then.
But suppose such intentions are present then, and suppose that participants do not freely decide which button to press because, in fact, they do

ANOTHER SCIENTIFIC THREAT TO FREE WILL?

435

not make a decision about which button to press: instead they unconsciously acquire an intention to press a certain button long before it seems
to them that they intend to press it.' Should compatibilists be terribly
worried? Not at all. It is very difficult to reason persuasively from alleged
or imagined findings about cases in which, as the agents realize, they have
no reason to favor either option over the other to the conclusion that the
same sort of thing would be found in cases in which the agents are far
from indifferent about their options. Presumably, automatic tie-breaking
mechanisms are at work in many cases in which we are indifferent
between or among the available options, as I have observed elsewhere
(Mele 2009, 83); and it is rash to assume that what happens in situations
featuring indifference is also what happens in situations in which unsettledness about what to do leads to careful reasoning about what to do.
Even if some action-ties are broken for us unconsciously well before we
are aware of what we "intend" to do, it certainly does not follow from this
that we never make effective conscious decisions. (And, of course, compatibilists are not at all troubled by the observation that these decisions
have "neurological antecedents.")
What about libertarianism? Libertarian theories about free will divide
into three kinds: event-causal, agent-causal, and noncausal. Typical eventcausal libertarian theories assert that agents never perform free actions
unless some of their actions are indeterministically caused. Whereas the
laws of nature that apply to deterministic causation are exceptionless,
those that apply most directly to indeterministic causation are instead
probabilistic. Tj^ically, events like deciding to help a stranded motorist
as distinct from the physical actions involved in actually helpingare
counted as mental actions. Suppose that Moe's decision to help a stranded
motorist is indeterministically caused by, among other things, his thinking
that he should help. Given that the causation is indeterministic, he might
not have decided to help given exactly the same internal and external conditions. In this way, event-causal libertarians seek to secure the possibihfy
of doing otherwise that they require for free action, or for fundamentally
free action (that is, free action that does not derive its freedom solely from
earlier free actions the agent performed).
Agent-causal hbertarian theories about free will assert that agents themselvesas opposed, for example, to agents' motivational and representational statesare causes of free actions. According to these theories, the
meaning of "free will" is such that exercises of it require "agent causation."

436

ALFRED MELE

Causation may be regarded as a relation between cause and effect. In

ordinary event causationfor example, an avalanche cmshing a cabin
both the cause and the effect are events. These events are connected by the
relation causation. In agent causation, an agent is connected by the relation
causation to an action (or intention) and that connection is not reducible
to a connection between states or events and the action (or intention).
Whereas most agent-causal libertarians prefer their agent causation straight,
some mix it with event causation in a theory about the production of free
actions (see Clarke 2003).
Noncausal libertarian theories about free will reject a popular view
about intentional action. Most theories about what intentional actions are
include a causal condition. According to these causal theories, all intentional
actions are events that are caused in a certain distinctive range of ways
either deterministically or indeterministically. For example, it may be
claimed that what it is to be an intentional action is to be an event that is
suitably caused by motivational and representational states. Noncausal
libertarians reject this idea. They insist both that the deterministic causation
of an action precludes its being a free action and that uncaused events can
be intentional actions.
Suppose, for the sake of argument, that actions are caused and that
the study at issue by Soon and colleagues proves that brain activity
occurring about ten seconds prior to button pressing actions gives participants in their experiment about a 0.6 probability of pressing a certain
button. Obviously, event-causal libertarians would not see this as a threat
to the claim that these people freely press the buttons they press nor to the
claim that they freely decide to press them. Antecedent probabilities of
decisions and other actions are comforting to event-causal libertarians
though, of course, antecedent probabilities of 0 or 1 would trouble them.
And, like compatibilists, event-causal libertarians have no need to reject
the claim that our actions "have causal antecedents that extend [far] back
in time" (Vargas 2009)as long as it is not being claimed that those antecedents are deterministic causes of our actions. If it is supposed, for the
sake of argument, that the participants in the Soon et al. study do not
actually decide which button to press and instead unconsciously acquire
an intention to press a certain button about ten seconds before they press
it, event-causal libertarians can help themselves to the compatibilist
response to this supposition sketched above.

ANOTHER SCIENTIFIC THREAT TO FREE WILL?

437

To the extent to which the study by Soon and colleagues provides

evidence that all actionsincluding decisionsare caused and that brain
events are among the causes of all actions, any libertarians who restrict
the sphere of fi-ee actions to uncaused actions or to agent-caused actions
that do not have brain events among their causes are in trouble. I have no
sympathy for libertarian views of these kinds.
One point to be emphasized about the data reported in Soon et al.
(2008) is that they pose no threat tofi-eewill on some leading philosophical accounts of what "fi-ee will" means. This is so even on the assumption
that all actions are caused and that all actions have brain events among
their causes. A second point to be emphasized harks back to the discussion
in section 2 of folk understandings offi-eewill. That discussion suggests
that the data at issue also do not threaten free will as the majority of nonspecialists conceive of it.
I said that my primary aim in discussing Soon et al. (2008) was to illustrate a general point. Here it is: Before we make a judgment about
whether particular data threaten the existence of free will, asking
ourselves how fi-ee will would need to be understood in order for the
threat to be a genuine one may often prove useful.
4. Conclusion
Undoubtedly, a number of things are among the causes of my persisting
in reading scientific work onfi-eewill and in and vmting about it. As I see
it, they include my belief that, as yet, there is no serious scientific threat
to free will, my belief that the false news that scientists have shown that
free will is an illusion is potentially dangerous, and my belief that I should
continue to try to counter that news. (Another cause, in my opinion, is my
persisting curiosity about how decisions, intentions, and actions are caused;
and some scientific work on "fi-ee will" sheds valuable hght on these issues.)
The main moral of the present article is that scientific arguments for the
nonexistence of free will that depend on the premise that, in Vargas's
words, "the bare fact that there are neurological antecedents to conscious
decisions" (2009) is incompatible withfi-eewill are far from persuasive.'"
Alfred Mele
Florida State University

438

ALFRED MELE

NOTES
1. Libet himself believed that free will is possible in a limited domain. He asserts that
"if the 'act now' process is initiated unconsciously, then conscious free will is not doing
it" (2001, 62; see 2004, 136). But he also claims that once we become aware of our
decisions or intentions, we can exercise free will in vetoing them (2004, 137-49). Some
people follow him part of the way. They accept the thesis about when and how decisions
to act are made but reject the window of opportunity for free vetoing as an illusion
(Wegner 2002, 55; Hallett 2007).
2. This probe read as follows:
In 2019, scientists finally show exactly where decisions and intentions are found in the brain and
how they are caused. Our decisions are brain processes, and our intentions are brain states. Also,
our deeisions and intentions are caused by other brain processes.
In 2009, John Jones saw a 20 dollar bill fall from the pocket of the person walking in fi-ont of
him. He eonsidered returning it to the person, who did not notice the bill fall; but he decided to
keep it. Of course, given what seientists later discovered, John's decision was a brain process
and it was caused by other brain processes.

The survey also asked whether students were taking their first philosophy class after high
school. The breakdown was 58 yes and 28 no. This difference did not have a statistically
significant effect on their answers about free will. About 88% of the first group and 82%
of the second group answered yes to the question about free will.
3. Of the participants who saw the physicalist story first (N = 43), 79.07% answered
yes to the question about that story, and 25.58% answered yes to the question about the
compliance drug story. The figures for those who saw the stories in the reverse order were
68.09% vs. 17.02%. Students who were taking their first philosophy course (N = 53) and
those who were not (N = 37) gave very similar responses: grand averages for yes answers
were 71.70% vs. 22.64% for the first-time students and 75.68% vs. 18.92% for the others.
4. Also see Stillman, Baumeister, and Mele (2011).
5. A comment on the expression "among the causes" may be usefiil. Suppose I had
deleted "among" or "among the" from the following string of words: "our conscious . . .
intentions are at least sometimes among the causes of corresponding actions." Some
readers would have inferred that I was entertaining the hypothesis that sometimes a
conscious intention is the only cause of a corresponding action. That is not a hypothesis
that I wish to entertain (see Mele 2009, 111).
6. Jackson 2000 is an excellent brief critical review of various relevant philosophical
positions that highlights the metaphysical nature of the debate.
7. "Compatibilism" and "incompatibilism" are often used as well to refer to views
about the conceptual relationship between moral responsibility and determinism.
8. I should add that I am officially agnostic about the issue that separates compatibilists from incompatibilists.
9. Freely deciding to press a button is one thing, and freely pressing it is another. I
forego discussion of free button pressing here.
10. I am grateftil to Manuel Vargas for motivating me to write this paper and for
comments on a draft. Some of the material in this article was presented at Peking University and Brown University. I am grateftil to my audiences for their input. This paper was
made possible through the support of a grant from the John Templeton Foundation. The

ANOTHER SCIENTIFIC THREAT TO FREE WILL?

439

opinions expressed in this publication are my own and do not necessarily reflect the views
of the John Templeton Foundation.

REFERENCES
Baumeister, R., E.J. Masicampo, and C.N. DeWall 2009. "Prosocial Benefits of Feeling
Free: Disbelief in Free Will Increases Aggression and Reduces Helpfulness," Personality and Social Psychology Bulletin 35: 260-68.
Clarke, R. 2003. Libertarian Accounts of Free Mil, Oxford: Oxford University Press.
Dennett, D. 2003. Freedom Evolves, New York: Viking.
Feltz, A., E. Cokely, and T. Nadelhoffer 2009. "Natural Compatibilism versus Natural Incompatibilism: Back to the Drawing Board," Mind and Language 24: 1-23.
Fisher, C. 2001. "If There Were No Free Will," Medical Hypotheses 56: 364-66.
Hallett, M. 2007. "Volitional Control of Movement: The Physiology of Free Will,"
Clinical Neurophysiology 118: 1179-92.
Haynes, J.-D. 2011. "Beyond Libet: Long-term Prediction of Free Choices from Neuroimaging Signals," in W. Sinnott-Armstrong and L. Nadel, eds. Conscious Will and
Responsibility, Oxford: Oxford University Press, 85-96.
Jackson, F. 2000. "Psychological Explanation and Implicit Theory," Philosophical Explorations I: 83-95.
Kim, J. 2003. "Supervenience, Emergence, Realization, Reduction," in M. Loux and D.
Zimmerman, eds.. The Oxford Handbook of Metaphysics., Oxford: Oxford University Press, 556-84.
Libet, B. 1985. "Unconscious Cerebral Initiative and the Role of Conscious Will in
Voluntary Action," Behavioral and Brain Sciences 8: 529-66.
. 2001. "Consciousness, Free Action and the Brain," Journal of Consciousness
Studies 8: 59-65.
. 2004. Mind Time. Cambridge, MA: Harvard University Press.
Mele, A. 2006. Free Will and Luck. New York: Oxford University Press.
. 2009. Effective Intentions. New York: Oxford University Press.
Monroe, A., and B. Malle 2010. "From Uncaused Will to Conscious Choice: The Need to
Study, Not Speculate About People's Folk Concept of Free Will," Review of Philosophy and Psychology 1:211-24.
Montague, P.R. 2008. "Free Will," Current Biology 18: R584-85.
Nahmias, E. 2011. "Why 'Willusionism' Leads to 'Bad Results': Comments on Baumeister, Crescioni, and Alquist," Neuroethics ^:\1-1A.
Nahmias, E., S. Morris, T. Nadelhoffer, and J. Turner 2005. "Surveying Freedom: Folk Intuitions About Free Will and Moral Responsibility," Philosophical Psychology 18:
561-84.
. 2006. "Is Incompatibilism Intuitive?" Philosophy and Phenomenological
Research. 73: 28-53.
Nahmias, E., J. Coates, and T. Kvaran 2007. "Free Will, Moral Responsibility, and
Mechanism: Experiments on Folk Intuitions," Midwest Studies in Philosophy 31:
214-^2.
Siegfried, T. 2008. "The Decider," Science News 174: 28-30.

440

ALFRED MELE

Soon, C , M. Brass, H. Heinze, and J. Haynes 2008. "Unconscious Determinants of Free

Decisions in the Human Brain," Nature Nerosdence 11: 543-45.
Stillman, T., R. Baumeister, and A. Mele 2011. "Free Will in Everyday Life: Autobiographical Accounts of Free and Unfree Actions," Philosophical Psychology 24:
381-94.
Vargas, M. 2009. Review of Effective Intentions, Notre Dame Philosophical Reviews
http://ndpr.nd.edu/review.cfm?id=17385.
Vohs, K., and J. Schooler 2008. "The Value of Believing in Free Will: Encouraging a
Belief in Determinism Increases Cheating," Psychological Science 19: 49-54.
Wegner, D. 2002. The Illusion of Conscious Will, Cambridge, Mass.: MIT Press.
. 2004. "Prcis o The Illusion of Conscious Will" Behavioral and Brain Sciences
27: 649-59.
-. 2008. "Self is Magic," in J. Baer, J. Kaufrnan, and R. Baumeister, eds. Are We
Free? Psychology and Free Will. New York: Oxford University Press.
Youngsteadt, E. 2008. "Case Closed for Free Will?" ScienceNOWDaily News 4/14/2008.
Zimmerman, D. 2006. "Dualism in the Philosophy of Mind," in D. Borchert, ed. Encyclopedia of Philosophy, 2nd ed., Detroit: Thomson Gale, vol. 3, 113-22.

Copyright of Monist is the property of Hegeler Institute and its content may not be copied or emailed to
multiple sites or posted to a listserv without the copyright holder's express written permission. However, users
may print, download, or email articles for individual use.