Monday, August 8, 2011

A reliabilist moral argument for the truth of some religious belief

Let P be the process of genetically or mimetically producing belief in non-empirical claims in order to enhance social cooperation.  Let moral realism be the claim that we know some non-trivial moral truths.  Consider this argument:
  1. (Premise) Evolutionary process P is the relevant process that produced both our religious and our moral beliefs.
  2. (Premise) If all religious beliefs are false, P is unreliable (since roughly half of the basic types of beliefs produced by P are then false).
  3. So, if all religious beliefs are false, then the relevant process that produced our moral beliefs is unreliable. (By 1 and 2)
  4. (Premise) Beliefs produced by an unreliable relevant process are not knowledge. (This is a consequence of reliabilism.)
  5. So, if all religious beliefs are false, we lack moral knowledge. (By 3 and 4 plus the analytic truth that knowledge requires belief.)
  6. (Premise) If moral realism is true, we have moral knowledge. 
  7. So, if all religious beliefs are false, moral realism is false. (By 5 and 6)
  8. So, if moral realism is true, some religious beliefs are true. (By 7)
And moral realism is true.

Now, I am not sure whether 1 is true.  But I think the main alternative to 1 is that God produced some of our religious or some of our moral beliefs or both.  And if that alternative is true, then some religious beliefs are true, namely those that say that God exists.  So my uncertainty about 1 does not harm the argument.  I am also not sure about the reliabilist premise 4, and that's more serious.

I think the naturalist reliabilist who wants to deny 8 will accept that P produced our religious and moral beliefs, but say that P is not the epistemologically relevant process.  The relevant process is, perhaps, the sub-type of P which is the genetic or mimetic production of beliefs in moral claims in order to enhance social cooperation.  I think this identification of the relevant sub-type is objectionably ad hoc.


Mike A Robinson said...

Professor I really enjoyed this post. I’m really trying, without success, to find ways around your argument. What other groups, along with the nihilist who denies moral realism, might assert that this argument lacks persuasive power? Are many naturalists in academia moral anti-realists?

Daniel said...

Dr. Pruss, Some naturalists might say that while religious beliefs are non-empirically derived from P, moral beliefs should be only be based on a process that is empirical. So, if something like Sam Harris' theory of morality were to prove correct, an alternate process to P could be proposed without being objectionably ad hoc.

At the same time, the arguments that objective morality is empirical have so far proved to be rather flimsy (IMHO).

Alexander R Pruss said...

The one hope I see for an empirically observable morality is if we have a moral sense, whereby we directly perceive moral truths. But such a sense would not be any of our physical senses, since our physical senses are sensitive to purely physical states of affairs, and moral states of affairs aren't purely physical. Thus, such a view could only be held by a non-naturalist.

Alexander R Pruss said...


Well, the anti-reliabilist in epistemology will deny 4.

I am a bit sceptical of 4 myself, but I am inclined to accept some version of the principle that if we know our beliefs come from an unreliable process, then the beliefs aren't knowledge. This won't support the full argument, but some version may yet survive.

Alexander R Pruss said...


Another move is Gene Witmer's here.

Mike A Robinson said...

Professor Pruss: Thanks so much for the added information and I really enjoyed the article: “God, evolutionary psychology and moral realism” as well as the related post:
“EAAN in the case of moral knowledge;” especially what you stated: “I think one can use EAAN-type arguments for a more limited conclusion, namely that if naturalism and evolution are true, then certain important kinds of knowledge are seriously threatened, specifically moral (and maybe more generally normative) knowledge (I think certain kinds of modal and metaphysical knowledge are also threatened, and it may be that metaphysical naturalism falls within the class of threatened knowledge).”
I want to re-read and study that article and the related posts (Gene’s criticism and your responses).

A bit off topic, Dr. Pruss would it be plausibly defensible for me to maintain:
“If it is possible that there is a moral absolute in one possible world, since it is an absolute, the moral absolute would be true in all possible worlds or necessary.”
If you have time to respond, I don’t presume a long answer. I am already so grateful for your posts and answers you have already provided.

Clifton said...

It seems that whether or not the naturalist is committed to (1) depends on how broadly or narrowly we construe evolutionary processes. On a narrow construal (one on which an evolutionary process is a specific adaptation in response to specific selective pressures) I don't see what's stopping the naturalist from saying that our moral beliefs are the result of one evolutionary process and our religious beliefs are the result of another.

Surely there's some story to told on which our moral beliefs are the result of pressures that selected for *true* moral beliefs, while our religious beliefs are the result of pressures that merely selected for religious belief, true or not. Whether there would be any reason to believe such a story is another matter.

But if the argument is trading on a broader construal of an evolutionary process (the entire evolutionary history of an organism, perhaps) why is that the construal that should be of interest to a reliabilist?

Alexander R Pruss said...


I am thinking here of something in between the very general and the very specific as the relevant process, something like "producing a non-empirical belief as a way of discouraging defection in prisoner's dilemmas."

But I feel the force of your worry.