In Chapter 11 of The Coherence of Theism (revised edition, hereafter: COT), Richard Swinburne argues (among other things) that the following sentence makes a coherent statement:
(PG) There is a person who is perfectly good.
Perfect goodness is one of the divine attributes that theists use to define the word ‘God’. In order to show that the statement ‘God exists’ makes a coherent statement, Swinburne attempts to show that each of the divine attributes that are used to define the word ‘God’, and to clarify the meaning of the statement ‘God exists’, can be used to make coherent statements.
Swinburne takes himself to have shown (in previous chapters of COT) that the following sentence makes a coherent statement:
(BOP) There is a person who is both omniscient and perfectly free.
Swinburne’s method of showing a sentence to express a coherent statement is to show that some other statement which is known to be coherent entails the statement in question. So, since Swinburne has previously argued that (BOP) makes a coherent statement, all he needs to do in Chapter 11 to show that (PG) makes a coherent statement, is to show that (BOP) entails (PG).
The following is an email exchange that I had recently with Richard Swinburne about a key argument in Chapter 11 of COT on this issue.
=============
UPDATE 12/13/12: I received a brief response from Richard Swinburne to my email dated 12/8/12. See this response at the end of this post.
=============
=======================
On 23/11/2012 21:19, Bradley Bowen wrote:
Mr. Swinburne,
I have recently been studying Chapter 11 of your book The Coherence of Theism (revised edition), and have a question about your argument for the claim that “it is logically necessary that an omniscient and perfectly free being be perfectly good.” (COT, p.188)
I have put your understanding of the divine attribute of perfect goodness into a formal definition:
A person P is perfectly good if and only if (a) P is so constituted that P always does the morally best action (when there is one), and (b) P is so constituted that P does no morally bad action. (see COT, p.184)
I have also attempted to state your argument (see COT, p.208) as follows:
(PFO1) An omniscient person O will know of any action whether or not that action is morally good or morally bad, and will know this because of O’s nature (i.e. O has the attribute of omniscience).
(PFO2) A perfectly free person F will do those actions that F believes are morally good and avoid those actions that F believes are morally bad, and will act this way because of F’s nature (i.e. F has the attribute of perfect freedom).
(PFO3) It is logically possible for a person to be both omniscient and perfectly free (at the same time).
Therefore:
(PFO4) Any person P who is both omniscient and perfectly free (at the same time) will do those actions that are morally good and avoid those actions that are morally bad, and will act this way because of P’s nature (i.e. P has the attributes of omniscience and perfect freedom).
There appears to be a disconnection between the conclusion of this argument, which is in terms of doing actions that are “morally good” and the definition of the divine attribute of perfect goodness, which is in terms of doing the “morally best action (when there is one)”.
How do you bridge the apparent logical gap between the idea of doing “morally good” actions and doing “the morally best” action?
Bradley Bowen
Kirkland, Washington
USA
=============================
Date: Tue, 27 Nov 2012 10:48:21 +0000
From: richard.swinburne
To: bbowen
Subject: Re: The Perfect Goodness of God
Dear Bradley,
Thank you for your email. What I wrote, and what you quote from me, is that a perfectly good being ‘always does the morally best action’ (when there is one). I emphasize the ‘when there is one’. Sometimes among the actions open to such a being will be two equal best actions, that is actions such that they are both better than any other action open to that being and each is just as good as the other. That is a situation with which we are often ourselves confronted. Some beings (including God, but not including humans) may also have a choice among an infinite number of actions, each of which is less good than some other action which he could do. For example, God has a choice of how many solar systems to create; and on the assumption that solar systems are good things, then the more of them the better. So however many he creates, he could do better by creating more. In that case, since it is not logically possible that such a being could do either the best action or an equal best action, his perfect goodness would consist simply in doing an action of the relevant kind (e.g. creating solar systems), but it wouldn’t be the best or equal best action, but it would be a good action. What you have labelled (PFO4) does follow from the premises which you state; but I could have drawn the stronger and more complicated conclusion that P will always do an action which is morally the best if there is one, or an equal best action (when there is no best) if there is one, or (if there is no best or equal best action) a good action. With best wishes
Richard Swinburne
=======================================
On 27/11/2012 19:39, Bradley Bowen wrote:
Mr. Swinburne,
Thank you for your response to my question about the argument for your claim that “it is logically necessary that an omniscient and perfectly free being be perfectly good.” (COT, p.188)
Your key point:
“What you have labelled (PFO4) does follow from the premises which you state; but I could have drawn the stronger and more complicated conclusion that P will always do an action which is morally the best if there is one, or an equal best action (when there is no best) if there is one, or (if there is no best or equal best action) a good action.”
To draw this stronger conclusion, it seems to me that stronger premises would be required, but I believe that you accept the stronger premises.
Here is one way to strengthen the premise about the relationship of omniscience to perfect goodness:
(PFO1′) An omniscient person O will know of any action whether or not that action is morally good or morally bad,
and if an action is morally good will know whether it is the morally best action among available alternative morally good actions,
and will know this becauseof O’s nature (i.e. Ohas the attribute of omniscience).
A similar adjustment could be made to the premise about the relationship between perfect freedom and perfect goodness:
(PFO2′) A perfectly free person F will do those actions that F believes are morally good and avoid those actions that F believes are morally bad, and F will do the action that F believes is the morally best action among available alternative morally good actions (when there is such an action), and will act this way because of F’s nature (i.e. F has the attribute of perfect freedom).
Do you agree that the premises need to be strengthened in order to draw the stronger conclusion?
Bradley Bowen
Kirkland, Washington
USA
=======================================
Date: Sat, 1 Dec 2012 12:34:29 +0000
From: richard.swinburne
To: bbowen
Subject: Re: The Perfect Goodness of God
Dear Bradley,
Unless I have missed something, your (PFO1′) is the same as your (PFO1), and your (PFO2′) is the same as your (PFO2), and so they don’t constitute any strengthening of the argument. You are however right that in order to get the ‘stronger and more complicated conclusion’ which I described in my previous email, I do need a stronger premise. The requisite premise is: a perfectly good being will – in so far as it can – do what it believes is the best action, or (if it believes that there is no best action but an equal best) do what it believes to be the equal best action, or (if it believes that there is no best, and no equal best) will do what it believes to be a good action. Given that premise and the premise about omniscience in the form that an omniscient being will have true beliefs about which actions are best or equal best or good, and a premise that an omnipotent being will be able to do such actions, the conclusion follows. With best wishes
Richard Swinburne
================================
From:Bradley Bowen
Sent:Sat 12/08/12 7:46 PM
To:Richard Swinburne
Mr. Swinburne,
Thank you for your suggested revision of my reformulation of an argument from p.208 of The Coherence of Theism (revised ed.):
You are however right that in order to get the ‘stronger and more com
plicated conclusion’ which I described in my previous email, I do need a stronger premise. The requisite premise is: a perfectly good being will – in so far as it can – do what it believes is the best action, or (if it believes that there is no best action but an equal best) do what it believes to be the equal best action, or (if it believes that there is no best, and no equal best) will do what it believes to be a good action.
Here is how I would reformulate the argument, based on your suggested enhancement:
(PFO1*) An omniscient person O will know of any action whether or not that action is a morally best action for O or an equal best action for O, or just a good action (that is neither a best nor an equal best action) for O, or a morally bad action for O, and will know this because of O’s nature (i.e. because O has the attribute of omniscience).
(PFO2*) A perfectly free person F will do an action – in so far as it can – when F believes the action is a morally best action for F or (if it believes that there is no best action but an equal best) do what it believes to be an equal best action for F, or (if it believes that there is no best, and no equal best) will do what it believes to be a good action for F, and will not do any action it believes to be a morally bad action for F, and will do these actions (and refrain from morally bad actions) because of F’s nature (i.e. becuase F has the attribute of perfect freedom).
(PFO3*) It is logically possible for a person to be both omniscient and perfectly free (at the same time).
Therefore:
(PFO4*) Any person P who is both omniscient and perfectly free (at the same time) will do an action that is a morally best action for P or (if there is no best action but an equal best) do an equal best action for P, or (if there is no best, and no equal best) will do a good action for P, and will not do any action that is a morally bad action for P, and will do these actions (and refrain from morally bad actions) because of P’s nature (i.e. becuase P has both the attribute of omniscience and the attribute of perfect freedom).
I believe the qualification “in so far as it can” in (PFO2*) falls out because an omniscient being who judges and action to be a morally best action for itself to perform is also judging that action to be one that is possible for itself to perform, since ought implies can, and similar reasoning applies to other moral categories (equal best, morally good, morally bad).
Do you think this is a reasonable and enhanced reformulation of the argument on p. 208 of The Coherence of Theism?
Bradley Bowen
Kirkland, Washington
USA
==================================
Date: Mon, 10 Dec 2012 10:28:22 +0000
From: richard.swinburne
To: bbowen
Subject: Re: The Perfect Goodness of God
*That is a very good fomulation of my argument! Best Wishes – Richard Swinburne
==============================
