Thursday, May 5, 2011

Morality and Possible Worlds, Part 1

Note: This is a paper I presented at the ISCA annual meeting in Wake Forest, NC.

            The problem of evil has been an issue in Christianity for as long as anyone, seemingly, can remember. I will spare the well-known details of its origination other than to say it is one of the chief objections against the existence of an omnibenevolent God. Underlying the objection and often expressed is the question: “why?” Why did God allow the world to be this way?
            I saw a “demotivator” poster which had a picture of a boxer receiving a vicious blow to the face which presumably left him staggered and beaten. Underneath was the (slightly) humorous caption, “not all pain is gain.” Most, if not all of us, would grasp the meaning of this quite experientially. This is precisely why the problem of evil can be a problem. The world does seem to have pointless suffering; it does seem to have senseless violence. But does it really?
            William Lane Craig points out that even if evil is gratuitous, God may be justified in allowing it. Further, Craig also believes “the evil, though objective, wouldn’t be gratuitous.”[1] In either case, then, the evil allowed by God does not count against his moral worth or character. This paper will suggest God is morally justified in creating this actual world over other worlds which he could have created.
            The paper will first examine the concept of a possible world. It then will assess criteria for moral preferability among different possible worlds. A brief look at Gottfried Leibniz’ conception of the best possible world will follow, along with God’s moral obligation to create such a world, if it exists. Finally, a positive argument will be presented to show God is morally justified in creating this actual world.
Possible Worlds
            First, if one is to understand God’s moral justification in creating the actual world then he must understand the concept of a possible world. A possible world is “a maximal description of reality, or a way reality might be.”[2] Possible worlds are governed, in general, by broad logical possibility.[3] For instance, there is a possible world in which I choose not to write this paper. There is a possible world in which one’s parents never meet and so he is never born. There is a possible world in which every Thursday every person on planet Earth shaves his right arm, and so on. It is also notable that what is physically impossible may itself be logically possible, as these are not identical.[4]
            It follows then the actual world in which we live was itself one of these nearly-infinite possible worlds.[5] From these, God could choose to instantiate any he so desired. In relation to creating conscious creatures as his image-bearers (Gen. 1:26), it would have been important to endow such creatures with the ability to make moral choices. This follows from God’s being a moral agent himself (as the objective grounds of morality).
            It is this ability to make free and morally-responsible decisions/actions that tend to delimit possible worlds into feasible worlds. Kenneth Keathley puts it this way: “He [God] knows what reality would be like if He had created a world without you or me in it. . .God possesses a perfect knowledge of all feasible worlds—all possibilities which would accomplish what He wanted to have happen.”[6]
            Further explaining the concept of feasible worlds in relation to Christian particularism, Craig notes, “so long as people are free, there is no guarantee that everybody in such a world would be freely saved.”[7] So it seems, given free actions, certain worlds are logically possible but not feasible (or able to be put in actualization by God [who is bound by his nature, which grounds logic and truth]).
            Craig gets even more specific in a different article. “He [an objector] must show that the circumstances under which various individuals would freely receive Christ are compossible, so that all persons in some possible world would freely receive Christ and be saved.”[8] The worlds inaccessible to God in a feasible sense are just those worlds which contain non-compossible truths; that is, worlds which contain truths about moral decisions/actions which cannot both be true. If Gary chooses to act morally right at time t-1 in circumstances C1, but C1 entails Lisa does what is morally right at t-1 only in the case that Gary does what is wrong there, then a world in which both Gary and Lisa do right simultaneously given C1 is not feasible. C1 is a broadly logically possible world, but it is infeasible for God to create given morally-free and responsible actions.
            Some allege God is responsible for circumstances in which people make moral choices (both with good and bad applications). The problem with such a view, according to Baggett and Walls, is that “it presumably involves people’s genuinely free choices and their consequences. Just because God foreknows the content of our decisions doesn’t mean he’s responsible for determining that content, nor does it preclude the ability to do otherwise.”[9]
            Both the ideas and implications of possible and feasible worlds will come into play throughout this paper. It is easy to think of alternative possible worlds; simply imagine the events of the world until now remaining uniform with one miniscule difference. However, considerations as to feasible worlds are much more difficult epistemologically.
Moral Preferability Among Worlds
            Imagine no constraints on one’s knowledge as to feasible worlds. At a minimum, consider the possibility that there are multiplied possible worlds which are also feasible, so that many complete descriptions of reality were available to God for creation. What standard should be used to judge the moral preferability of one world to another? That is, what makes it true that world W1 is better than W2 in a moral sense? Should the standard be what is ultimately best for all moral agents? An analysis of certain proffered axioms should take place concerning moral preferability among worlds.
An Analysis of Proffered Axioms
            1. Any world W1 is morally preferable to any other world W2 provided W1 contains more human flourishing or well-being than W2.
            This is not a typical theistic axiom when it comes to discussions of moral preferability. Indeed, this is the domain of atheistic naturalists who also believe in objective morality, such as Sam Harris.[10] Because of this it would be unfair for an atheist or skeptic of God to press this axiom as incumbent upon the theist in any defense against the problem of evil. That said this axiom does not resist critical analysis.
            First, there are internal problems with the “well-being” axiom. An unfortunate result can be illustrated by the famous “utility monster” (UM) thought experiment.[11] While proponents of the well-being axiom (and certainly Harris) would object to this axiom’s characterization as old-line utilitarianism, it nonetheless remains that well-being entails happiness on some level. Even if it were not the same problem attending utilitarianism would do so to the well-being axiom as well.
            All one must do to engender this result is to alter the thought experiment a little bit. Instead of the UM’s consuming greater and greater amounts of happiness by killing, torturing, destroying, and maiming, suppose the UM consumes greater and greater amounts of well-being. By doing these aforementioned actions, the highest level of overall well-being is achieved (so long as the UM is a human or group of humans). Such a scenario is logically possible (even if not likely). Nozick asks, “Is it all right…?”[12] Our moral intuition, which is itself a good source of moral epistemology, recoils at the thought.[13] Of course it is never “all right” to harm and murder and maim, even if it increases the overall state of human well-being!
            If it is never all right to harm others in this way to increase one’s own well-being, even in the face of increasing the overall well-being of humans, speaks to this axiom’s falsehood. Even if one were to amend the axiom’s use of “overall” to mean “every human being” there remains a problem. For would not our moral intuitions be repulsed by a logically-possible world in which everyone’s well-being and flourishing were increased by these acts?[14] Another attendant problem on this view is that not only are such actions in these worlds preferable, they are also obligatory.[15]
            Harris himself acknowledges a potential problem with this individualistic view when he writes, “It is clear that we face both practical and conceptual difficulties when seeking to maximize human well-being. . .[there are principles] that at their extremes. . .[are] hostile to the other.”[16] He goes on to say, “in this case, rapists, liars, and thieves would experience the same depth of happiness as the saints.”[17] In these cases, then, a world is morally preferable to another even in the event that the latter world has more well-being than the former.
            Second, the well-being axiom has external problems. A Christian theist would not claim that God has, as the ultimate goal of morality, a paradigmatic well-being axiom. Paul Copan alludes to this, claiming, “a moral universe and human dignity are best explained in the context of a morally excellent, worship-worthy Being as their metaphysical foundation.”[18] That is, the goal of life is not derived from human well-being but from conforming to a standard of morality (since the two are not identical). In that case, any world which contains a greater adherence to the objective moral standard of values (rooted in God’s nature) and commands (which flow in accordance with those values and not contrary to them) is actually preferable to a world of well-being without such adherence.
            2. Any world W1 is morally preferable to any other world W2 provided W1 contains more morally-good acts than W2.
            This particular axiom enjoys a sort of intuitive support found throughout the world. In fact, it is a common refrain one hears when discussing matters of spirituality and ethics in regards to salvation. “If my good deeds outweigh my bad deeds, then I will get into Heaven.” In much the same way, if more good deeds than bad deeds are preferable for salvation, then surely they are for choosing among worlds as well.
            The potential problem is that this seems to open the door for any number of odd-world scenarios. First, consider this: suppose in W2 there is, on average, one good act done for every man, woman, and child on earth every day. If this were our actual world this would equate to over six billion good acts per day (at this present time). However, in this case it would seem that W1 would be preferable if only there were more people (in total) existing throughout the world’s history (provided it was identical in every other respect [which is logically possible]). Preferring a world almost purely on the basis of its being more populated is, in moral terms, very nearly arbitrary.
            Second, this axiom does not account for the moral evil and wrong done between two or more worlds. Suppose W1 has the most morally-good actions of any possible world (for the sake of argument). Further suppose this entails also the most morally-evil or wrong actions done among any worlds that are possible. In that case should one be so quick to label W1 as a morally preferable world to the world with the second-most morally-good actions?
            Third, this axiom fails to account for salvation. It is true that Heaven and final salvation are part of this actual world, so that the number of those saved counts toward the overall moral good in each case. However, since God is just, the number of those sent to Hell is good in each case as well. In this case, such a world may be morally preferable to another world even in the case that all of its members ended up in Hell.[19] God would not and does not view that world as morally preferable to one in which there are indeed saved people (1 Tim. 2:4; Ezekiel 33:11). In that case, then, such an axiom must be incomplete at best and false at worst.
            3. Any world W1 is morally preferable to any other world W2 provided W1 contains more morally-good acts and saved individuals than W2.
            This axiom appears quite within the range of biblically-minded people’s comfort. This combines the second axiom above with an answer to one of the objections against that axiom. Since God prefers the salvation of individuals to their damnation, and a good God prefers good acts to bad ones, their conjunction provides a way of achieving God’s goals in accordance with moral preferability. But is this true?
            It seems there may be a problem with this axiom as well. In dealing with a similar axiom, Craig points out this type of world may preclude worlds we would consider to be better. “As a loving God, He wants to minimize the number of the lost: He wants hell to be as empty as possible.”[20] An objector may be confused on this point. He may be tempted to think, “is this world not such a world? After all, people are either saved or lost. Therefore, maximizing the saved automatically reduces the lost.” This would be confused. The axiom merely states that any world containing more saved people and good actions is preferable to any world which contains less of both; it specifies nothing of the number of those who are in Hell.
            As a thought experiment, consider W1 to be a world in which, throughout history, four billion people come to a saving relationship with God. Consider W2 to be a world in which 3,999,999,999 come to know God. Further suppose both W2 and W1 to contain the same number of morally good acts.[21] Now suppose W1 contains three billion souls in Hell, while W2 has two billion. Are we really willing to say W1 is better than W2 after all? While this axiom is likely on the right track, it seems it too must be rejected as incomplete.
Axiom in Light of God’s Grounding of Objective Morality
            A discussion of insufficient axioms has taken place; however, now an axiom must be presented that will be acceptable with respect to moral preferability among worlds. Moral preferability of any non-relativistic stripe will be objective and independently binding; that is, one is trying to establish what really is better given a set of circumstances. The question of whether or not God is constrained to create such a world will be discussed later. What follows is suggested to be the Axiom of Moral Preferability (AMP).
            4. Any world W1 is morally preferable to any other world W2 provided W1 contains an optimum number and balance of saved individuals in comparison to W2.
            This axiom will almost certainly entail many morally-good actions in light of free will. The AMP differs from (3) in that it does not simply promote the number of saved individuals in a given world but the optimum balance as well.[22] Craig alludes to this when he says, “It is possible that in order to create the actual number of persons who will be saved, God had to create the actual number of persons who will be lost. It is possible . . . any other possible world which was feasible for God the balance between saved and lost was worse.”[23]
            In theism and Christianity, God is the absolute standard for objective morality. So inextricably tied to moral values is God that Baggett and Walls remark, “The force of the moral argument is that theism is no more outlandish or outrageous than many of our most cherished moral convictions.”[24] Since God’s nature is the standard of objective moral value and God wants all men to be saved, any world with a higher number of saved individuals which contains an optimum balance of those individuals is preferable to a world with a lower number or containing a lesser balance.
            Defeat of an axiom is possible if there is a counterexample. Is there such an example for the AMP? Suppose W1 contained four billion saved out of six billion in a world’s history. Suppose also that W2 held one less saved person than W1 out of an identical number of those existing. It seems the AMP holds in this situation.
            What if the parameters changed a little? Suppose W1 contained the same number and ratio of saved-to-lost. Further suppose W2 contained 4,000,000,001 out of six billion. Indeed, one would argue that W2 fits the AMP now. Rather than a counterexample, we have simply swapped places between W1 and W2.
            As a final attempted counterexample, what if the proportions were only slightly different? Suppose W2 contained 4,000,000,001 saved out of a world’s history of 6,000,000,002.[25] Despite the balance still favoring W1 there is one more precious soul saved (though one more goes to spend eternity in Hell). This is the most challenging to the AMP.
            I would respond that such a decision between worlds is, at best, vague; it is not at all clear God would prefer W2 to W1 in this case. Vagueness counts against any objection as a defeater.[26] I also think it reasonable that God would prefer W1 even in this circumstance, since W2 involves sending a person to heaven and to hell in a sort of one-to-one correspondence; a balance that is in this scenario less than optimal. With the AMP firmly in place, this paper shall consider the argument from the best possible world.

                [1] William Lane Craig, “Question #196,” <>, accessed April 14, 2011.

                [2] William Lane Craig, Reasonable Faith 3rd ed. (Wheaton, IL: Crossway, 2008), 183. It is also important to note that whatever appears in a possible world can be either contingent or necessary (and in either case it is necessarily possible), and what appears in no possible world is of necessity impossible (e.g., married bachelors, square triangles, 2+2=97, etc.).

                [3] Bob Hale, “Absolute Necessities,” in Philosophical Perspectives, Metaphysics. Vol. 30, 10, (1996:), 93-117. This means they are internally consistent. In strict logical possibility something must be both internally consistent and compatible with other truth. So, while “the U.S. president is a prime number” is broadly logically possible, there nonetheless is no possible world where the U.S. president is a prime number (because of the truth that U.S. presidents are persons), and hence it is not strictly logically possible.

                [4] Ibid.

                [5] I say “nearly-infinite” because any world which contains logical impossibilities is by definition not a possible world; if there are logically-impossible worlds then these reduce the number of the set of all possible worlds to something less than infinite (an infinite series has no end).

                [6] Kenneth Keathley, Salvation and Sovereignty: A Molinist Approach (Nashville, TN: B&H Academic, 2010), 17-18. While Molinism is a distinctive approach in this paper, it is not necessary to the argument of the paper as a whole. So long as any consistent account of the argument proffered is given, it matters not the methodology. Developing an argument for Molinism or any other interpretive framework is not the goal of this paper.

                [7] William Lane Craig, “How Can Christ Be the Only Way to God?” <>, accessed April 14, 2011.

                [8] William Lane Craig, “No Other Name: A Middle Knowledge Perspective on the Exclusivity of Salvation Through Christ,” in Faith and Philosophy, Vol. 6 (1989:), 172-188.

                [9] David Baggett and Jerry L. Walls, Good God: The Theistic Foundations of Morality (New York: Oxford, 2011), 244 n27.
                [10] William Lane Craig and Sam Harris Debate, “Is Good from God?” at the University of Notre Dame, April 7, 2011. Also, see Harris’ book, The Moral Landscape: How Science Can Determine Human Values (New York: Free Press, 2010), 6.

                [11] Robert Nozick, Anarchy, State, and Utopia (n.p.: Basic Books, 1974), 41. “Utilitarian theory is embarrassed by the possibility of utility monsters who get enormously greater gains in utility from any sacrifice of others than these others lose. . .Maximizing the total happiness requires continuing to add persons so long as their net utility is positive and is sufficient to counterbalance the loss in utility their presence in the world causes others.”

                [12] Ibid., 41-42.

                [13] One may question if intuition is a good source of knowledge. For that, I shall provide a basic argument for intuition as a priori knowledge:
                1. If we can hold justified true beliefs independently of any process or perception, then we have intuitive knowledge.
                2. We can hold justified true beliefs independently of any process or perception.
                3. The laws of logic are justified upon their examination (application of empiricism).
                4. Inference is an application of the laws of logic.
                5. Inference must be used upon application of empiricism.
                6. If (3-5), then the laws of logic must be justifiably known.
                7. If (3-6), then the justification is known logically prior to empiricism.
                8. If (3-7), then (2) is true.
                9. Therefore, we have intuitive knowledge ([3-7, 8], MP from [1-2]).

                [14] To say this is not logically possible is question-begging.

                [15] Paul Copan, When God Goes to Starbucks (Grand Rapids, MI: Baker Books, 2008), 15.

                [16] Harris, The Moral Landscape, 187.

                [17] Ibid., 189-190.

                [18] Paul Copan, “God, Naturalism, and the Foundations of Morality,” in The Future of Atheism ed. Robert Stewart (Minneapolis: Fortress Press, 2008), 142.

                [19] Alexander Pruss holds a great but brief discussion on why even the people in Hell owe thanks to God; it is because of the moral good of existence itself! Alexander Pruss, “No One Would be Better Off Not Existing,” <>, first accessed April 4, 2011.

                [20] Craig, “No Other Name.”

                [21] This would be possible if the number of saved performed more good acts than they would in the other world. In any case, this is not logically impossible.

                [22] Additionally, it is worth noting that focusing only on the balance between saved and lost while ignoring the number of those saved is to miss the point of preferability as well. As Craig notes in “No Other Name,” “Even if we grant that God could have achieved a better ratio between saved and lost, it is possible that in order to achieve such a ratio God would have had to so drastically reduce the number of the saved as to leave heaven deficient in population (say, by creating a world of only four people, three of whom go to heaven and one to hell).”

                [23] Craig, “No Other Name.”

                [24] Baggett and Walls, 28.

                [25] Out of pure mathematical interest, W1 holds a balance of saved-to-lost at 66.666666666666666666666666666667%, while W2 has 66.666666661111111112962962962346%.

                [26] At most, it may undercut some warrant we have for belief in the AMP. It would not be a defeater for such a belief, however. See Alvin Plantinga, Warranted Christian Belief (New York: Oxford, 2000), 366.

All posts, and the blog Possible Worlds, are the sole intellectual property of Randy Everist. One may reprint part or all of this post so long as: a) full attribution is given (Randy Everist, Possible Worlds), b) all use is non-commercial, and c) one is in compliance with the Creative Commons license at the bottom on the main page of this blog.

No comments:

Post a Comment

Please remember to see the comment guidelines if you are unfamiliar with them. God bless and thanks for dropping by!