Part III: Threats to free will from neuroscience

Compatibilists think that we are morally responsible for our actions even if we inhabit a deterministic universe. They argue that moral responsibility does not require magic powers (i.e. the unconditional ability to do otherwise), but instead, requires that an agent possesses certain psychological capacities C (e.g. rational deliberation, self-control, mental time travel) and that those capacities are not constrained by any excessive internal (e.g. strong addictive desires) or external forces (e.g. being held at gunpoint). If we are to understand ‘free will’ as a placeholder for whatever gives us moral responsibility, compatibilists would argue that the unconstrained exercise of our psychological capacities is an exercise of free will.

On this view, free will is not just an all-or-nothing feature, it also comes in degrees. Worms don’t have free will, children have some, and psychologically normal adult humans have quite a bit of free will, at least some of the time. Very young children (1-2 years old) don't have free will because their psychological capacities aren't sufficiently developed. Psychopathic adults don't have as much free will as normal adults because their psychological capacities are impaired. The extent to which normal adult humans have free will—in this compatibilist sense—is a matter of controversy. There are many scientists who think that recent findings undermine our belief in free will. That is to say, they believe that the real causes of our actions are something other than what we think they are (e.g. unconscious processes, environmental factors, brain chemistry). Before reviewing those findings, it might be useful to consider a range of hypotheses one could formulate regarding the extent to which we have free will:

1) Streams of freedom: C regularly causes our behavior.
2) Flickers of freedom: C rarely causes our behavior.
3) Zero freedom: C never causes our behavior.

I do not take these three hypotheses to be exhaustive, by any means, but only useful in illustrating the debate among philosophers and scientists. I suspect most of the population would endorse something pretty close to 1. Free will skeptics hold that 3 is true and will often cite brain imaging studies for support (e.g. Libet 1983, Soon et al. 2008). These studies are taken to show that unconscious processes—processes for which we are not, and cannot be aware of—drives our behavior. If unconscious processes are the sole causes of our behavior, then our conscious processes don’t cause our actions, even if we may think that they do. I will now briefly summarize the basic methodology of such experiments.

Subjects are hooked up to electrodes are asked to make a simple decision without thinking about their reasons for doing so. For example, deciding between flexing their left or right wrist. The subjects are also told to carefully monitor a fast moving clock to determine when exactly they came to their random decision. Afterwards, the experiments compare the time to which the subjects reported their intention to act, to the electrical activity in their brain.

Across a number of experiments, EEG data show increased brain activity in the motor cortex 300ms before the subjects report their intentions. The brain activity that precedes the conscious decision is taken to demonstrate that the subjects’ decision making is carried out by unconscious or random brain processes, rather than by the subjects themselves. Scientists then make the following inference. These findings generalize to all of our decisions. Thus, these experiments demonstrate that we don’t ever actually have free will.

I believe that the scientists and philosophers who cite these kinds of studies in support of free will skepticism make a hasty generalization. The conclusions of the experimenters may be good explanations for what’s going on in these specific experiments, but how much application do they have in the real world? After all, there is at least some reason to think that a subset of our decisions and actions aren’t based on random or unknown processes for which we are unaware. Surely they sometimes involve conscious reasoning, carefully weighing the pros and cons of a tough decision. For example, whether or not we should have kids, which career to pursue, and which conclusions we ought to draw from neuroscientific studies. Free will skeptics would of course deny these claims, but the burden of proof is on them to demonstrate that their generalizations hold true. 

One might argue that our decision making processes still interact with brain processes for which we are unaware. Surely unconscious processes are going to interact with our conscious ones, but it doesn’t follow that our conscious processes never cause us to do anything. If an unconscious process can cause behavior, why couldn’t a conscious one as well? Many of the claims regarding causation involve implicit assumptions about the nature of causality. Thus, scientists and philosophers are not only making claims about the details of the experiments, they are also making claims about metaphysics. There is much more to say about causation and its relation to mental phenomena, but it would be more suited for a post of its own.

In conclusion, compatibilist free will would be threatened only if the findings in these experiments really did generalize to all of our actions. If our conscious thoughts never gave rise to action, then we would not have free will. At this point, such a conclusion is unwarranted, to say the least. In the next post, I will discuss another possible threat to compatibilist free will. Namely, findings in the field of social psychology that are taken to show that our behavior is determined by situational factors that we are unaware of.

Works cited

Libet, B., Gleason, C. A., Wright, E. W., & Pearl, D. K. (1983). Time of conscious intention to act in relation to onset of cerebral activity (readiness-potential). Brain, 106(3), 623-642.

Soon, C. S., Brass, M., Heinze, H. J., & Haynes, J. D. (2008). Unconscious determinants of free decisions in the human brain. Nature neuroscience, 11(5), 543-545.


*This post is part of a series on free will. 

Transgenderism and Science

Liberals tend to think of themselves as belonging to the “party of science”. To justify use of this title, liberals will point out how conservatives tend to reject mainstream climatology (e.g. humans are warming the planet) and well-established biology (e.g. evolution). But conservative critics allege that liberals reject mainstream science too. Specifically, the science behind the safety of GMOs, vaccines, nuclear power, and most recently, transgenderism—the idea that gender and biological sex can come apart. I agree that anti-scientific views are held by both parties (perhaps even equally so), but I think it is a mistake to lump transgenderism into the anti-science bin. While I think transgenderism raises some scientific questions, I don’t think liberals, or transgender individuals, tend to believe things that are contradicted by well-established science.

At a recent event at Ferris State University, conservative pundit Ben Shapiro responded to a student who asked a question about transgender rights. A video of this exchange* was uploaded to youtube shortly after. Here are some of the central claims Shapiro makes in his response to the student:

(1) Transgender individuals are making claims that are in conflict with well-established science and deny basic biological facts (1:15).

(2) Sex and gender are coextensive. They cannot come apart (0:15).

(3) The high transgender suicide rate is to be explained by the condition itself, rather than by discrimination, rejection, and stigmatization (starts around 3:20).

In this post, I will argue that the first claim is false, that the second is not well-established by the science, and that justification for the third claims rests upon a misrepresentation of the evidence and faulty reasoning.

(1)

Early on in the exchange, Shapiro expresses his knee-jerk reaction to transgender individuals: “You are not the sex to which you claim to be” (0:59).

If taken literally, he is attacking a straw man, as that is not the claim transgender people are making. Transgender people are claiming that their gender does not align with their sex—not that their sex isn’t what the doctors determined at their birth. If you were to ask them about their chromosomes or sex organs, they would not say something false.

Transgender individuals insist there is an important distinction between ‘sex’ and ‘gender’. Sex is to be understood as the possession of all or almost all of a number of biological properties (e.g. genetics, sexual organs, hormone production) associated with being either male or female, whereas gender is the possession of all or almost all of a number of psychological properties and behavioral dispositions (e.g. identity, expression, masculinity vs. femininity) that are typically associated with one’s sex. Put more simply, sex is what’s between the legs, gender is what’s between the ears. The distinction is made not only by transgender people, but can be found in any psychology or sociology textbook.

Transgender individuals are alleging that there can be a mismatch between one’s sex and gender profiles. That is to say, it’s possible that a man** can be born in a female body and vice versa. The American Psychological Association takes a similar stance, defining transgender as “an umbrella term for persons whose gender identity, gender expression or behavior does not conform to that typically associated with the sex to which they were assigned at birth.”

(2)

I take it that Shapiro will either reject that there is a distinction between ‘sex’ and ‘gender’ or that the distinction doesn’t validate the beliefs of transgender individuals. I say this because, later in the video, Shapiro claims that, “Gender is not disconnected from sex.” (0:15) What exactly does he mean by this?

One interpretation is that he is claiming that ‘gender’ and ‘sex’ are coextensive. That is to say, anything that is a biological male or female must have the corresponding gender profile by necessity. If the two classes of properties always go together, the distinction between sex and gender effectively falls apart. But if Shapiro is saying that sex and gender are coextensive, then he is making a scientific claim, one that is by no means well-established science or a basic biological fact. His claim would then be asserting that a scientific hypothesis about the relation between sex and gender is true, without any evidence.

It is true that the coextension hypothesis is in direct conflict with what transgender people are saying. Transgender people believe that sex/gender mismatching is possible and that their life experiences attest to that fact. (In the scientific literature, the possibility of mismatching is treated as a hypothesis, known as “the brainsex theory of transsexualism”). At best, liberals and transgender individuals would be denying that the coextension hypothesis is true. But that does not amount to denying basic biological facts, as it would in the cases of identifying as a different age or species (Shapiro’s analogies to age and species start around 2:15).

If Shapiro is making a more semantic claim, that ‘sex’ and ‘gender’ are synonymous, then he’s just wrong. This would be true even if we assume that sex and gender are co-extensive. ‘Sex’ and ‘gender’ would still apply to different aspects of the same referent. The property of having a heart is coextensive with the property of having a kidney. But having a heart does not mean the same thing as having a kidney.

One might still argue that the coextension thesis is more plausible because it is supported by “common sense” intuitions about sex and explains why many use ‘sex’ and ‘gender’ interchangeably. I think that most people’s life experiences are consistent with the coextension thesis, only because most people identify with the gender associated with their sex. If mismatching can happen, but very infrequently (0.3% of people), one would not expect many people to have encountered evidence contrary to it. Hence, the life experiences of most people cannot count as strong evidence in favor of the coextension thesis, as it is consistent with the possibility that mismatching occurs in rare circumstances.

I see no reason to think, from the armchair, that the coextension thesis is plausible. (Perhaps those who make certain theological assumptions might disagree.) To the contrary, I think there is evidence that provides some support for the mismatching hypothesis. Namely, the testimony of transgender people.

Imagine a world where the vast majority of people have observed time and time again that ravens are black. The common sense view would be that ravens are just black by nature and that they can’t come in any other color. But say that there was a very small group of people claiming to have seen a white raven. How should the vast majority of people respond to such claims? One response would be skepticism. Most people have only seen black ravens. Therefore, it is unlikely that white ravens exist and very likely that the individuals claiming otherwise are mistaken. Another response would be to accept that white ravens might exist, but not to take a stand on whether or not they do. I think the latter attitude is more justified than the former.

Similarly, I think accepting the possibility that mismatching is possible is reasonable given the testimony of transgender people. We have no good reason to think that they are lying about their subjective experiences. Additionally, the average person does not know enough about genetics or biology to determine whether or not mismatching is possible. The specific biological bases of gender (if there in fact are any) are still hotly contested, even among scientists who specialize in the fields relevant to these questions (e.g. endocrinology, developmental psychology, neurobiology).

Shapiro also draws attention to the fact that transgender individuals are at much higher risk for depression and anxiety disorders than the general population. Perhaps one way to improve their well-being would be to give them the benefit of a scientific doubt and accept them for how they express themselves and for how they identify. However, Shapiro takes issue with this suggestion, not only because he thinks trans people are fundamentally mistaken about the science, but because he thinks referring to them by their professed gender identity won’t do them any good (Starts around 4:04).

(3)

Shapiro argues that it makes very little difference to the well-being of trans individuals whether or not we recognize or respect their gender identities. To support this contention, he notes that transgender individuals commit attempt suicide at much higher rates than cisgender individuals (40% vs. 4%), and that a recent study found that “it makes no difference, virtually no difference statistically speaking, as to whether people recognize you as a transgender person or not, which suggests there’s a very high comorbidity between transgenderism, whatever that mental state may be, and suicidality, that has nothing to do with how society treats you” (3:44). If it makes no difference whether or not we recognize someone as transgender, then Shapiro reasons that their suicide rate must be due to the psychological condition itself.

The UCLA study does not say what he says it does. The study found that there were two closely related kinds of risk factors that lead to suicidality among transgender people: “rejection, discrimination, victimization, and violence related to anti-transgender bias and serious mental health conditions” (Haas et al., 2014). Furthermore, the authors conclude that, “Based on prior research and the findings of this report, we find that mental health factors and experiences of harassment, discrimination, violence and rejection may interact to produce a marked vulnerability to suicidal behavior in transgender and gender non-conforming individuals.” I think there is an obvious connection between the negative societal effects and mental health issues: the former causes the latter. But claims of causation require lots of evidence. So, I will leave that up to the relevant experts to determine.

Shapiro follows up on this point by bringing up the low suicide rates among blacks (starts around 4:25). While Shapiro talks exclusively about bullying, I take it that he would maintain that even if you were to factor in any of the other negative societal effects (e.g. discrimination), you would still fail to account for the high transgender suicide attempt rate. He reasons that if things like bullying and discrimination increase the risk of suicide, blacks should be more suicidal than whites since they are bullied and discriminated against more often. But they’re not more likely to commit suicide. Therefore, the way society treats you has nothing to do with suicidality. This isn’t a very good argument.

First, blacks are not a very good comparison group to transgender individuals. Blacks do not face rejection from their families for coming out as black, they are not regularly accused of having a mental illness (or being delusional), and they are not denied health care or work because of their identity. A much better comparison group would be homosexuals, whom have been subjected to similar negative societal effects, albeit to a lesser degree now. Here, we see that the data undermines Shapiro’s argument.

The suicide attempt rates among homosexuals are now between ten and twenty percent, lower than the rates among trans people (Haas et al., 2014; King et al., 2008). However, the rates used to be much higher. For instance, in the 1970’s, a time when homosexuality was significantly frowned upon, a large-scale survey of over five thousand homosexuals found that 40% had attempted suicide (Jay, 1979).*** Since the 1970’s, homosexuals have become much more accepted into society and they are also now less likely to attempt suicide. Two questions arise. Why did the suicide rate of homosexuals decline over the past forty years? And why is the suicide rate still much higher than heterosexuals today?

Societal effects (esp. rejection and disapproval by close family and friends) probably play a significant role in explaining the suicide trends among these groups. A recent study in Canada found that transgender people who had supportive parents were 57% less likely to attempt suicide (Bauer et al., 2015). Those who experienced low levels of hatred and abuse were 66% less likely. Similar suicide risk factors have been also been found among homosexuals (Haas et al., 2010; Van Bergen et al., 2013). Furthermore, a recent study published in JAMA looked at suicide attempts among homosexual teenagers from 2004-2015 (Raifman et al. 2017). The study found that their rate of suicide attempts dropped by 14% in the states where gay marriage was legalized. In states where gay marriage was illegal, the suicide rates remained constant throughout this 11 year period. Again, these findings intuitively support a causal hypothesis regarding suicidal tendencies and social acceptance. Finally, to speak to Shapiro’s specific claims about bullying, a recent meta-analysis concluded that bullying, in general, is a significant risk factor for suicidal ideation and attempts (Van Geel et al., 2014).

Issues surrounding gender and sexual orientation will always arouse conflict between those on the right and left. We hold many opposing beliefs about morality, the existence of God, and how society should be structured, and those beliefs affect how we will respond to scientific claims. We will tend to either ignore findings that contradict our beliefs, or we will deny that they are true. In order to resolve our disagreements about science, we have to be honest about our own beliefs and motives, and be just as honest when we try to represent the views of those we disagree with. Building up straw men, misrepresenting scientific studies, and engaging in tasteless ridicule is not going to narrow the political divide.

References

Bauer, G. R., Scheim, A. I., Pyne, J., Travers, R., & Hammond, R. (2015). Intervenable factors associated with suicide risk in transgender persons: a respondent driven sampling study in Ontario, Canada. BMC Public Health, 15(1), 525.

Haas, A. P., Eliason, M., Mays, V. M., Mathy, R. M., Cochran, S. D., D'Augelli, A. R., ... & Russell, S. T. (2010). Suicide and suicide risk in lesbian, gay, bisexual, and transgender populations: review and recommendations. Journal of homosexuality, 58(1), 10-51.

Haas, A. P., Rodgers, P. L., & Herman, J. L. (2014). Suicide attempts among transgender and gender non-conforming adults. work, 50, 59.

Jay K, Young A (1979). The Gay Report: Lesbians and Gay Men Speak Out About Sexual Experiences and Lifestyles. New York, NY: Summit Books.

King et al. 2008 King, M., Semlyen, J., Tai, S. S., Killaspy, H., Osborn, D., Popelyuk, D., & Nazareth, I. (2008). A systematic review of mental disorder, suicide, and deliberate self harm in lesbian, gay and bisexual people. BMC psychiatry, 8(1), 70.

Raifman, J., Moscoe, E., Austin, S. B., & McConnell, M. (2017). Difference-in-differences analysis of the association between state same-sex marriage policies and adolescent suicide attempts. JAMA pediatrics, 171(4), 350-356.

Van Bergen, D. D., Bos, H. M., van Lisdonk, J., Keuzenkamp, S., & Sandfort, T. G. (2013). Victimization and suicidality among Dutch lesbian, gay, and bisexual youths. American journal of public health, 103(1), 70-72.

Van Geel, M., Vedder, P., & Tanilon, J. (2014). Relationship between peer victimization, cyberbullying, and suicide in children and adolescents: a meta-analysis. JAMA pediatrics, 168(5), 435-442.

* Timestamps are for the video featured on the article linked to near the beginning of the article: http://dailycaller.com/2017/02/10/ben-shapiro-uses-age-to-counter-transgenderism-argument-why-arent-you-60-video/

** By ‘man’, I mean an individual with all or almost all of the psychological properties and behavioral dispositions typically associated with being a male. While this may be contrary to ordinary linguistic practice, this is an accurate way of describing what transgender people are claiming.  

***Same suicide attempt rate as transgender persons today

Cat people

Virtually everyone agrees that is seriously wrong to kill an innocent person. But is it seriously wrong to kill an innocent potential person (e.g. a newborn baby)? In this post, I will outline an interesting argument developed by Michael Tooley (1984) that bears directly upon this question. The argument’s conclusion is that killing potential persons is morally equivalent to not having kids. If one accepts the argument’s chain of reasoning, then either one must accept 1) that it is not seriously wrong to kill a potential person, or 2) that it is seriously wrong not to have kids.

Before fleshing out the argument, I must first discuss a moral principle that Tooley invokes.

Tooley appeals to, what is called, the moral symmetry principle (MSP). The principle is stated very generally to apply to all actions and omissions, but a particular application of the principle entails that there is no morally relevant difference between killing someone and letting them die, provided that one holds fixed all of the morally relevant factors (e.g. the agents’ intentions) and subsequent consequences (see also the equivalence thesis). 

One might at first think the principle is implausible because of obvious counterexamples. For instance, shooting someone with a gun is not morally equivalent to letting someone get shot by someone else. Tooley argues that the principle seems counterintuitive only when one considers cases that involve significantly morally different factors. In the provided example, there are several sources of possible differences between the two cases. One might think that the intentions of the two agents are different. In the first case, the subject shoots someone with a gun, presumably because they intend to kill the person. In the second case, it is possible that the person does not want the person to die, but refrains from acting out of fear. 

Tooley presents several cases where all of the morally relevant factors are the same, allowing one to see the intuitive appeal of MSP. Consider the following two scenarios:

“(1) Jones is about to shoot Smith, when he sees that Smith will be killed by a bomb unless he warns him. Jones’s reaction is: ‘How fortunate—that will save me the trouble of killing him myself.’ So Jones allows Smith to be killed by the bomb, even though he could easily have warned him. (2) Jones wants Smith dead, and shoots him.” (Tooley 1984, pg. 191).

Did the Jones in (1) commit an action that is morally worse than the Jones in (2)? MSP states that the events described in (1) and (2) are morally equivalent. I will let the reader decide whether or not their intuitions align with this principle. (I myself find it fairly intuitive.) With a basic understanding of the principle now on the table, it’s time to turn to Tooley’s argument.

The kitten argument

Imagine that scientists invent a special chemical that turns kittens into potential persons. After some time, the kittens injected with the chemical will come to develop all of the psychological capacities found in humans that we take to be morally relevant (e.g. Self-awareness, rationality, future oriented desires, a biographical sense of one’s self existing over time, etc.). In this world, scientists conduct an experiment in secret using four subjects: three kittens (A-C) and a human infant that is an orphan (D).

Step 1:
Kitten A is intentionally not injected with the special serum.

Kitten B is injected with the special serum but is then injected with a chemical that neutralizes its effect.

Tooley, appealing to the moral symmetry principle, argues that the action involving kitten B is morally equivalent to withholding the chemical from kitten A. That is to say, neutralizing the chemical is morally equivalent to withholding it. Tooley is careful not to say whether these actions are wrong or not. He just points out that they are morally equivalent actions according to MSP.

Step 2:

Suppose that after the experiment is completed, the scientists decide to kill A and B. Given the properties and potentialities they now possess, there does not seem to be a morally relevant difference between acts of killing A or B. It then follows (from Step 2) that there is no moral difference between the combined actions of withholding the chemical from A and then killing it and neutralizing an active chemical in B and then killing it.

Step 3:

Kitten C is injected with the special serum. This time, no neutralizing chemical is added, but it will take a couple of years before it will develop personhood.

Scientists later decide that they do not want Kitten C to reach full personhood but have run out of the neutralizing chemical. Instead, they decide to neutralize the effect of the chemical by killing Kitten C in such a way that it does not undergo any pain or suffering.

If neutralizing the chemical in Kitten A, and then killing it, was not seriously wrong, then neutralizing the chemical by directly killing Kitten C is not seriously wrong. The only difference between the two cases is the number of steps it takes to kill the kittens. Number of steps in an action doesn’t seem to be a morally relevant difference. 

But, via the law of transitivity (LT), if there is no moral difference between killing B or C, then there is no moral difference between killing C (a potential person) or A (a normal cat that was never injected with anything).

Step 4:

A number of behavioral experiments are conducted with the human infant. After the experiments are over, the scientists have no more need for the test subject and do not want it to develop into a person. They decide to kill the baby in such a way that it does not undergo any pain or suffering.

If there is no moral difference between killing C or B, then there is no moral difference between killing B or killing the human baby (D). Moreover, if there is no moral difference between killing B or A, then there is no moral difference between killing D and withholding the chemical from A and then killing it.

“it is prima facie no more seriously wrong to kill a human organism that is a potential person, but not a person, than it is intentionally to refrain from injecting a kitten with the special chemical, and to kill it instead.” (Tooley 192)

A more condensed and formalized version of the argument is as follows:

P1. Withholding the special serum from a kitten is morally equivalent to neutralizing an active serum in a kitten (via MSP).
P2. There is no moral difference between withholding the chemical from A and then killing it and neutralizing an active chemical in B and then killing it.
P3. Neutralizing the effects of the serum by killing the kitten is morally equivalent to neutralizing the chemical by injecting it with a neutralizing chemical and then killing it.
Lemma: Killing a potential person is morally equivalent to intentionally withholding the serum from a kitten and then killing it (via 2, 3 and LT).
P4. There is no intrinsic moral difference between the act of killing a potential cat person and a potential human person
C. There is no moral difference between withholding a personhood chemical from a kitten and killing a potential human person (via lemma, 4 and LT).

While Tooley frames the discussion within a science fiction context, one could understand the act of injecting a kitten with the personhood chemical as analogous to bringing children into the world; withholding the personhood chemical from a kitten as analogous to not having children.

I take it that most would deny that is seriously wrong not to have kids. But if we are to accept that, it would follow that it isn’t seriously wrong to kill a potential person (via abortion or active euthanasia). But the belief that killing potential persons is seriously wrong is another moral claim that widely accepted. Tooley’s thought experiment gives rise to a dilemma with two seemingly unappealing horns to choose from. Either one must accept that it is not seriously wrong to kill human infants, or it is seriously wrong not to have kids.

If the argument is sound, one of these common sense intuitions is wrong. For those who cannot accept either horn of the dilemma, I think the most promising way to defuse the argument would be to try and show that the moral symmetry principle (that is used to derive P1) is implausible. I think the argument is valid, and that its premises are all plausible. Hence, I think I am faced with having to reject one of the common sense intuitions. To make that choice, I would have to carefully assess the two moral claims to see what rational support they may or may not have. In a future post, I may just have to carry out that analysis.

References

Tooley, M. (1984). Abortion and Infanticide. Blackwell Publishing Ltd.

Part II: What compatibilism ought to look like

In the previous post, I argued that the truth of determinism threatens certain conceptions of free will, but not others. That is to say, if determinism were found to be true, there would be some sense in which we don’t have free will. In this post, I will motivate a compatibilist conception of free will that is not threatened by determinism, and argue that it is a legitimate position to hold in the free will debate.

‘Free will’ and ‘moral responsibility’ are closely related concepts. It is often thought that moral responsibility requires the presence of a certain kind of agency. For example, it is thought that young children and non-human animals lack the kind of agency required for moral responsibility but that normal adult humans possess it under certain conditions. Thus, ‘free will’ can be understood as a placeholder for the kind of agency that is required for moral responsibility. There have been at least four proposed features to take its place.

1)      the ability to do otherwise (Kant)

2)    a power that makes us the ultimate source of actions and decisions (Kane)

3)     the ability to do what you want (Hume, Hobbes)

4)    the exercise of a set of psychological capacities (e.g. conscious decision making and self-control) without any internal or external constraints (Frankfurt, Dennett, Nahmias)

The first two candidates are considered to be incompatibilist, in that we cannot have those abilities or powers if determinism is true. The latter two candidates are compatibilist, in that the truth of determinism does not entail that we lack them.

Compatibilists have been accused of changing the subject. It is argued that 1 & 2 satisfy the conditions for the everyday notion of ‘free will’ that most people share. 3 & 4 do not satisfy those conditions and are thus, are not suitable for filling the ‘free will’ placeholder. Thus, compatibilists are changing the subject and talking about something else.  

It is often thought that folk intuitions should carry some weight in our philosophical theorizing, but many agree that they don’t settle the free will/moral responsibility debate. If most people think of ‘free will’ and ‘moral responsibility’’ in terms of 1 and 2, and do not think 3 and 4 are relevant to these concepts, then compatibilists would be in trouble. However, the empirical findings do not indicate that we’re in such a situation (Nahmias et al. 2006, Knobe and Nichols 2009, Nahmias et al. 2014).

If we continue in treating ‘free will’ as a neutral placeholder then the question shifts from ‘is X in accord with what everyday people think?’ to ‘is X the kind of agency that makes us morally responsible?’ The answer to the first question may constrain our theorizing about ‘free will’, but it is the answer to the second question that will make or break compatibilism.

Figuring out whether we have the kind of agency required for moral responsibility may seem straightforward, but philosophers also disagree as to how to understand ‘moral responsibility’. Understanding these differences will further clarify why many compatibilists get accused of cheating. Moral responsibility is typically characterized in one of two ways:

1)      Consequentialist MR- An agent deserves blame or praise because of consequentialist or contractualist considerations (e.g. to protect society from future harm).

2)    Basic desert MR- An agent deserves blame or praise just because they had performed the action.

In defending compatibilism, one has to state what, exactly, is compatible with determinism. Is it the kind of free will that gives us (1) or the kind that gives us (2)?

With two conceptions of moral responsibility on the table, there are now two kinds of compatibilism one could defend:

C1. Determinism is compatible with consequentialist MR
C2. Determinism is compatible with Basic desert MR

While most philosophers identify as compatibilists, it isn’t clear how many subscribe to C1 or C2. But as Derk Pereboom (2014) points out, C1 is not a philosophically interesting position to hold, as virtually everyone would be a compatibilist under that characterization. If no one disagrees with the thesis, then there isn’t an interesting philosophical debate to be had. The adoption of C1 in response to claims of incompatibilism is a lot like the following scenario.

Imagine the following conversation between two subjects, S1 and S2:

S1: I think Tom and Sarah are incompatible with one another.

S2: I think Tom and Sarah are compatible with one another.

S1: I guess we’ll have to agree to disagree.  

Problem: They’re actually talking about two different individuals named ‘Tom’.

Some philosophers seem to be in the same boat as S1 and S2. Here’s an illustration with philosophical content:

Incompatibilist: I think that determinism is incompatible with being morally responsible.

Compatibilist: I think that determinism is compatible with being morally responsible.

Incompatibilist: We have a substantive philosophical disagreement.

Problem: There’s an equivocation over the use of the term ‘moral responsible’. The two speakers mean different things.

Dan Dennett, a prominent compatibilist, seems to defend C1. He argues that holding people responsible can be justified for instrumental reasons. It’s useful to hold people accountable for their actions because it encourages good behavior and discourages bad behavior. Furthermore, Dennett points out that our social practices of blaming and praising others for their actions is one that is good, and possibly necessary for a functional society. But if that’s the only justification for holding someone responsible, then Dennett is clearly not defending C2. Dennett’s case effectively translates to saying that we ought to hold people as if they are responsible in order to achieve good end (AKA, a defense of C1). Following Pereboom, I think the compatibilist needs to defend C2 in order for there to be a substantive philosophical debate. Compatibilists who defend C1 can rightfully be accused of changing the subject, but there are plenty of compatibilists who do defend C2 (e.g. Nahmias). 

If the compatibilist conceptions of free will are enough to give us moral responsibility—in the basic desert sense—then there is no bait and switch, and compatibilism is a legitimate position to hold. In the following post, I will provide a defense of compatibilism against a number of empirically informed objections. It is said that certain findings in neuroscience and social psychology undermine our sense of agency. I will argue that these findings do not convincingly demonstrate that we lack responsibility for all of our actions. 

Works cited

Björnsson, G., & Pereboom, D. (2014). Free will skepticism and bypassing. Moral psychology, 4, 27-35.
       

Murray, D., & Nahmias, E. (2014). Explaining away incompatibilist intuitions. Philosophy and Phenomenological Research, 88(2), 434-467.

Nahmias, E., Morris, S. G., Nadelhoffer, T., & Turner, J. (2006). Is incompatibilism intuitive?. Philosophy and Phenomenological Research, 73(1), 28-53.

Nichols, S., & Knobe, J. (2007). Moral responsibility and determinism: The cognitive science of folk intuitions. Nous, 41(4), 663-685.

*This post is part of a series on free will.