Saturday, November 16, 2013

Review: The Coherence of Theism by Richard Swinburne

Introduction. In The Coherence of Theism (henceforth CoT), Richard Swinburne is concerned with examining whether the central doctrines of classical theism[1] are coherent (that is, free from contradiction). It is important to understand what this limited goal means for the theist/atheist debate. If the atheist can convincingly claim that theism is incoherent, then the game is over: theism cannot be true. For the theist to show that theism is coherent is only a partial victory. Then theism is in some sense possibly true. Further grounds would have to be given to show that it is in fact true. [2]

Part I: Religious Language. In Part I, Swinburne sketches various criteria for coherence. The basic means for testing whether or not some proposition is (in)coherent is to ‘unpack’[3] the concept and see what results. This is necessary because while some contradictions are explicit (The author of Hamlet did not write Hamlet) many others are implicit. Swinburne closes out this section by tackling a handful of topics, including ways in which theology involves ‘mundane’ uses of words (i.e., theological terms are defined using words that are close to their more pedestrian, ‘every day’ usages) and more ‘analogical’ or ‘stretched’ uses of words. Even when theology is using words in more mundane ways, the properties involved (e.g., power, goodness) may involve those properties occurring in unfamiliar combinations[4]. Swinburne helpfully notes that this occurs not only in theology, but also in science.[5] Using non-religious examples, Swinburne illustrates how mundane language implies only similarities, not absolute sameness. Although myself and my desk are alike in some ways (we both are composed of matter, have weight, occupy space) we are very different in other ways (I am made of flesh and bone, can lose weight through diet and exercise, and breathe—none of which apply to my desk).

In discussing more analogical language, Swinburne proposes that such language be understood as follows. When we propose some property X, we consider standard examples of X and ~X. We then argue for an extended or analogical use of X which we can call X2. We can approach understanding X2 by saying that X2 resembles the standard examples of X more than it does the standard examples of ~X. Use of such analogical language makes demonstrations of (in)coherence a lot more difficult—a point Swinburne will return to at the end of the book. Again, as with more mundane uses of language, Swinburne points out that such analogical uses also occur in science (e.g., in Quantum Theory).

While Swinburne acknowledges that dealing with the Ultimate Fact will inevitably involve use of analogical language and the ‘mystery’ card, he argues that resorting to analogical language should occur only when absolutely necessary. Many (but not all) of the divine attributes, he thinks, can be shown to be coherent without having to play the mystery card. [6]

Part II: A Contingent God. Next, Swinburne turns to a discussion of those divine attributes which he believes can be shown to be prima facie coherent without appeal to analogy/mystery. He does so in a cumulative fashion. For example, his chapter headings include ‘An Omnipresent Spirit’, ‘Free and Creator of the Universe’, ‘Omnipotent’, ‘Omniscient’, ‘Perfectly Good and a Source of Moral Obligation’, and ‘Eternal and Immutable’. In the first chapter just listed, he examines the (in)coherence of God being an omnipresent spirit, in the next he examines the (in)coherence of a God being an omnipresent spirit who is also free and creator of the universe, etc. He proceeds by considering what it means to be an embodied being like ourselves, gradually removes those attributes which prevent us from being omnipresent, and argues along the way that no incoherence emerges. Swinburne argues fairly convincingly that these traditional doctrines, as he defines them, do not entail any apparent contradictions either in themselves or in co-occurring.[7] His particular definitions of the doctrines are for the most part fairly traditional[8], with the exception of Swinburne’s arguments for open theism[9]. The most controversial discussion of the divine properties will, for most theists, occur in Part III.

Part III: A Necessary God. Swinburne then considers two final claims theists wish to make about God: first, that God is a necessary being and, secondly, that God necessarily possesses the attributes that he does. [10] Swinburne first examines six different kinds of necessity[11], and then sees which of these senses can be used to coherently state the two aforementioned claims about God. Swinburne’s analysis of types of necessity results in the following verdicts: God’s existence is necessary in terms of causal ultimacy not logical necessity, and God’s possessing the characteristics he does is one of a posteriori necessity. (As Swinburne’s arguments for God being causally ultimate rather than logically necessary are more controversial, I relegate the discussion of God’s attributes being a matter of a posteriori necessity to a lengthy footnote.[12])

Logical necessity is the type of necessity, Swinburne thinks, that most theists wish to claim for God. This means that ‘God exists’ is true simply in virtue of what the statement is. It does not, unlike a posteriori necessity[13], require any presuppositions about further facts. If ‘God exists’ were logically necessary, Swinburne argues, this would mean that its denial (‘God does not exist’) would be necessarily false—that is, would be incoherent. However, neither the statement itself nor any propositions which it is known to entail seem to be incoherent and thus there is no evidence that ‘God exists’ is logically necessary.[14] Swinburne also rejects the other types of necessity he has examined, and concludes that God is best thought of as the ‘ultimate Brute Fact’: that fundamental datum of reality which is the explanation for all else that is, but which itself has no explanation.[15] Swinburne concludes that the atheist should turn their guns to demonstrating that theism is untrue rather than incoherent. In the last and final chapter, Swinburne argues that the type of God he has argued for is worthy of worship, both because he is perfectly good and because, as the causally ultimate Being, we owe all we have and are to him.

Conclusion. Reading The Coherence of Theism was exhilarating, for several intertwined reasons. First, it sets the stage for Swinburne’s later fantastic work. Second, it demonstrates how systematic the man is in his approach to philosophy of religion. Consider: he tackles whether or not theism is possibly true (The Coherence of Theism), then looks at grounds for thinking theism to be (probably) true (The Existence of God), then looks at the interplay between faith and reason (Faith and Reason) before turning to his famed series on specifically Christian doctrines (Responsibility and Atonement, Revelation, The Christian God, Providence and the Problem of Evil) and finally to the evidence for Jesus (The Resurrection of God Incarnate). Third, in many ways The Coherence of Theism demonstrates the resurgence of theism in modern philosophy. Swinburne’s recommendation that atheism turn its guns on the truth of theism rather than its coherence seems to be commonly accepted now. Whether one accepts all of Swinburne’s qualifications on the divine attributes is another matter.[16] Again, a fantastic work from one of Christianity’s great thinkers.


Apologetics 315 Book Reviewer Latter Day Inkling is a U.S.-based research psychologist for the military. He is especially interested in epistemology and natural theology.


[1] That is, those doctrines that many theists (whether Jewish, Islamic, or Christian) would espouse.
[2] Obviously, however, if theism is true that is grounds for thinking that it is coherent—a point that becomes relevant later.
[3] In philosopher language, a proposition is described in such a way that it becomes clear(er) what further propositions it entails. If a contradiction emerges—that is, is entailed by—the original proposition, the original proposition is itself contradictory, is hence incoherent, and is hence false.
[4] I.e., God as a bodiless being, God being omnipotent, etc.
[5] See The Coherence of Theism, page 53.
[6] Swinburne spends some time discussing and dismissing logical positivism/verificationism. As this is considered a dead issue in current philosophy, I skipped this material.
[7] In more technical language, the traits appear not only possible but compossible. That is, not only is omnipresence internally coherent but bundling it together with the other properties does not result in incoherence.
[8] I.e., Swinburne considers the only limit to God’s omnipotence one of logical possibility. He also argues that moral properties ‘supervene’ on natural properties so that some states of affairs are necessarily good or bad, whether or not God exists. Other states of affairs are contingently good or bad, depending on God’s commands and hence God’s existence. Swinburne also argues strongly against a ‘strong’ sense of immutability ala Aquinas, which, he feels, seems incoherent: how could a strongly immutable God interact with a changing world?
[9] Swinburne cites Nelson Pike’s influential argument for the incompatibility of exhaustive divine foreknowledge and human libertarian freedom.
[10] “The first such claim is that God does not just happen to exist. It is not a matter of fortunate accident that there exists a God; he exists necessarily. The other is that God is necessarily the kind of being which he is; God does not just have the properties which he does. It is not by chance that he is omnipotent or omniscient. Being omnipotent or omniscient is part of God’s nature.”—page 242
[11] Logical necessity, a posterior necessity, necessity of the past, causal ultimacy, everlasting existence, and physical necessity. Understanding each of these is best approached via examples—see Chapter 13.
[12] Swinburne’s discussion of a posteriori necessity is a model of lucidity, albeit a discussion that demands careful (re)reading. He proceeds via a series of examples drawn from Kripke and others. E.g., consider your car. What attributes are essential and which are nonessential for it to remain a car? However, the discussion of God’s essential characteristics is more ambiguous because some of the criteria proposed for material objects (continuity of matter) are not available for an immaterial being. Swinburne also argues summarily in CoT (and at much greater length in The Evolution of the Soul) that what makes a person the same person is an immaterial matter-the possession of the same soul. This leads Swinburne to propose a criterion of ‘continuity of experience’ by which we may judge A to be the same person as B. This yields a possible incoherence in defining God’s possession of his attributes as ‘a posteriori’ necessary. This incoherence, he thinks, can only be avoided by at long last resorting to analogical use of the term ‘person’. He notes yet again that such analogical use of language occurs in non-theological settings (i.e., science). Ultimately he concludes that it is difficult if not impossible to determine the in(coherence) of God’s necessarily possessing these properties. He therefore posits that the most promising avenue for the theist to pursue in demonstrating coherence here is to give grounds for believing the doctrine to be true. He essentially does this, briefly, in his is There a God?, location 649 Kindle edition: “It seems to me that it is simpler to postulate not merely that God is eternally infinitely powerful, knowledgeable, and free, but that he is so essentially. If we were to say that it is only an accident that God is infinitely powerful, etc., if we allow that God could, if he so chose, abdicate. He could reduce himself to a being of limited power. He could even commit suicide…..But all of that would make it much less the foundational brute fact that our God is the course of all that is. It would need to be explained why God had not already limited his powers or committed suicide….All that no longer requires explanation if we suppose that God is essentially omnipotent, omniscient, perfectly free, and eternal.” Swinburne is thus arguing that, on the grounds of simplicity (evidence of something’s being true) we have evidence that such a supposition (i.e., God possessing his traits necessarily) is coherent.
[13] E.g., ‘The number which is the number of the planets is greater than 6.’ Because the number of the planets is 9, this proposition is necessary. See page 244.
[14] Swinburne also argues, much more quickly, that the world seems to be contingent but that nothing logically contingent can derive from anything logically necessary. This is a vast topic in its own right, with some heavyweight philosophers (e.g., Peter Van Inwagen) agreeing with Swinburne and others (William Lane Craig, Alvin Plantinga) demurring. These conflicting sets of intuitions are part of the reason why Swinburne and Van Inwagen disavow the contingency argument, while Craig and Plantinga promote it.
[15] Swinburne sees this as a great-making property of God, contrary to many classical theists. “For the theist the existence of God is a tremendous thing, the most fundamental truth about the universe. It seems to trivialize it to say that it holds for the same reason as does the truth that all bachelors are unmarried.”—page 275
[16] One who disavows Swinburne’s open theism but thinks that his theodicy in Providence and the Problem of Evil still works, for example, will a fortiori think that a theist who holds a more robust view of God’s omniscience will a fortiori have even more confidence in their own theodicy. In assessing The Coherence of Theism, if one thinks that Swinburne is correct in that the God described by Swinburne is worthy of worship may be, again a fortiori, even more worthy of worship if God in fact is the ontological foundation for moral facts. For argument that even necessary (i.e., moral) truths cannot exist a se but are rooted in God, see Good God by Baggett and Walls.

3 comments :

Paul McCauley said...

If the ontological argument is right then would it not be true that showing the coherence of theism is more than a partial victory? If theism can be true then it must be true (if the ontological argument goes through). Is that correct?

Peter Schaefer said...

Plantinga's MOA (Modal Ont Arg) does have that implication. If it is possible that God exists (exists in some possible world) then as a necessary being he exists in all possible worlds...including the actual. Note that Swinburne has no faith whatsoever in ontological arguments (or indeed deductive atgents) for theism.

Paul McCauley said...

Gotcha. Thanks Peter.

Post a Comment

Thanks for taking the time to comment. By posting your comment you are agreeing to the comment policy.

Blog Archive

Amz