Flickers of Freedom

Cyber Monday

In the next few days my online time will move from concerns with Flickers posts to Christmas shopping.  I greatly appreciate all the helpful comments I received throughout the month and am thankful to Thomas for giving me this opportunity.  Only a few years ago I started focusing on issues in free will instead of core aspects of metaphysics.  So, in many ways, my thinking about these issues is still quite naive and a work in progress.  For people like me, both the insights that can be gained by the comments that people posted and some of the connections I was able to form throughout the month are very valuable.  I strongly encourage graduate students, new PhDs or people who recently moved from concerns in other areas of philosophy to the topic of free will to contact Thomas about being a featured author.  Thanks to everyone who helped make this experience valuable for me.    

 

The Summer of Sixty Nine

1969 was an ideal time to be a compatibilist.  Most people in the debate still believed that the ability to do otherwise was consistent with determinism.  Yet, thanks to publication of Frankfurt's "Alternate Possibilities and Moral Responsibility", there were also good reasons to believe that even if the incompatibilist could show that the ability to do otherwise was not consistent with determinism, agents could still be morally responsible in a deterministic world.  This two pronged defense of compatibilism seemed to give the compatibilist an amazing advantage.  What I want to suggest here is a way to return to the good ole days. 

Today, classical compatibilists and semi-compatibilists seem to spend almost as much time arguing with each other as confronting incompatibilists.  Dispositional compatibilists frequently argue that Frankfurt cases are nothing more than finkish dispositions.  A live wire can be attached to a safety device that would cut the circuit upon sensing the approach of a conductor.  While the wire is live, it would not deliver a charge to a conductor that made contact with it.  Instead, it would have lost that dispositional property upon sensing the approaching conductor.  Dispositional compatibilists often argue that the same thing is true in a Frankfurt case.  Jones has the ability to do otherwise before the device kicks in.  Yet, would lose that ability once the device is activated. 

 Yet, is this really a problem for Frankfurt's argument?  Frankfurt argues that both on a counterfactual account of the ability to do otherwise and on the libertarian variety, the ability to do otherwise is not required for moral responsibility.  It seems to me that his argument was extraordinarily successful.  What Frankfurt did not argue is that the ability to do otherwise is not necessary for moral responsibility on any conceivable account of abilities.  He never claimed to have a crystal ball.

The strongest defense of compatibilism is to adopt a compatibilist account of the ability to do otherwise while embracing the lesson we learn from Frankfurt cases.  Frankfurt teaches us that neither the incompatibilist nor the conditional compatibilist accounts of the ability to do otherwise are necessary for moral responsibility (not that PAP is false, but that it is false on the interpretations of the ability to do otherwise that were present when Frankfurt wrote).  So, even if the incompatibilist can prove that her account of the ability to do otherwise is the correct account, she still has more work to do in order to show that we cannot be morally responsible in a deterministic world.  Embracing a compatibilist account of the ability to do otherwise adds the advantage that even if the incompatibilist could show how Frankfurt cases fail, she would still have more work to do.  She would have to show why an incompatibilist account of the ability to do otherwise is better than the compatibilist account.  Shouldn't compatibilists and semi-compatibilists strive to return to the good ole days when compatibilism about the ability to do otherwise was extraordinarily popular, and Frankfurt added additional support for compatibilism.     

Related articles

My Plan

The Direct Argument

Neo-Aristotelian Dispositional Compatibilism

What the laws of nature are ought to impact how we think of free will.  For example, Humean accounts seem congenial for being a compatibilist about the ability to do otherwise.  What appears to be the most popular non-Humean account, however, also provides some support for compatibilism. 

Dispositional compatibilists note that fragile glasses that are never stressed and never shatter can exist in a deterministic world.  While it is determined that the glass will never shatter, it still has the liability of doing so.  Similarly, a strong agent in a deterministic world may never lift a feather.  Yet, being strong, the agent has the ability to do so.  Similarly I am confident that the chair of the mathematics department at my university is able to solve the equation "2+2=x."  I am not, however, confident that the chair of the mathematics department at my university is currently awake.  There seems to at least be a sense of 'able' that stems from the dispositional properties of agents.

One manner of objecting to dispositional compatibilism is to note that two different modal concepts may be in play.  One is grounded in the deterministic laws of the world, the other is grounded in dispositional properties.  Further, one might insist that the important sense of 'ability' for the free will debate is the one grounded in the laws and not the one grounded in dispositional properties.           

Yet, on the main non-Humean account of laws of nature, this response to dispositional compatibilism makes no sense.  (It is noteworthy that Vihvelin defined dispositions in terms of laws.  By doing so, she cannot embrace the main non-Humean account of laws.  Consider the conditional, 'if electrons attracted each other atoms would collapse.'  Since this conditional seems non-trivially true, we seem committed to the claim that there are worlds where electrons attract each other.  So, while it may be physically necessary that electrons repel each other, it is not metaphysically necessary. Yet, an equally plausible thing to claim is that if electrons attracted each other, they wouldn't be electrons.  Part of what it is to be an electron is to have a negative charge.  Further, part of what it is to have a negative charge is to repel other negatively charged particles.  If an object attracts other objects of its kind, then necessarily that object is not an electron.  When we claim that 'if electrons attracted each other atoms would collapse' what we really mean is that if a different object played the electron role in an atom and that object attracted objects of its kind, then atoms would collapse.  While we ought to charitably interpret the claim 'if electrons attracted each other atoms would collapse' in a manner so that it comes out true, it is equally essential to be more critical when we try to derive the more significant conclusion that it is only a contingent matter that electrons repel each other. 

If part of what it is to be an electron is to behave in a certain manner, then electrons essentially have some dispositional properties.  Electrons act as they do because they are what they are.  Contrary to Descartes' claim when he formed the concept of laws of nature, objects do not need to wait for God to dictate laws to them in order to know how to behave.  Instead, the behavior of electrons stem from their natures. 

The neo-Aristotelian account of laws of nature claims that laws of nature are grounded in the essential dispositional properties of natural kinds.  If this account is correct, then thinking about laws just is thinking about dispositional properties.  If thinking clearly about dispositional properties leads to a compatibilist account of abilities, then thinking clearly about laws does so as well.  On this account, we cannot respond to the dispositional compatibilist that there is both a sense of 'ability' grounded in dispositions and another grounded in the laws.  This is because in both cases we are really thinking about the exact same thing.

My main purpose in this post is not to defend either the neo-Aristotelian view of laws of nature nor to defend dispositional compatibilism.  Doing either of these things would require much more space than would be manageable for this post.  Instead, the point is that how we think about free will depends upon what we think about laws of nature.  Thinking, instead, of merely the sentences that describe laws of nature blurs our capacity to think more clearly about free will.   Seriously considering free will requires thinking seriously about what laws of nature are.

Related articles

My Plan

Iffy Abilities

Iffy Abilities

Incompatibilists frequently chide compatibilists for providing iffy accounts of the ability to do otherwise.  They insist that properly thinking about agent's abilities requires that we hold the laws and the past fixed.  Yet, when we explore our capacity to reason modally, there are excellent reasons to believe that incompatibilists are making a huge error. 

Philosophers frequently get things epistemically backwards.  We too frequently concentrate on when we learned a formal semantics for a concept and then assume that the order in which we learn the formal semantics maps the order in which we learn the basic concept.  A great example of this occurs in the case of modality and counterfactual conditionals.  We learn the formal possible worlds semantics for metaphysical and nomic necessity prior to learning the formal semantics for counterfactual conditionals.  This frequently causes philosophers to think of reasoning about metaphysical and nomic necessity as being prior to reasoning about counterfactuals.  Nevertheless, recent work in the psychology of modal reasoning indicates that this is a mistake.   In fact, there are good reasons to believe that our capacity to think about metaphysical and nomic modalities (i.e. our ability to reason philosophically) is an exaptation from our ability to reason counterfactually.    (An excellent book to begin thinking about the psychology of modal reasoning is Hoerl, McCormack & Beck's "Understanding Counterfactuals, Understanding Causation" James Woodward's paper in that collection is significantly important for philosophers wading into the psychological literature on this topic). 

Counterfactual reasoning develops surprisingly early.  Children develop this capacity somewhere around the age of 3.  If counterfactual reasoning has intimate ties to causal reasoning, it is not too surprising that it would develop so early.  In fact, even as young as two children engage in games of make believe.  If I were a lion, or a dragon, or a doorstop, this is how I'd behave and they act accordingly. Yet, at these young ages, reasoning about nomic or metaphysical necessities is still absent.  Further, if nomic reasoning is reasoning about what is possible given the laws of nature, it is not only much later that individuals start thinking about nomic possibilities but it is also significantly later that societies start thinking about nomic necessity.  The very concept of a law of nature does not arise until Descartes.       

Similarly, the method in which we learn about (and teach) metaphysical necessity relies upon a prior capacity to reason counterfactually.  Consider how you try to convince your students that there might be propositions that are metaphysically necessary but only a posteriori knowable.  Usually, we have the student suppose that there is a material that looks like, tastes like, and smells like water.  In fact, this substance acts exactly like water until we attempt to discover what its molecular composition is.  When doing the molecular analysis we discover that it is composed of XYZ instead of H2O.  Is that substance water?  The question we are asking them is if there were a substance like XYZ would it be water. 

In fact, reasoning about metaphysical and nomic necessity just is reasoning about counterfactuals.   Just as we can define necessity in terms of possibility, we can define other forms of modal reasoning in terms of counterfactual reasoning.  Something is metaphysically necessary if and only if it would happen no matter what.  In other words, P is metaphysically necessary if and only if any antecedent counterfactually implies P (p. 24 of Lewis's Counterfactuals and Chapter 5 of Timothy Williamson's The Philosophy of Philosophy have great discussions of this connection). 

So, what does this discussion have to do with free will?  Any adequate account of the ability to do otherwise must make this concept a modal concept.  If an account of the ability to do otherwise requires that we neither alter the past nor alter the laws of nature in a deterministic world, then this account demands that we do not employ the resources of counterfactual reasoning when we examine whether an agent could have done otherwise.  If all modal reasoning is a species of counterfactual reasoning, then these accounts demand that we do not think of the ability to do otherwise modally.  Restricting our attention to only worlds with the same past and the same laws in a deterministic world does not provide us with a modal notion but instead is merely a long winded way of asking what actually occurred. 

One might object that reasoning about nomic necessity requires that we hold the laws of nature fixed.  After all, one might naively believe that something is nomically possible if and only if it occurs in some world that shares the same laws of nature as the base world and nomically necessary if and only if it occurs in every world that shares the same laws as the base world.  Yet, this is clearly an inadequate account of nomic modalities.   Consider a very simplistic world, one in which all that exists is a single stationary electron with no forces acting upon it.  What will occur at the next moment?  The correct answer is that it will stay where it is at.  A poor answer is that nomically anything follows from that assumption because the assumption is nomically impossible. 

Yet, on a Humean account, the world depicted cannot share the same laws as our own.  After all, the summary that provides the best balance of simplicity and strength for that world would contain many less laws than our own world.  Laws about how negatively charged objects effect positively charged objects would add gratuitous complexity to the summary that is not offset by increases in strength.  On the neo-Aristotelian account laws are grounded in dispositional properties essentially held by natural kinds.  Since there are less fundamental natural kinds in the simplistic world, it contains less laws than our own.  On Armstrong's account of laws of nature, laws are necessitation relations between universals.  Armstrong's account of laws, conjoined with his Aristotelian position about universals, entails that there are less laws in the simplistic world.  After all, there are less universals related by necessitation relations.  On all the main accounts of laws of nature, we cannot hold the laws of nature fixed when considering nomic necessity.     

Incompatibilists frequently chide classical compatibilists for adopting constitutionally iffy accounts of the ability to do otherwise.  Yet, if all modal reasoning is a species of counterfactual reasoning the more serious complaint is that incompatibilists do not offer constitutionally iffy accounts of the ability to do otherwise.  If you were asked to solve the equation 6+3=x without employing any mathematical reasoning it seems that you cannot succeed.  Yet, this teaches us noting interesting about mathematics.  Similarly, if you were asked to consider whether the ability to do otherwise is compatible with determinism without employing your capacity to reason modally you would be unable to do so.  Yet, this teaches us nothing about agent's abilities.  Since incompatibilists demand that we hold the laws and the past fixed when we consider the ability to do otherwise, incompatibilists are demanding that we consider what an agent is able to do without employing our capacity to reason modally.  So far, it seems that the only modal accounts of the ability to do otherwise are compatibilist accounts.  

 

The Consequence Argument

The Consequence Argument assumes that nobody has any control over the laws of nature or what occurred in the distant past.  Rule Beta□  states that if an agent has no control over P and has no control over the fact that P logically entails Q, then the agent has no control over Q.  Since the laws of nature and a complete description of the events at any moment entail all true propositions in a deterministic world, the Consequence Argument concludes that nobody has any control over anything in a deterministic world. 

 Part of what makes responding to the Consequence Argument more challenging than the Direct Argument is that there may be sound versions of the argument.  Frequently, the Consequence Argument is stated in terms of being able to render propositions false.  Nobody can render the laws of nature or propositions about the distant past false.  Further, these propositions entail all true propositions in a deterministic world.  So, nobody can render any proposition false in a deterministic world.  Perhaps, these versions of the argument are sound.  Yet, incompatibilists employ the argument to demonstrate that an agent has no control over any proposition in a deterministic world.  Having some control the states of affairs that make a proposition true, however, is not the same thing as having the capacity to render that proposition false.  Perhaps much of the Consequence Argument's appeal stems from semantic ascent.

 Suppose again that Bob and Sue independently rob the same bank.  Sue cannot render the proposition Bob robbed the bank false.  Further, Sue cannot render it false that the proposition Bob robbed the bank logically entails that either Bob robbed the bank or Sue robbed the bank.  So, Sue cannot render it false that either Bob robbed the bank or Sue robbed the bank.  This seems unproblematic.  How can Sue render the disjunction false unless she can render both disjuncts false?  Similarly, consider the argument form employed in the last post: 

1. Ns P (Assumption for > introduction)

2. Ns □(P > (PvQ)) (Beta□, v introduction, alpha)

3. Ns (PvQ) (1, 2)

4. Ns P > NRs (PvQ) (> introduction, 1–3)

When we read Beta□ as transferring an inability to render a proposition false, this argument form seems unproblematic.  After all, if P is true and S lacks the capacity to render P false, then S lacks the capacity to render PvQ false as well.  Yet, having at least some control over the states of affairs that make a proposition true is not the same thing as having the capacity to render a proposition false.  An agent who produces a truthmaker for a proposition and has the capacity to refrain from producing a truthmaker for that proposition has a significant amount of control over the states of affairs that the proposition is about even if she lacks the capacity to render that proposition false.        

 Bob has no control over the fact that Sue robbed the bank.  Further, the fact that Sue robbed the bank entails that someone robbed the bank.  So, according to an ontological interpretation of Beta□ Bob has no control over the fact that someone robbed the bank.  Yet, Bob robbed the bank, and could have done otherwise.  Bob provided one of the truthmakers for the claim that at lesat one of them robbed the bank, and could have done otherwise.  Rule Beta□  entails that when two different agents independently produce truthmakers for a proposition neither has any control over the state of affairs that the proposition is about.  Yet, typically when two different agents produce truthmakers for a proposition both have at least some control over the states of affairs that the proposition is about.            

 In standard preemption cases an agent produces a truthmaker for a proposition, could have refrained from producing a truthmaker for the proposition, yet lacks the capacity to render that proposition false.  Ravizza's counterexample to B exploited this fact.  Similar problems for B□ are easy to generate once we recognize that Ravizza's case is merely an instance where the logic of semantic relations differs from the logic of truthmakers. 

McKay and Johnson (1996) produced a counterexample to Beta (without the box).  Suppose that a fair coin had never been flipped.  While Jane only once had an opportunity to flip the coin, she freely decided not to.  By Rule alpha, Jane has no control over the logically necessary truth that 'if the coin did not land on heads, then if the coin did not land on tails, then the coin did not land on heads or tails.'  Further, since the coin is a fair coin, Jane has no control over the direction in which the coin lands.  So, Jane has no control over the fact that the coin did not land on heads.  By Rule Beta it follows that Jane has no control over the proposition 'If the coin did not land on tails, then the coin did not land on heads or tails.'  Yet, again since Jane has no control over the direction in which a fair coin lands, Jane has no control over the fact that the coin did not land on tails.  A second employment of rule Beta generates the result that Jane has no control over the fact that the coin did not land on heads or tails.  Yet, Jane certainly has some control over this fact.  After all, had she flipped the coin, it would have landed on either heads or tails.  The standard response to McKay and Johnson has been to replace Rule Beta with Rule Beta□.

Yet, Johnson and McKay's counterexample is merely one manner of demonstrating that the logic of truthmaking and not the logic of truthbearing ought to be central to the free will debate.  In the truthmakers literature there is a debate over whether something can be a truthmaker for a disjunction without being a truthmaker for either disjunct.  Read (2000) employs a case where the only horses in a race are Valentine and Epitaph.  The race's occurrence guarantees that either Valentine or Epitaph wins.  Yet, the race's occurrence neither guarantees that Valentine wins nor guarantees that Epitaph wins.  So, while the race's occurrence makes it true that either Epitaph or Valentine wins it does not make either disjunct true.  The same point could be made with the coin example.  Flipping the coin makes it true that the coin lands on either heads or tails.  Yet, since the coin flip does not guarantee what direction the coin would land, flipping the coin neither makes it true that the coin lands on heads nor makes it true that the coin lands on tails.  The reason the same arguments can be employed when considering free will and truthmaking is that they share the same logic. 

 Interestingly, both in the Direct Argument and the Consequence Argument counterexamples were produced for the early transfer rules precisely where the logic of truthmaking would expect them to occur.  In both cases, the response has been to merely add a box.  Yet, this does not resolve the underlying problem with these arguments.  Free will and moral responsibility are about the world and not merely our descriptions of it.  

In cases where a disjunction has two true disjuncts, an agent can have some control over the states of affairs that the disjunction is aobut without having control over both disjunts.  Yet, this is inconsistent with Beta□ of the Consequence Argument.  Beta□ has gone unchallenged for too long.  It is invalid. 

 

 

The Direct Argument

 

In the next two posts I plan on discussing why taking the distinction between metaphysical and semantic claims seriously generates problems for the Direct Argument and the Consequence Argument.   The argument in this post stems from my paper, “Truthmakers and the Direct Argument.” While I’d prefer using my time as a featured author to get assistance on some of my current projects, one of my main concerns is how a similar line of argument can be employed against the transfer principles of the Consequence Argument.  This will be the topic of my next post.  Yet, since the Direct Argument is a significantly easier target than the Consequence Argument it seems useful to explore the easier problem first. 

 Rule B of the Direct Argument claimed that if an agent is not responsible for P and not responsible for the fact that if P is true then Q is true, then the agent is not responsible for Q.   Ravizza, however, employed cases of causal overdetermination to generate counterexamples to B.  In response to Ravizza's challenge, Ted Warfield suggested replacing B with B□ which claims that if an agent is not responsible for P and it is logically necessary that P entails Q then the agent is not responsible for Q.  There are, however, cases of logical overdetermination that are as problematic for B□ as Ravizza's challenge was for B. 

  B□ only seems plausible when we concentrate on propositions instead of concentrating on what makes those propositions true.  Propositions are either true or false.  Yet, when we shift our attention to what makes a proposition true instead of merely concentrating on whether the proposition is true, more options arise.  Something can make a proposition true, make the negation of that proposition true, or neither make the proposition nor its negation true.  An agent can have the ability to produce a truthmaker for a proposition, and have the ability to refrain from producing a truthmaker for a proposition, without having the capacity to make that proposition false.  Such cases frequently produce counterexamples to  B□.  Here, I shall explore problems for B□ that occur for existential quantification and disjunctions. 

Suppose that Bob and Sue independently rob the same bank.  The fact that Bob robbed the bank logically entails that either Bob or Sue robbed the bank.  Yet, according to Rule A of the Direct Argument, Sue cannot be responsible for a logically necessary proposition.  So, Sue is not responsible for the fact that Bob robbed the bank logically entails that either Bob robbed the bank or Sue robbed the bank.  Further, Sue is not responsible for the fact that Bob robbed the bank (after all, their actions were done independently).  So, by Rule B□ Sue is not even partially responsible for the fact that either Bob or Sue robbed the bank.  Similar reasoning demonstrates that Bob is also not even partially responsible for the fact that either Bob or Sue robbed the bank.  While both Bob and Sue are partially responsible for the fact that at least one of them robbed the bank, Rule B□ of the Direct Argument entails that neither are responsible. 

The Direct Argument has an extremely implausible exclusivity built into it.  When a disjunction has two true disjuncts someone can be at least partially responsible for the disjunction by being responsible for only one of its disjuncts.  Yet, B□ entails that someone can only be even partially responsible for a disjunction with two true disjuncts by being responsible for both disjuncts.  Suppose that an agent is responsible for Q and not responsible for P.  The following argument form demonstrates that the agent is not even partially responsible for PvQ.     (Symbols are not transferring for me from my word document so I am using > in place of a horseshoe.)

1. NRs P (Assumption for > introduction)

2. NRs □(P > (PvQ)) (v introduction and Rule A)

3. NRs (PvQ) (B□, 1, 2)

4. NRs P > NRs (PvQ) (> introduction, 1–3)

So, B□ entails that an agent can only be even partially responsible for a disjunction with two true disjuncts if she is at least partially responsible for both disjuncts. 

 While both are partially responsible for the fact that at least one of them robbed the bank, Rule B□ provides the result that neither are even partially responsible for this disjunction.  A similar problem occurs for existential quantification.  The fact that Bob robbed the bank logically entails that someone robbed the bank.  By Rule A, Sue is not responsible for the fact that Bob robbed the bank logically entails that someone robbed the bank.  Since Sue is not responsible for Bob's actions, Sue is not responsible for the fact that Bob robbed the bank.  So, by Rule B□ Sue is not even partially responsible for the fact that someone robbed the bank.  Similar reasoning shows that Bob is also not even partially responsible for the fact that someone robbed the bank.  Yet, clearly when the police discover that someone robbed the bank and want to discover who is responsible, the correct answer is not nobody.  Both Bob and Sue are at least partially responsible for the fact that someone robbed the bank, but this is inconsistent with Rule B□ .      

Ravizza, M. 1994. “Semi-Compatibilism and the Transfer of Nonresponsibility,” Philosophical

Studies, 75, 61-93

Warfiled, T. 1996. "Determinism and Moral Responsibility are Incompatible," Philosophical Topics, 24 p. 215-226

 

My Plan

I wanted to thank Thomas for allowing me to be the featured author for this month. Throughout the month, I plan on discussing the Direct Argument, the Consequence Argument, how the metaphysics of laws of nature ought to change the free will debate, whether there is an incompatibilist account of the ability to do otherwise, and how the standard account explaining the apparent directionality of time  ought to change the free will debate.  While I hope to cover a wide range of topics, each of these discussions stem from a common theme.  In this post, I will explore the general concern that gives rise to the posts I shall be making throughout the month. 

Much of the free will literature looks odd from the perspective of a smart undergraduate student. Frequently, however, they learn to speak the language the free will debate is carried out in and those of us immersed in it long enough tend to lose track of why the debate initially seemed strange. Nevertheless, perhaps the debate goes wrong precisely where we teach students not to worry. Perhaps the free will debate can be improved by taking seriously those areas that initially seem strange instead of learning to ignore the oddity.    

To see why much of the free will literature looks odd when first encountered, consider how we typically define determinism.  A world is deterministic if and only if a complete description of the events that occur at some moment in time combined with the laws of nature entail all true propositions.  Notice, however, that entailment is a semantic notion.  P entails Q if and only if it is not possible for P to be true and Q to be false.  Yet, we must also restrict P and Q to only those sorts of things that can have truth values.  After all, it is not possible for my left shoe to be true and my right shoe to be false.  Yet, my left shoe does not entail my right shoe.  Entailment is a relationship between the bearers of truth.  So, if the laws of nature entail anything, laws of nature must be propositions or sentences.  Yet, nobody has ever been constrained by a proposition.  Propositions do not cause anything to occur.  Similarly, on this definition, determinism is not about relationships between the events that occur in a world.  Instead, determinism is about relationships between descriptions of these events.       

A similar concern can be seen when we think of where we attribute moral responsibility in arguments like the Direct Argument.  In the Direct Argument, we transfer a lack of responsibility for the truth of a proposition to a lack of responsibility for the truth of any proposition that it entails.  Yet, we never praise of blame an agent for the truth vale of a proposition.  People are responsible for what they do and the states of affairs that they bring about, not directly for descriptions of the world.  We act in the world, not on the semantic plane.  We might blame Sally for killing Kate.  We might even blame Sally for making the proposition "Kate was killed" true.  We do not, however blame Sally for the proposition "Kate was killed" being true.  It seems that it is a category mistake to attach responsibility to propositions instead of to those things that make the proposition true.  Responsibility is attached to acts and events in the world and not directly to propositions or descriptions of the world.    

Prior to Kripke's work it was quite common for philosophers to slide between metaphysical, epistemic, and semantic claims.  After all, it was commonly believed that all necessary truths are also a priori and analytic.  Kripke's demonstration that there are necessary truths that are only a posteriori knowable provided one of the first clues that sliding between semantic, epistemic, and ontological concepts can be troubling. 

More importantly, theorists in the truthmaker literature have recently reached a consensus (something rare among philosophers) that truthmaking does not follow classical entailment. (Rodriguez-Pereyra 2006, p. 187).    While there have been many objections to the claim that truthmaking is preserved through classical entailment, I only have space to consider one of the earliest criticisms here.  Consider the explosion problem for necessary truths (Restall, 1996).  Suppose that truthmaking were preserved through classical logic.  In that case, if o is a truthmaker for P, and P classically entails Q, then o is a truthmaker for Q.   Clearly, this post makes it true that 'this post exists.'  Yet, since necessarily true propositions are entailed by any proposition, 'this post exists' classically entails that '2+2=4'.  If we assume that truthmaking is preserved under classical entailment, it follows that this post makes it true that 2+2=4.  Yet, clearly, this post does not make it the case that 2+2=4, so we have excellent reason to believe that truthmaking is not preserved through classical entailment.  

If the logic of truthmaking is distinct from classical entailment, and classical entailment is the appropriate logic for semantic issues, then we have even greater reasons to be leery when discussions of metaphysical topics like free will and moral responsibility are addressed as if they were debates about propositions and entailment relations.  Given how extensively semantic ascent is employed in the free will literature, this problem seems like it ought to have significant ramifications for the free will debate.  After all, the very definition of determinism that is most frequently employed in the free will literature ought to be enough to provide reason to think that something has gone fundamentally wrong in the debate. 

The reason that so much of the free will debate is carried out in its current manner is that much of the contemporary free will debate took its current shape when the linguistic turn was still quite popular.  During the linguistic turn, it was thought that we could avoid difficult metaphysical questions by replacing them with questions about semantics.  Free will is one of the most challenging philosophical topics because so many other areas of philosophy bear on what we ought to think about free will.  If we could bracket these difficult topics through the process of semantic ascent, the free will debate becomes significantly easier.  It would be nice if we could ignore what laws of nature are and instead merely focus on propositions about laws of nature when considering free will.  Yet, the failure of the linguistic turn, lessons we learned from Kripke, and the contemporary truthmakers literature coming to the conclusion that the appropriate logic for semantic concerns is not the same as the appropriate logic for ontological concerns all ought to make us deeply suspect of this change in topic.   

I am convinced that incompatibilism about moral responsibility and incompatibilism about the ability to do otherwise both gain a significant amount of their appeal through an illicit use of semantic ascent.  If I am right, much of the contemporary free will debate needs to be radically revised. 

Restall, G., 1996, “Truth-makers, Entailment and Necessity,” Australasian Journal of

Philosophy, 74: 331–340.

Rodriguez-Pereyra, G. (2006). Truthmakers. Philosophy Compass, 1, 186–200.

Passing the Torch!

With Halloween now behind us and the grueling holiday season now formally upon us (whether we're ready or not), November is finally here. While this means the semester is almost over, it also means that it is time to once again pass the torch from one Featured Author to the next. Before we do, I first wanted to thank Justin Coates for doing a splendid job during this past month of stoking the dialectical fires here at Flickers! Combing philosophical rigour with a touch of humor made for some fun and illuminating posts. It's exciting to see how much interesting work is being done by the latest generation of action theorists and moral philosophers. Hopefully, this means our already interesting field will continue to grow and thrive! So, please join me in thanking Justin for his contributions during October.

That said, it's now time to hand off the philosophical baton to our next Featured Author--Professor Charles Hermes.  As a former fellow graduate student at FSU with Charles, I can personally attest to how passionately he is committed to philosophical discourse! While we have focused on different topics in our respective dissertations under the tuteleage of Al Mele, I always enjoyed bantering back and forth with Charles about all issues metaphysical.

These days, Charles is a visiting assistant professor in the philosophy department at the University of Texas--Arlington. Charles specializes in metaphysics and philosophy of action.  Charles’s main philosophical interests have been on laws of nature, dispositions, counterfactuals, and truthmaking.  Recently, however, Charles has turned the majority of his attention to how these debates can influence how we ought to think about free will. Needless to say, his work should be right up our readers' alleys here at Flickers of Freedom. So, please join me in welcoming him to the show!  I look forward to yet another month of engaging and illuminating philosophical conversation!

p.s. As I mentioned a few weeks ago, please help both Charles and the blog more generally by taking a few minutes to use the social media icons at the bottom of each post to spread the word.  One of the chief benefits of the Featured Authors series for junior philosophers is that it can increase the visibility of their work.  So, please help by doing your part! It takes very little effort for you and it can make a big impact for the authors.

p.p.s. Here is the upcoming schedule for the Featured Author Series:

December:           Peter Tse  

January:               Justin Caouette

February:             V. Alan White

If you are interested in doing a stint as a Featured Author, please send me an email. Similarly, if you have some suggestions for others who might be interested--especially people who are not already regular contributors to the blog--please let me know.  I am looking to start filling up the line up for 2015!