Thursday, March 5, 2020

Upbringing and responsibility

Consider these two stories:

  1. Alice grew up in a terrible home. Her mother abused her. Her father taught her by word and deed that morality is the advantage of the stronger. Her parents forced her to join them in their manifold criminal enterprises. Alice lacked good role models. When Alice turned 17, she grew wings and flew away.

  2. Bob grew up in a terrible home. His father abused him. His mother taught him by word and deed that morality is the advantage of the stronger. His parents forced him to join them in their manifold criminal enterprises. Bob lacked good role models. When Bob turned 17, he rebelled and turned away from the life of crime.

The first story is unlikely and unbelievable. The second is unlikely but believable. This suggests that we don’t literally think that a really bad upbringing makes it literally impossible to do the right thing.

As an argument against determinism, this is rather weak, though. For even the determinist will say that there are many factors left out of story (2), and it could be that one of those left out factors that caused Bob to rebel.

4 comments:

Grigory Aleksin said...

On the subject of moral responsibility. I am long been thinking about the following argument. Although I am not sure it is really that strong:
(1) If Naturalism is true, Determinism must be true (premise)
(2) Free Will and Determinism are incompatible (premise)
(3) If Moral Responsibility exists, then Free Will must exist (premise)
(4) Moral Responsibility Exists (premise)
(5) Free Will exists (by (3) and (4))
(6) Determinism is false (by (2) and (5))
(7) Naturalism is false (by (1) and (6))
The argument is pretty deep since it raises all sort of questions, but I think that premise (1) is least secure.

Alexander R Pruss said...

I think premise (1) is false. But I think the following premise is true:

1a. If naturalism is true, then indeterminism is no friendlier to free will than determinism.

(Or, more rigorously: If naturalism is true and if determinism is incompatible with free will, then there is no free will even if determinism is not true.)

The intuition is that naturalistic indeterministic processes just add noise, and that doesn't give freedom.

Grigory Aleksin said...

Thank you for your suggestion! I am not sure that my argument is actually very helpful. Is there a good way to argue that free will is incompatible with naturalistic indeterminism? I can see it intuitively, but... would I argue that free will requires some form of agent causation to be compatible with indeterminism? and since agent causation fails to integrate with a naturalistic worldview, then it is just noise?

Then again, this argument could be turned into one for theism/supernaturalism:
I will use Bayes’ theorem to develop my idea. This argument will undoubtedly be more contentious because I use an equiprobability principle. However, I believe that this is justified and will argue as such below.

Key Terms:
SN: Supernaturalism
N: Naturalism
MR: Moral Responsibility
D: Determinism

Bayesian Formulation:
P(SN|MR)=P(MR|SN)(PSN)/P(MR)=P(MR|SN)(PSN)/[P(MR|SN)P(SN)+P(MR|N)P(N)]

P(MR|SN)is uncontroversially high. It seems to me that a supernaturalist solution offers greater scope for indeterminism and has the resources necessary to support agent causation. Furthermore, there is greater scope for avoiding the causal closure principle.

Equiprobability Principle: since the prior probability of supernaturalism and naturalism are not conditioned on our background information, there is little one can say about this probabilities. For this reason, I adopt the following principle:

If there are two mutually exclusive possibilities that are exhaustive and mutually exclusive, thus forming a partition of the sample space, and there is no information pertaining to their likelihood, it is assumed that their a priori probability is roughly equal. That is inductive, it can be stated that each probability is equal to 1/N, where N is the number of possibilities in the partition.

For this reason, I assume that the a priori probability of naturalism and supernaturalism is roughly equal, that is P(N)=P(SN)=1/2. Thus cancellation in the equation ensues. I factorise out the prior probabilities of naturalism and supernaturalism which then cancel out thus leaving the following equation:

P(SN|MR)=P(MR|SN)/[P(MR|SN)+P(MR|N)]

The problem here is then whether indeterminism is more compatible with naturalism or supernaturalism? Peter Van Inwagen would probably claim that here lies a mystery and that therefore the argument has no good evaluation. I am again not sure. I suppose it does indeed come down to whether indeterminism is more likely to be compatible with free will under naturalism or under supernaturalism. I favour the latter, but as before I do not know enough to show it.

Grigory Aleksin said...

I think that I made a few mistakes with my notation. Proof-reading is indeed a useful thing to do...