7 ms·

Could someone with better understanding explain compatibalism to me? I’m struggling to grasp how it can allow for both determinism and free will.

skissane 8y ago
There is no single universally agreed definition of the term "free will". So, it is possible to define "free will" in such a way that it is compatible with determinism (e.g., "my will is free if its decisions are the product of my ordinary processes of thinking–even if those ordinary processes are determined by natural laws–as opposed to the product of any extraordinary external interference such as drugs, brainwashing, mind control implants, etc"). Part of the dispute is over whether such a definition of "free will" matches people's everyday pre-philosophical usage, and also whether it works for the purposes of other disciplines in which the concept of "free will" might be invoked, such as ethics or law. Compatibilists would say it is close enough, and can be made to work for those purposes. Incompatibilists say it is a radical redefinition, and undermines those purposes.
- MrScruff 8y ago
  So it seems the crux of the matter is how we regard the process of a person making a decision. A compatibilist would say that for the purposes of discussion we describe this event as the exercising of will, even though from a physics perspective the outcome could be predetermined. Is that correct? If so that seems similar to a recent discussion on Bohr's view of the irreducibility of biology to physics. https://news.ycombinator.com/item?id=18104508 https://news.ycombinator.com/item?id=18104508
  - skissane 8y ago
    Certainly that's one way a compatibilist could put it – that "free will" is part of psychology, which ultimately reduces to/emerges from/supervenes upon lower layers of biology/chemistry/physics which may be fundamentally deterministic. (Although exactly how deterministic they are may depend on which interpretation of quantum mechanics is correct.) A point worth adding – there are different forms of determinism. Physical determinism says people's decisions are predetermined by physical laws and initial conditions. Theological determinism says people's decisions are predetermined by God's will. A compatibilist could be either. (Although most compatibilists you'll encounter these days are in the physical not theological camp.)
- naasking 8y ago
  I think you pretty fairly describe the current landscape of this debate, which is unusual in itself. Kudos! > Part of the dispute is over whether such a definition of "free will" matches people's everyday pre-philosophical usage, and also whether it works for the purposes of other disciplines in which the concept of "free will" might be invoked, such as ethics or law. Fortunately, this is an empirical question that has been explored by experimental philosophy [1]. It turns out that most lay people employ Compatibilist reasoning, so I think this question has also been answered. [1] The latest that I've seen on the topic: https://www.researchgate.net/publication/274892120_Why_Compatibilist_Intuitions_are_not_Mistaken_A_Reply_to_Feltz_and_Millan https://www.researchgate.net/publication/274892120_Why_Compa...
  - fjsolwmv 8y ago
    Lay people, yes, but lay people also believe in ghosts and angels. Lay people believing it doesn't make it true. Scientific reasoning shows compatibilism is inaccurate, But just how like how belief in an scientifically absurd afterife can make people happy, being scientifically correct is not equivalent to having a good life. And this is a potential definition of the essence religion or wait, properly opposed to science: faith can make you happy even when you are wrong, and losing that faith makes you unhappy. Therefore, having faith is the correct choice for an ideal human life. It's right to be wrong.
    - naasking 8y ago
      > Lay people, yes, but lay people also believe in ghosts and angels. Lay people believing it doesn't make it true. I suggest you read the study and others like it. If you flat out ask people what they believe, with no further context, they tend to support incompatbilist intuitions. But when they test their responses to moral questions, they employ Compatibilist moral reasoning, and agree with claims that are Compatibilist.
- clairity 8y ago
  i'm leaning toward a belief in an outcome where we eventually prove strong determinism false, simply because of the probabalistic underpinnings of the universe. in that scenario, compatibalism would be an acceptance of a weak determinism based on the likeliness of probabalistic outcomes (particularly at larger scales), rather than an absolute certainty.
garmaine 8y ago
Communication over the internet is hard and I’m not sure if I can convey the idea briefly in a HN post, but the idea is simple. The assumption that free will and determinism are at odds with each other is ungrounded when you actually look into it. “Non-determinism” implies randomness which is NOT what we are talking about when we say something like “Susie made a decision.” She makes a decision for reasons, and those reasons are the determinant (cause) of her choice. When you look into what decision making actually is, I.e. the procedures that underlying it, you find that it MUST be the case that our thinking is mostly if not all deterministic for free will to be of any meaning at all. So really, the question is “why wouldn’t they be compatible?” And it says a lot about our introspective blind spots that we have for so long thought that free will and determinism were at odds with each other.
Veedrac 8y ago
To put it shortly but slightly inaccurately, compatibilism basically comes from the idea that "free will" is implied by "moral responsibility", and then it is noted that moral responsibility and determinism are compatible. For more detail, https://plato.stanford.edu/entries/compatibilism/#FreWil https://plato.stanford.edu/entries/compatibilism/#FreWil.
- naasking 8y ago
  > To put it shortly but slightly inaccurately, compatibilism basically comes from the idea that "free will" is implied by "moral responsibility", and then it is noted that moral responsibility and determinism are compatible. It's actually the other way around: if you acted according to your own free will, then you are morally responsible for your choice.
  - Veedrac 8y ago
    No, read the link I gave. (Note also that I'm explaining the dominant arguments, not endorsing them.) If it's still not clear why I put the implication that way around, note that not everyone thinks that you have a moral responsibility in the way Peter Singer argues, but basically every philosopher thinks it's wrong to kick a dog.
    - naasking 8y ago
      I've been studying and debating this issue for years, I know what it means. From your own link, "As a theory-neutral point of departure, then, free will can be defined as the unique ability of persons to exercise control over their conduct in the manner necessary for moral responsibility". In other words, making a choice of your own free will entails moral responsibility, not the other way around like you said. I don't know what Singer has to do with any of this.
      - Veedrac 8y ago
        > "As a theory-neutral point of departure, then, free will can be defined as the unique ability of persons to exercise control over their conduct in the manner necessary for moral responsibility" Exactly... "necessary", not "sufficient". https://en.wikipedia.org/wiki/Necessity_and_sufficiency https://en.wikipedia.org/wiki/Necessity_and_sufficiency https://plato.stanford.edu/entries/necessary-sufficient/ https://plato.stanford.edu/entries/necessary-sufficient/
majewsky 8y ago
The English Wikipedia article was not very enlightening, so I went over to the German Wikipedia which includes this helpful example: > According to [David] Hume, free will is not the ability to choose otherwise given the exact same inner and outer condition. He considers it the hypothetical ability to choose otherwise, had the actor been in a psychologically different state because of different wishes or beliefs. (Sorry for the rough translation, but I think it brings the point across.) In general, the debate seems to be on what you take "free will" to mean. Philosopher/neuroscientist Joscha Bach [1] defines it as "the ability to do what one has recognized as right", contrasting it e.g. with compulsory behavior. [1] https://bach.ai https://bach.ai, although the source of the quote is the German podcast https://alternativlos.org/42/ https://alternativlos.org/42/
- rwnspace 8y ago
  Regarding your quote of Joscha Bach, I wonder what we make of Executive Function disorders (ADHD) and their relationship to free will. Particularly, I wonder if philosophers with executive function disorders are more likely to prefer determinism...
intralizee 8y ago
Compatibilists are inconsistent determinists. They claim we're "free" when we're doing what we want; not being forced by external circumstances to do something we don't want. Basically the idea of being free if you're a puppet and love your strings. I find the idea ignores how everything is one entity and what encompasses the puppet, "being the stage & props" must also be loved by the puppet for one to be free. So the perfect life fairytale is needed to be free?
- naasking 8y ago
  Are you suggesting that you're constantly doing things you don't want all the time? And that you have no power to change this?
  - eli_gottlieb 8y ago
    I think he's suggesting tossing out personal feelings or values ("the strings") as having no actual normative worth, which leaves me wondering what he thinks would be worth having free will over.
Ephem 8y ago
As Wikipedia phrases Harry G. Frankfurt (1969): "A compatibilist can believe that a person can choose between many choices, but the choice is always determined by external factors" A bit sloppily put, a compatibilist believes a choice is free if the agent making it is not coerced or constrainted, even if the choice that the agent will make is determined. Different compatibilists approach this as either being descriptive (this is how we tend to actually make moral judgements), or normative (free will is defined in these terms and moral responsibility for our actions follows from that). Most incompatibilists agree that "freedom to act" is a necessary criterion for free will, but not that it is sufficient.
- Ephem 8y ago
  Personally I would argue that we descriptively have different definitions of free before a choice and after. Because the future has not yet happened, all possible future choices are indeed free in a sense, even if one of them is predetermined. When we look back, I think we tend to think of free in a different way, that is, given that a choice has already been made, we can now say that it was the only choice that could have happened and did (determinism or not, I think we often tend to think this way in ordinary life). When we judge moral responsibility, we tend to think of the actor in that moment of making the choice. At that moment the choice was still "free", because the future had not yet happened, and so we ascribe moral responsibility to these actions.
  - garmaine 8y ago
    What does it mean for a choice to be “free” if it is predetermined?
- MrScruff 8y ago
  So is it the case that the compatibilist imagines three possible classes of decision. 1) Where the decision is coerced. 2) Where the decision is not coerced and so ostensibly 'free' but determined by external factors (ie. the laws of Physics). 3) Where the decision is genuinely 'free', thus violating the laws of Physics as we know them. Compatibilists would reject 3) but would describe 2) as the exercising of 'free will'. Is that correct?
mac01021 8y ago
"You can do what you want, but you can't want what you want." - Lawrence of Arabia
- leoreeves 8y ago
  That's very similar to Schopenhauer's view: "Man can do what he wills but he cannot will what he wills."
brownbat 8y ago
The Stanford Encyclopedia of Philosophy is a great resource for common sense and exhaustive coverage of these sorts of things: https://plato.stanford.edu/entries/compatibilism/ https://plato.stanford.edu/entries/compatibilism/ Classical free will, going back to Hobbes, is really just that sometimes we act unencumbered by external forces. Determinism posits that no one can act other than how one ultimately does. Those could be compatible. There is a difference between me deciding to stay in a room after careful deliberation, and me being chained to a wall in that room in spite of my protest. Even if the future is fixed, there's still a difference between a person being forced to do something by an external actor and a person doing something they would ordinarily do.
- fjsolwmv 8y ago
  And that Hobbesian definition was ignorant of modern neuroscience.
  - brownbat 8y ago
    Modern neuroscience wouldn't tell you very much that's useful about external forces. All Hobbesian free will is committed to is that being chained to a wall changes your circumstances. Being kidnapped deprives you of meaningful options. There is a separate version of free will that insists that agents are the ultimate cause of their actions, rather than mediated causes, which would not be consistent with determinism. But that is not the only definition of free will (and I don't think a very common one, except by incompatabilists). A lot of this hinges on what people actually mean by the individual phrases. If you sometimes engage in deliberation, and sometimes deliberation changes your mind, then even if that deliberation would have always inevitably come out the same way, you still know that in those cases deliberation affected your subsequent actions. Your deliberation (informed though it may be by external factors) is sufficient for most sensible versions of agency. (EDITED to be more comprehensive.)
fjsolwmv 8y ago
Compatibilism rests upon the concept of philosophical zombies. It says that since you can't know someone else's internal state, you can assume they have free will, because assuming they don't gave free will doesn't change anything you can observe. It basically says that free will is an unanswerable question, so it swaps in a different question that is answerable. Its akin to p-values in statistics (probability of evidence given hypothesis) being used in place of a what we really want to know (probability of a gypothesis given evidence).
adrianN 8y ago
I believe this SMBC provides an easy to understand explanation of the concept: https://www.smbc-comics.com/comic/compatibilism https://www.smbc-comics.com/comic/compatibilism
alan-crowe 8y ago
The intuition of conflict comes from the two person case. The slave is in chains, the master wields his bloody whip. The master determines that the slave will pick cotton. The slave is not free to refuse. The single person case has two descriptions of the person. The chatty, social person, who tells us of his thoughts and feelings and why he chose tea instead of coffee. Contrast that with the account in terms of atoms and molecules. Neurons fire, muscles contract, etc. Where have we seen this before? In the kinetic theory of gasses. The pressure of the hot gas in a heat engine pushes the piston down. Alternatively the molecules bounce off the top of the piston, hammering it down with countless tiny hammer blows. Which account is true? Why not both? In the single person case the differences between the two accounts are so large that we are sucked into seeing them as rival accounts. But philosophers have never managed to clarify why the accounts are incompatible. Both accounts run out of steam. We don't really know the secrets of our own hearts and often resort to rationalizing the choices that we freely make, after the event. The atoms and molecules account asserts that those choices are in principal determined, and offers a wonderful excuse, in the exponential sensitivity to initial conditions, for why determinists cannot be expected to make predictions. Sure, we can say that our free choices are determined by atoms and molecules following the laws of physics. But take the prediction problem more seriously. Imagine that I have a gadget that predicts whether you will chose tea or coffee. If I tell you in advance, you can chose the other way, defeating the prediction. The philosophical problem arises due to viewing the atoms and molecules that make up your body as the gadget that predicts your choices. You cannot escape their power over your actions. On the other hand the point seems fatuous. Those atoms and molecules are you. "You cannot do other than what you do" is just a triviality because the two accounts are two accounts of a single underlying reality. They are bound to be compatible, no matter how much the customary language jars and creates the appearance of conflict.

skissane 8y ago

There is no single universally agreed definition of the term "free will". So, it is possible to define "free will" in such a way that it is compatible with determinism (e.g., "my will is free if its decisions are the product of my ordinary processes of thinking–even if those ordinary processes are determined by natural laws–as opposed to the product of any extraordinary external interference such as drugs, brainwashing, mind control implants, etc"). Part of the dispute is over whether such a definition of "free will" matches people's everyday pre-philosophical usage, and also whether it works for the purposes of other disciplines in which the concept of "free will" might be invoked, such as ethics or law. Compatibilists would say it is close enough, and can be made to work for those purposes. Incompatibilists say it is a radical redefinition, and undermines those purposes.

MrScruff 8y ago
So it seems the crux of the matter is how we regard the process of a person making a decision. A compatibilist would say that for the purposes of discussion we describe this event as the exercising of will, even though from a physics perspective the outcome could be predetermined. Is that correct? If so that seems similar to a recent discussion on Bohr's view of the irreducibility of biology to physics. https://news.ycombinator.com/item?id=18104508 https://news.ycombinator.com/item?id=18104508
- skissane 8y ago
  Certainly that's one way a compatibilist could put it – that "free will" is part of psychology, which ultimately reduces to/emerges from/supervenes upon lower layers of biology/chemistry/physics which may be fundamentally deterministic. (Although exactly how deterministic they are may depend on which interpretation of quantum mechanics is correct.) A point worth adding – there are different forms of determinism. Physical determinism says people's decisions are predetermined by physical laws and initial conditions. Theological determinism says people's decisions are predetermined by God's will. A compatibilist could be either. (Although most compatibilists you'll encounter these days are in the physical not theological camp.)
naasking 8y ago
I think you pretty fairly describe the current landscape of this debate, which is unusual in itself. Kudos! > Part of the dispute is over whether such a definition of "free will" matches people's everyday pre-philosophical usage, and also whether it works for the purposes of other disciplines in which the concept of "free will" might be invoked, such as ethics or law. Fortunately, this is an empirical question that has been explored by experimental philosophy [1]. It turns out that most lay people employ Compatibilist reasoning, so I think this question has also been answered. [1] The latest that I've seen on the topic: https://www.researchgate.net/publication/274892120_Why_Compatibilist_Intuitions_are_not_Mistaken_A_Reply_to_Feltz_and_Millan https://www.researchgate.net/publication/274892120_Why_Compa...
- fjsolwmv 8y ago
  Lay people, yes, but lay people also believe in ghosts and angels. Lay people believing it doesn't make it true. Scientific reasoning shows compatibilism is inaccurate, But just how like how belief in an scientifically absurd afterife can make people happy, being scientifically correct is not equivalent to having a good life. And this is a potential definition of the essence religion or wait, properly opposed to science: faith can make you happy even when you are wrong, and losing that faith makes you unhappy. Therefore, having faith is the correct choice for an ideal human life. It's right to be wrong.
  - naasking 8y ago
    > Lay people, yes, but lay people also believe in ghosts and angels. Lay people believing it doesn't make it true. I suggest you read the study and others like it. If you flat out ask people what they believe, with no further context, they tend to support incompatbilist intuitions. But when they test their responses to moral questions, they employ Compatibilist moral reasoning, and agree with claims that are Compatibilist.
clairity 8y ago
i'm leaning toward a belief in an outcome where we eventually prove strong determinism false, simply because of the probabalistic underpinnings of the universe. in that scenario, compatibalism would be an acceptance of a weak determinism based on the likeliness of probabalistic outcomes (particularly at larger scales), rather than an absolute certainty.

garmaine 8y ago

Communication over the internet is hard and I’m not sure if I can convey the idea briefly in a HN post, but the idea is simple. The assumption that free will and determinism are at odds with each other is ungrounded when you actually look into it. “Non-determinism” implies randomness which is NOT what we are talking about when we say something like “Susie made a decision.” She makes a decision for reasons, and those reasons are the determinant (cause) of her choice. When you look into what decision making actually is, I.e. the procedures that underlying it, you find that it MUST be the case that our thinking is mostly if not all deterministic for free will to be of any meaning at all. So really, the question is “why wouldn’t they be compatible?” And it says a lot about our introspective blind spots that we have for so long thought that free will and determinism were at odds with each other.

Veedrac 8y ago

To put it shortly but slightly inaccurately, compatibilism basically comes from the idea that "free will" is implied by "moral responsibility", and then it is noted that moral responsibility and determinism are compatible. For more detail, https://plato.stanford.edu/entries/compatibilism/#FreWil https://plato.stanford.edu/entries/compatibilism/#FreWil.

naasking 8y ago
> To put it shortly but slightly inaccurately, compatibilism basically comes from the idea that "free will" is implied by "moral responsibility", and then it is noted that moral responsibility and determinism are compatible. It's actually the other way around: if you acted according to your own free will, then you are morally responsible for your choice.
- Veedrac 8y ago
  No, read the link I gave. (Note also that I'm explaining the dominant arguments, not endorsing them.) If it's still not clear why I put the implication that way around, note that not everyone thinks that you have a moral responsibility in the way Peter Singer argues, but basically every philosopher thinks it's wrong to kick a dog.
  - naasking 8y ago
    I've been studying and debating this issue for years, I know what it means. From your own link, "As a theory-neutral point of departure, then, free will can be defined as the unique ability of persons to exercise control over their conduct in the manner necessary for moral responsibility". In other words, making a choice of your own free will entails moral responsibility, not the other way around like you said. I don't know what Singer has to do with any of this.
    - Veedrac 8y ago
      > "As a theory-neutral point of departure, then, free will can be defined as the unique ability of persons to exercise control over their conduct in the manner necessary for moral responsibility" Exactly... "necessary", not "sufficient". https://en.wikipedia.org/wiki/Necessity_and_sufficiency https://en.wikipedia.org/wiki/Necessity_and_sufficiency https://plato.stanford.edu/entries/necessary-sufficient/ https://plato.stanford.edu/entries/necessary-sufficient/

majewsky 8y ago

The English Wikipedia article was not very enlightening, so I went over to the German Wikipedia which includes this helpful example: > According to [David] Hume, free will is not the ability to choose otherwise given the exact same inner and outer condition. He considers it the hypothetical ability to choose otherwise, had the actor been in a psychologically different state because of different wishes or beliefs. (Sorry for the rough translation, but I think it brings the point across.) In general, the debate seems to be on what you take "free will" to mean. Philosopher/neuroscientist Joscha Bach [1] defines it as "the ability to do what one has recognized as right", contrasting it e.g. with compulsory behavior. [1] https://bach.ai https://bach.ai, although the source of the quote is the German podcast https://alternativlos.org/42/ https://alternativlos.org/42/

rwnspace 8y ago
Regarding your quote of Joscha Bach, I wonder what we make of Executive Function disorders (ADHD) and their relationship to free will. Particularly, I wonder if philosophers with executive function disorders are more likely to prefer determinism...

intralizee 8y ago

Compatibilists are inconsistent determinists. They claim we're "free" when we're doing what we want; not being forced by external circumstances to do something we don't want. Basically the idea of being free if you're a puppet and love your strings. I find the idea ignores how everything is one entity and what encompasses the puppet, "being the stage & props" must also be loved by the puppet for one to be free. So the perfect life fairytale is needed to be free?

naasking 8y ago
Are you suggesting that you're constantly doing things you don't want all the time? And that you have no power to change this?
- eli_gottlieb 8y ago
  I think he's suggesting tossing out personal feelings or values ("the strings") as having no actual normative worth, which leaves me wondering what he thinks would be worth having free will over.

Ephem 8y ago

As Wikipedia phrases Harry G. Frankfurt (1969): "A compatibilist can believe that a person can choose between many choices, but the choice is always determined by external factors" A bit sloppily put, a compatibilist believes a choice is free if the agent making it is not coerced or constrainted, even if the choice that the agent will make is determined. Different compatibilists approach this as either being descriptive (this is how we tend to actually make moral judgements), or normative (free will is defined in these terms and moral responsibility for our actions follows from that). Most incompatibilists agree that "freedom to act" is a necessary criterion for free will, but not that it is sufficient.

Ephem 8y ago
Personally I would argue that we descriptively have different definitions of free before a choice and after. Because the future has not yet happened, all possible future choices are indeed free in a sense, even if one of them is predetermined. When we look back, I think we tend to think of free in a different way, that is, given that a choice has already been made, we can now say that it was the only choice that could have happened and did (determinism or not, I think we often tend to think this way in ordinary life). When we judge moral responsibility, we tend to think of the actor in that moment of making the choice. At that moment the choice was still "free", because the future had not yet happened, and so we ascribe moral responsibility to these actions.
- garmaine 8y ago
  What does it mean for a choice to be “free” if it is predetermined?
MrScruff 8y ago
So is it the case that the compatibilist imagines three possible classes of decision. 1) Where the decision is coerced. 2) Where the decision is not coerced and so ostensibly 'free' but determined by external factors (ie. the laws of Physics). 3) Where the decision is genuinely 'free', thus violating the laws of Physics as we know them. Compatibilists would reject 3) but would describe 2) as the exercising of 'free will'. Is that correct?

mac01021 8y ago

"You can do what you want, but you can't want what you want." - Lawrence of Arabia

leoreeves 8y ago
That's very similar to Schopenhauer's view: "Man can do what he wills but he cannot will what he wills."

brownbat 8y ago

The Stanford Encyclopedia of Philosophy is a great resource for common sense and exhaustive coverage of these sorts of things: https://plato.stanford.edu/entries/compatibilism/ https://plato.stanford.edu/entries/compatibilism/ Classical free will, going back to Hobbes, is really just that sometimes we act unencumbered by external forces. Determinism posits that no one can act other than how one ultimately does. Those could be compatible. There is a difference between me deciding to stay in a room after careful deliberation, and me being chained to a wall in that room in spite of my protest. Even if the future is fixed, there's still a difference between a person being forced to do something by an external actor and a person doing something they would ordinarily do.

fjsolwmv 8y ago
And that Hobbesian definition was ignorant of modern neuroscience.
- brownbat 8y ago
  Modern neuroscience wouldn't tell you very much that's useful about external forces. All Hobbesian free will is committed to is that being chained to a wall changes your circumstances. Being kidnapped deprives you of meaningful options. There is a separate version of free will that insists that agents are the ultimate cause of their actions, rather than mediated causes, which would not be consistent with determinism. But that is not the only definition of free will (and I don't think a very common one, except by incompatabilists). A lot of this hinges on what people actually mean by the individual phrases. If you sometimes engage in deliberation, and sometimes deliberation changes your mind, then even if that deliberation would have always inevitably come out the same way, you still know that in those cases deliberation affected your subsequent actions. Your deliberation (informed though it may be by external factors) is sufficient for most sensible versions of agency. (EDITED to be more comprehensive.)

fjsolwmv 8y ago

Compatibilism rests upon the concept of philosophical zombies. It says that since you can't know someone else's internal state, you can assume they have free will, because assuming they don't gave free will doesn't change anything you can observe. It basically says that free will is an unanswerable question, so it swaps in a different question that is answerable. Its akin to p-values in statistics (probability of evidence given hypothesis) being used in place of a what we really want to know (probability of a gypothesis given evidence).

adrianN 8y ago

I believe this SMBC provides an easy to understand explanation of the concept: https://www.smbc-comics.com/comic/compatibilism https://www.smbc-comics.com/comic/compatibilism

alan-crowe 8y ago

The intuition of conflict comes from the two person case. The slave is in chains, the master wields his bloody whip. The master determines that the slave will pick cotton. The slave is not free to refuse. The single person case has two descriptions of the person. The chatty, social person, who tells us of his thoughts and feelings and why he chose tea instead of coffee. Contrast that with the account in terms of atoms and molecules. Neurons fire, muscles contract, etc. Where have we seen this before? In the kinetic theory of gasses. The pressure of the hot gas in a heat engine pushes the piston down. Alternatively the molecules bounce off the top of the piston, hammering it down with countless tiny hammer blows. Which account is true? Why not both? In the single person case the differences between the two accounts are so large that we are sucked into seeing them as rival accounts. But philosophers have never managed to clarify why the accounts are incompatible. Both accounts run out of steam. We don't really know the secrets of our own hearts and often resort to rationalizing the choices that we freely make, after the event. The atoms and molecules account asserts that those choices are in principal determined, and offers a wonderful excuse, in the exponential sensitivity to initial conditions, for why determinists cannot be expected to make predictions. Sure, we can say that our free choices are determined by atoms and molecules following the laws of physics. But take the prediction problem more seriously. Imagine that I have a gadget that predicts whether you will chose tea or coffee. If I tell you in advance, you can chose the other way, defeating the prediction. The philosophical problem arises due to viewing the atoms and molecules that make up your body as the gadget that predicts your choices. You cannot escape their power over your actions. On the other hand the point seems fatuous. Those atoms and molecules are you. "You cannot do other than what you do" is just a triviality because the two accounts are two accounts of a single underlying reality. They are bound to be compatible, no matter how much the customary language jars and creates the appearance of conflict.