What God Cannot Do – Part 2

There are two types of actions that God cannot do (see The Coherence of Theism, p.164):

T1. Actions that are logically impossible for any being to perform.

T2. Actions that are logically possible for some beings to perform, but logically impossible for God to perform.

No being, including God, could produce or discover a four-sided triangle. No being, including God, could produce or find a married bachelor. No being could produce or identify an even number that is greater than 1 but less than 2. No being, according to Swinburne, can change the past. These are actions of Type 1. Let’s call these intrinsically impossible actions.

There are many actions that I can perform, but which God cannot perform. These are actions of Type 2. For example, I can snap my fingers, brush my teeth, scratch my head, clear my throat, chew gum, close my eyes, eat a cheeseburger, wash my face, put my fist through a window, kick a soccer ball, throw a stone, and pinch myself, but God is not able to do any of these things, because God is a spirit, a bodiless person. I can do some things that God cannot do, because I have a body, and God does not. Some actions can only be performed by persons who have a body.

There are other examples of actions of Type 2. God cannot forget someone’s name, because God is omniscient. God cannot try to move some object but fail, because God is omnipotent. God cannot do something that is evil, because God is both omniscient and perfectly good. God cannot cause himself to cease to exist, because God is eternal. These are all actions that are possible for some beings to perform, but not for God, because of some property (divine attribute) that God possesses and which some other beings do not possess. Let’s call these actions extrinsically impossible actions.

Actually, it would be logically possible for God, who is a bodiless person, to brush his teeth, scratch his head, etc. if the property of being bodiless was just a property that God happened to have but was not an essential or necessary property of God.

If God is in fact bodiless, but things could have been otherwise so that God had a body, then logic would not constrain God from performing actions that require the possession of a body. So long as God remains bodiless, he will be unable to perform actions that require having a body, but if it is logically possible for God to have had a body or to become embodied, then it is logically possible for God to brush his teeth, scratch his head, etc.

Many theists and most theologians, however, want to claim that divine attributes, such as omnipotence, omniscience, perfect goodness, etc., are essential properties of God; that is to say, it is logically impossible for God to exist but to lack some of the divine attributes.

In any possible world in which it is true that ‘God exists’, the being who is God possesses all of the divine attributes or properties (Swinburne’s analysis of ‘God’ includes nine such properties). If the divine attributes are essential properties of God, then it is not just the case that God is in fact unable to brush his teeth or to forget a name, but it is logically impossible for God to brush his teeth or to forget a name.

To be continued…

About Bradley Bowen
  • http://www.blogger.com/profile/02447457531584294267 Joe

    It seems like the Paradox of the Stone argument commits the modal fallacy (in Wade Savage's paper, The Paradox of the Stone in The Philosophical Review it is the second premise).

    (P) God can create a stone which God cannot lift.
    (C) Therefore, God is not omnipotent.

    You can't derive God's lack of omnipotence from the mere possibility that God can create a stone God can't lift.

    (Unless God is necessarily omnipotent…I'm not sure how Swinburne's question-begging charge applies then.)

    See for example,

    http://www.sfu.ca/~swartz/modal_fallacy.htm#omnipotence

  • http://www.blogger.com/profile/16947798364523082547 Rich Griese

    Would be good to establish that there are gods before you begin attempting to talk about their attributes. People often make this mistake. Since no gods have been demonstrated, talking about them is a waste of time.

    Cheers! RichGriese@gmail.com (use for responses)

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    Joe said…

    You can't derive God's lack of omnipotence from the mere possibility that God can create a stone God can't lift.
    ============
    Response:
    That is, in essence, the solution that Swinburne offers for the paradox of the stone. So long as God is merely able to create a stone he cannot move but chooses not to exercise this ability, there is no problem. It is only if God chooses to exercise this ability that there will be a stone that God cannot move.

    But in that case, God would be making the choice to BECOME less than omnipotent. Thus, this has no bearing on whether at times prior to making this choice God was omnipotent. In otherwords, the lemma where God chooses to BECOME less than omnipotent by creating such a stone does not show that it was impossible for God to have been omnipotent.

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    Rich Griese said…

    Would be good to establish that there are gods before you begin attempting to talk about their attributes. People often make this mistake. Since no gods have been demonstrated, talking about them is a waste of time.
    ============
    Response:

    I believe the opposite to be the case. Before one tries to establish the existence of God, one must first clarify what is meant by the word 'God', and this requires that one talk about the alleged attributes of God.

    Richard Swinburne's book The Coherence of Theism focuses in on two questions: (1) What does the sentence 'God exists' mean? (2)Do believers make a coherent claim by uttering the sentence 'God exists'?

    These two questions need to be answered before one can properly provide or evaluate arguments for the existence of God.

  • http://www.blogger.com/profile/07559081710058635050 Pulse

    Copied from Part 1.

    Dianelos Georgoudis said…

    "God does not do absurd things such as creating a four-sided triangle or a stone too heavy to lift because God, being perfect, never wants to do absurd things. Similarly God never wants to do evil. So it’s not a question of what God *can* do (for God, being limitless could do anything whatsoever), but of what God *wants* to do. I think this far it is easy to agree."

    A) You are incorrectly conflating that which is impossible with that which is absurd. The two are not one and the same.

    B) You are asserting God's motives and desires without evidence. The question of God's sanity and benevolence is separate from the question of God's omnipotence.

    C) You are improperly redirecting the question by assuming the very thing that is under question, God's omnipotence. Motivation is irrelevant unless ability is first established.

    D) The statement "God can do anything God wants to do" is a very different and much weaker statement than "God can do anything". The former does not establish God's omnipotence because the very same statement can be applied to dirt. Dirt can do anything dirt wants to do.

  • http://www.blogger.com/profile/02447457531584294267 Joe

    Bradley Bowen said…

    "That is, in essence, the solution that Swinburne offers for the paradox of the stone. So long as God is merely able to create a stone he cannot move but chooses not to exercise this ability, there is no problem. It is only if God chooses to exercise this ability that there will be a stone that God cannot move.

    But in that case, God would be making the choice to BECOME less than omnipotent. Thus, this has no bearing on whether at times prior to making this choice God was omnipotent. In otherwords, the lemma where God chooses to BECOME less than omnipotent by creating such a stone does not show that it was impossible for God to have been omnipotent."

    Interesting…so where does this leave the Paradox of the Stone then?

  • http://www.blogger.com/profile/18095903892283146064 Richard Wein

    I've always been sceptical of the claim by some theists that God necessarily has certain properties by definition.

    Suppose it turned out that there is a being that did all the things God is claimed to have done in the Bible, and has all the properties normally associated with God, bar one: the being in question is vastly powerful but not omnipotent. Would those theists say, "Well, it seems God doesn't exist after all. But there exists some other being with very similar properties. What shall we call Him?" Of course they wouldn't. They would call that being God. But in that case omnipotence is not a necessary property of God.

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    Joe said…

    Interesting…so where does this leave the Paradox of the Stone then?
    ============
    Response:

    To be honest, I have not thought enough about the paradox of the stone to give you a firm answer.

    Swinburne's proposed solution seems correct to me in relation to his stripped-down concept of a "contingent God". If God just happened to be omnipotent (i.e. if being omnipotent was not a necessary or essential property of God) then God could choose to make himself less than omnipotent, and there would be no real problem from the paradox of the stone.

    Even if being omnipotent were essential to being God, God could still choose to shed his omnipotence, but this would, in effect, be committing suicide, because God would cease to exist upon losing his omnipotence. Nevertheless, essential omnipotence seems compatible with the paradox of the stone.

    However, if God's existence or eternality was a necessary or essential property of God, then God cannot destroy himself by eliminating one of his other essential properties.

    In this case, it would be the combination of both essential eternality and essential omnipotence that would rule out the lemma where God has the power to create a stone that he cannot lift. Such a God would not have this power.

    But then, if I were a theist, I would follow Swinburne's path by restricting the definition of 'omnipotence' so that it automatically excludes any actions that are logically incompatible with God's other divine attributes.

    The real heart of the matter thus would be: To what extent does such a narrowed conception of 'omnipotence' create significant liminations on what God could do?

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    Richard Wein said…

    Suppose it turned out that there is a being that did all the things God is claimed to have done in the Bible, and has all the properties normally associated with God, bar one: the being in question is vastly powerful but not omnipotent. Would those theists say, "Well, it seems God doesn't exist after all. But there exists some other being with very similar properties. What shall we call Him?" Of course they wouldn't. They would call that being God. But in that case omnipotence is not a necessary property of God.
    ===========
    Response:

    I agree with your claim about how people would likely respond to the described scenario. However, it is not clear to me that this establishes the conclusion that "omnipotence is not a necessary property of God".

    One could interpret the response to the described scenario as one where believers have decided to change the meaning of the word 'God' to fit with a new understanding of reality.

    If these believers were clear-headed and honest, they would say:

    "Yes, it is true that 'God exists'is false based on the meaning of the word 'God' as we previously used this word. But now, with this new information we have revised the meaning of the word 'God' a bit, so that in the new sense of the word the claim 'God exists' is now true."

    If believers in the scenario you descibe characterized their change of mind this way, would your conclusion still follow?

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    A second objection to Richard Wein's comment…

    According to Swinburne's account of the descriptive theory of proper names, the referent of the proper name 'God' is the individual, if there is one, who has MANY of the divine attributes discussed by Swinburne (e.g. omnipotence, omniscience, perfect freedom, etc.).

    On this understanding of the meaning of 'God', the meaning of 'God' could be tied to the property of 'essential omnipotence' and one could still identify or pick out the individual who is 'God' even if that individual was not essentially omnipotent.

    In other words, the connection between some proper names and their associated definite descriptions is a loose one that allows for some range of different beings to count as the referent of the name.

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    I hope that I'm not totally confused in my responses to Richard Wein, but I might be.

    It is challenging to keep clearly in mind all of the distinctions and complexities involved here:

    - the descriptive theory of proper names
    - essential properties vs. non-essential properties
    - de re vs. de dicto necessity
    - analysis of ordinary use of a word vs. proposed technical definitions

    (to name a a few key ideas that come to mind).

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    More reflections on Richard Wein's objection…

    A believer at one point might be convinced that God exists, and that God's existence was necessary existence; that is to say, that 'God exists' was a necessary truth, true in all possible worlds.

    Such a believer might then propose an analysis of the proper name 'God' as picking out a unique individual who possesses various divine attributes, including the property of 'necessary existence'.

    At some later point in time, this believer might be persuaded that the claim 'Some person has the property of necessary existence' is an incoherent statement. At that point, the believer would probably revise his proposed analysis of 'God' to drop the divine attribute of 'necessary existence' from the definite description that is associated with this name.

    Such a believer could then reasonably say the following:

    "Under my previous analysis of the proper name 'God' the statement 'God exists' would be false, for there is no person who has the property of 'necessary existence'. However, if we revise my previous analysis slightly, and drop the condition of 'necessary existence' from the list of divine attributes, then, on this new understanding of the word 'God' it would be true that 'God exists'."

  • http://www.blogger.com/profile/18095903892283146064 Richard Wein

    Bradley Bowen wrote:

    >If believers in the scenario you descibe characterized their change of mind this way, would your conclusion still follow?<

    Yes. As far as I'm concerned meanings don't change just because someone decides to give a new definition. Meanings are determined by mental states which cause people to use words in a particular way. People often incorrectly report their own meanings.

    As I see it, a word like "God" serves primarily to refer to an entity. And a definition serves primarily to identify that entity. That's why we can use "ostensive" definitions. For example, I might define "Barack Obama" merely by pointing at the man and saying "that's Barack Obama". Alternatively, I might identify him by defining him as "US president", but it doesn't follow that "US president" is a necessary property of Barack Obama, even if I perversely insist that it is. In the present case, it seems that the theists are still using the word to refer to the same entity as before (although one of their beliefs about that entity has changed). So I would say that the primary component of the meaning has not changed.

    Another component of meaning is a matter of the limits to which we are willing to extend our use of a word. If the entity we currently call Barack Obama turned out to be a shape-changing lizard, would we still call him "Barack Obama"? The answers to such questions can be unclear. We might well feel uncomfortable with calling him Barack Obama, but unwilling to say that such usage was distinctly right or wrong. Theists might say, "The entity we called God exists, but we're no longer willing to call it God, because it's turned out not to be omnipotent." But we agree that they probably wouldn't say that. They probably would be willing to use the word "God" to describe a non-omnipotent being.

    So it seems that the meaning of "God" has not significantly changed for your hypothetical theists. All that's changed has been (1) they have a different belief about God, and (2) they have dropped their erroneous claim that for them omnipotence is a necessary property of God by definition.

    If you still disagree, then please tell me what you understand by the assertion that God necessarily has some property by definition. I take it to mean that the relevant group of speakers would not call an entity "God" unless it had that property.

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    Richard Wein said…

    Theists might say, "The entity we called God exists, but we're no longer willing to call it God, because it's turned out not to be omnipotent." But we agree that they probably wouldn't say that. They probably would be willing to use the word "God" to describe a non-omnipotent being.

    ========
    Response:

    Yes. That sounds right to me.

    Although the word 'God' would not describe the being, but would be used to name and to identify the being.

    Assuming the descriptive theory of proper names applies to the word 'God', there would be a description associated with the name 'God' and the description would allow us to pick out or identify a particular being.

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    Richard Wein said…

    So it seems that the meaning of "God" has not significantly changed for your hypothetical theists. All that's changed has been (1) they have a different belief about God, and (2) they have dropped their erroneous claim that for them omnipotence is a necessary property of God by definition.

    If you still disagree, then please tell me what you understand by the assertion that God necessarily has some property by definition. I take it to mean that the relevant group of speakers would not call an entity "God" unless it had that property.
    =============
    Response:

    The way that Swinburne explains the descriptive theory of proper names, there does not appear to be a way to specify a particular necessary condition for the correct application of a proper name, at least not in cases where multiple criteria or properties are specified, as is the case with the proper name 'God'.

    When the description associated with a proper name involves multiple criteria or multiple properties, then the name is said to apply when there is an individual who has MANY of the specific properties given in the description.

    On this account of proper names, given a list of eight divine attributes there are eight possible combinations of the attributes in which seven apply and one does not (the 1st attribute might be the missing one, or the 2nd attribute, or the 3rd, etc.). There are even more possible combinations where only six attributes apply and two do not (I believe there are 28 different such combinations of attributes).

    The descriptive theory of proper names that Swinburne describes might be overly simple, and the logical relationship between descriptions and proper names might be more varied and complex than the idea that just MANY of the properties must apply in order to identify the correct referent of the name.

    Nevertheless, based on the simple form of the descriptive theory of proper names, it looks like no individual simple property could be a necessary condition for the correct application of the proper name 'God'.

    Swinburne uses criterial definitions in his analysis of the concept 'person' and also for the concept of 'having a body'. In using the version of the descriptive theory of proper names that he explains, Swinburne in essence connects the name 'God' to a criterially defined concept.

    In conclusion, if we want to make omnipotence a necessary condition for something counting as 'God' then I think we have to abandon the descriptive theory of proper names, or at least the simple version of it that Swinburne provides.

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    Richard Wein said…

    If you still disagree, then please tell me what you understand by the assertion that God necessarily has some property by definition. I take it to mean that the relevant group of speakers would not call an entity "God" unless it had that property.
    ==============
    Response:

    See my most recent post on de dicto vs de re necessity.

    To say that "the relevant group of speakers would not call an entity 'God' unless it had that property' appears to be talking about de dicto necessity.

    For example, to say that the relevant group of speakers would not call an entity 'God' unless it had the property of omnipotence, appears to be talking about de dicto necessity of having the property of omnipotence.

    Also talking about 'God' having the property of ominpotence 'by defintion' appears to be to talk about de dicto necessity of possessing this property.

    I'm a bit hesitant to buy into your proposed criterion of de dicto necessity. What people 'would' or 'would not' call 'God' seems to be a bit thin as a criterion. I think one needs to say something about why the people would or would not use the word in question, and not merely the facts about when the word is applied or withheld.

    The 'why' is needed to allow room for the misuse and abuse of language. People might withhold application of a word in certain circumstances for irrational or inadequate reasons.

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    Is omnipotence an essential property of God? I take it that personhood is an essential property of Bradley Bowen (me). If I cease to be a person, then I will cease to exist.

    So, assuming that exactly one divine being exists (a person with the properties of omnipotence, omniscience, and the other divine attributes discussed by Swinburne), and assuming that this being is currently omnipotent, would this being cease to exist if it became less than omnipotent tomorrow morning?

    Let's say that this being's power was reduced to the level of an ordinary human being.

    I, and most believers, would be inclined to call this being 'God' if it had possessed the various divine attributes for a significant period of time in the past (and had no competitors for the name 'God').

    I, and most believers, would be inclinded to say, on learning that the being had lost its omnipotence, that 'God no longer exists' but that the person who once was 'God' still exists.

    This suggests that being eternally omnipotent is not a necessary condition for being 'God', since the name 'God' would seem to apply to a being who was omnipotent for a long time, but not for eternity.
    It also suggests that being omnipotent (or at least nearly omnipotent) is a necessary condition for being 'God'.

    So, it seems possible to me that a person could be omnipotent and God one day, and continue to exist as a person but not be omnipotent and not be God the next day. If this is logically possible, then omnipotence would not be an essential property of God, but might well be a necessary condition for being 'God'.

  • http://www.blogger.com/profile/05211466026535549638 Bradley Bowen

    Richard Wein said…

    Theists might say, "The entity we called God exists, but we're no longer willing to call it God, because it's turned out not to be omnipotent." But we agree that they probably wouldn't say that. They probably would be willing to use the word "God" to describe a non-omnipotent being.

    =============
    Response:

    If we accept the following claim by Swinburne, then Richard's conclusion is unavoidable:

    1. 'God' is a proper name that denotes or picks out an individual, if there is such an individual, who satisfies MANY of the following descriptions:
    (a) is eternally a spirit (i.e. a bodiless person),
    (b) is eternally omniscient,
    (c) is eternally omnipotent,
    (d) is eternally perfectly good,
    (e) is eternally omnipresent,
    (f) is eternally the creator of the universe (i.e. has always had and always will have power to cause any logically contingent being to cease to exist),
    (g) is a source of moral obligation.

    Based on this assumption, it follows that if there were a being that satisfied all of the above descriptions except for (c), and if no other being satisfied all or nearly all of these descriptions, then that being would properly be called 'God', and the sentence 'God exists' would clearly be true.

    However, this clear implication of Swinburne's position seems questionable to me, and suggests to me that his analysis of 'God' is not quite right.

    Nevertheless, I think both Richards (Swinburne and Wein) are right in pointing out that there is some flexibility in the word 'God'. I'm just doubtful, at this point, that the nature of the flexibility has been correctly characterized by either Richard.


CLOSE | X

HIDE | X