Skip to main content
SpringerLink
Log in

Does Anselm beg the question?

  • Article
  • Published:
International Journal for Philosophy of Religion Aims and scope Submit manuscript

Abstract

Saint Anselm’s ontological argument for the existence of God, formulated nearly a millennium ago, continues to bedevil philosophers. There is no consensus about what, if anything, is wrong with it. Some philosophers insist that the argument is invalid. Others concede its validity but insist that it is unsound. A third group of philosophers maintain that Anselm begs the question. It has been argued, for example, that Anselm’s use of the name “God” in a premise assumes (or presupposes) precisely what has to be proved, namely, that God exists. Another tack is to argue that the premise that God is possible implies or presupposes the conclusion that God exists, or perhaps that, in order to know that God is possible, one must know that God exists. Just as no consensus has emerged about what, if anything, is wrong with Anselm’s argument, no consensus has emerged about whether the argument begs the question. In this essay, I focus on the second type of claim made by the third group of philosophers—the claim that Anselm’s argument begs the question by assuming, as a premise, that God is possible. In particular, I focus on the argument of the contemporary analytic philosopher William Rowe, who has claimed, since at least 1975, that Anselm’s ontological argument begs the question. I argue that Rowe’s argument fails.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Similar content being viewed by others

Notes

  1. Anselm (1033–1109) is said to have formulated the argument—“one of the most-discussed . . . in the history of philosophy” (King 2006, p. 215)—in 1077 or 1078. It appears in his work Proslogion.

  2. This appears to have been Gaunilo’s claim, when he sought to show, by using refutation by logical analogy, that something is wrong with the form of Anselm’s argument. I provide a qualified defense of Anselm against Gaunilo’s attack in Burgess-Jackson (1994).

  3. Philip E. Devine replies to this argument in Devine (1975).

  4. It is an interesting question whether people or arguments (or both) beg questions. My own view, as my title suggests, is that people (e.g., Anselm) beg questions while arguing, but nothing in this essay hinges on this understanding. For the sake of uniformity, I assume henceforth that arguments beg questions, because that is how William Rowe, whose work I go on to discuss, speaks.

  5. “Although surely some arguments are question-begging, it is by no means easy to say what this fault consists in or how it is related to circularity” (Plantinga 1974, p. 217; italics in original). “Few tactics are more common in philosophical argument than accusing one’s opponent of begging the question. Yet it is remarkable how little agreement there is among philosophers about just what such an accusation amounts to, or indeed about what begging the question is” (Biro 1977, p. 257).

  6. I take this reconstruction from Rowe (2007, pp. 41–42). Bracketed material is mine. Here is the pertinent passage of Chapter 2 of Proslogion (translated by M. J. Charlesworth):

    Even the Fool, then, is forced to agree that something-than-which-nothing-greater-can-be-thought exists in the mind, since he understands this when he hears it, and whatever is understood is in the mind. And surely that-than-which-a-greater-cannot-be-thought can-not exist in the mind alone. For if it exists solely in the mind, it can be thought to exist in reality also, which is greater. If then that-than-which-a-greater-cannot-be-thought exists in the mind alone, this same that-than-which-a-greater-cannot-be-thought is that-than-which-a-greater-can-be-thought. But this is obviously impossible. Therefore there is absolutely no doubt that something-than-which-a-greater-cannot-be-thought exists both in the mind and in reality.

    Anselm of Canterbury (1998, pp. 87–88) (italics in original).

  7. We might also say that the cost of nonbelief in the existence of God is nonbelief in at least one of the first three propositions.

  8. See, e.g., Rowe (1979) and Rowe (2010).

  9. See, e.g., Swinburne (1993, p. 4): “[I]t is coherent to suppose that there exists eternally an omnipresent spirit, who is perfectly free, the creator of the universe, omnipotent, omniscient, perfectly good, and a source of moral obligation—given certain qualifications on the ways in which certain of these terms are to be understood.” Some philosophers have made the more modest claim that “it is not irrational to believe that a perfect being is possible” (see Van Inwagen 2006, p. 21), in which case, given the validity of Anselm’s argument, “it is not irrational to believe that a perfect being exists.” The idea, one supposes, is that validity preserves not just truth, but rational belief (or, more precisely, the non-irrationality of belief).

  10. Oppy (1995, p. 54) makes a similar point. He says that on Rowe’s understanding of begging the question (which will be elaborated shortly), “all valid arguments beg the question.”

  11. An anonymous reviewer worries that this turns God into a thing. But surely God is a thing of some sort. Perhaps I should have used “entity,” “object,” or “being” instead of “thing.”

  12. Put differently, suppose, for the sake of argument, that God is in category B. Then, by the great-making principle, God might have been greater than God is. (God might have been in category A.) But God, by definition, is “the being than which none greater is possible.” Therefore, the being than which none greater is possible is a being than which a greater is possible. This is absurd, so the supposition—that God is in category B—is false.

  13. Compare “the numbers 4 and 5 are virtually identical.” To say that two things are virtually equivalent or virtually identical is to imply, conversationally, that they are not equivalent or identical.

  14. Rowe admits as much when he writes, “generally, one can know that a being is a possible being without having to know that the being actually exists” (Rowe 2009, p. 89). Later, he adds, “Normally, when we are asked whether it is possible that a particular being, X, exists in reality, all we need to know is that there is nothing contradictory or impossible in the proposition: ‘X exists in reality”’ (Rowe 2009, p. 90; italics in original).

  15. Quite obviously, neither the premise itself nor “a main conjunct of [the] premise” is identical to the conclusion.

  16. In 1988, in the pages of the periodical Faith and Philosophy, Rowe had an exchange with Georges Dicker. Dicker argued that Rowe’s reconstruction of Anselm’s argument omits a crucial premise, namely, that God exists in the understanding. When this premise is supplied, Dicker maintained, it is no longer the case that the argument begs the question, in which case Rowe’s attempted refutation of Anselm’s argument fails. Rowe replied to Dicker by distinguishing between Anselm’s argument and what he called “the Son of Anselm’s argument.” Only the latter, which Rowe thinks superior to the original (because less objectionable or “mysterious”), succumbs to the charge of question-begging by assuming that God is possible. Rowe then goes on the offensive against Dicker, arguing that Anselm’s argument, no less than the Son of Anselm’s argument, begs the question, albeit in a different way. Anselm’s argument, Rowe says, begs the question by assuming that God exists in the understanding. The Son of Anselm’s argument, by contrast, begs the question by assuming that God is possible. Dicker appears to be persuaded by Rowe’s argument, though he quibbles with Rowe’s characterization of his position. See Dicker (1988a), Rowe (1988), and Dicker (1988b). My criticism of Rowe in this essay applies to both of his arguments. In both cases, he fails to show that a proper subset of the premises cannot be known without knowing the conclusion. All he shows is that one cannot know all the premises without knowing the conclusion, but this, as I have said, is true of all valid arguments, or at least those valid arguments that are addressed to “logically competent targets” (Oppy 1995, p. 55). Surely we would not wish to say that all valid arguments—much less all valid arguments that are addressed to “logically competent targets”—are question-begging.

  17. The reader may wonder what could be wrong with Anselm’s argument if it is not question-begging. As I said at the outset, there are many strategies available to the critic, the most promising of which, in my opinion, are (1) denying the truth of the second premise, which asserts that God is a possible object, and (2) denying the truth of the third premise, which asserts the great-making principle. I believe the focus should be on the truth, not the independent knowability, of the premises. Another thing to keep in mind is that Anselm may not have viewed his argument as a proof, in Rowe’s sense. If he didn’t view it as a proof (or intend for it to be a proof) in Rowe’s sense, then it is no criticism of his argument that it fails as a proof, in Rowe’s sense. Anselm may, like Blaise Pascal in making the wager argument, have intended it to be an argumentum ad hominem, the aim of which is to show one’s interlocutor the logical implications of his or her beliefs, whatever the semantic or epistemic status of those beliefs. Another possibility is that Anselm’s objective was not “to convert the atheist but rather to lead an already formed Christian faith into a deeper understanding of its object” (Hick 2006, p. 17). In any event, “study of the Ontological Argument can be fruitful even though [read: even if] the argument itself fails” (Hick 2006, p. 18).

  18. Plantinga (1974, p. 221) agrees with Rowe that Anselm’s argument (in any of its formulations) fails as a proof, on the ground that the modal claim is neither “uncontested” nor supported by “uncontestable premisses.” But this, he suggests, is too high a standard. The question for Plantinga is not whether the modal claim is known, but whether it is rational to accept it. If it is rational to accept it, then, given the validity of the argument and given that it is rational to accept the other premises, it is rational to accept the conclusion.

  19. See Rowe (2007, p. 46), where he implies, or at least flirts with the idea, that premise 2 of Anselm’s argument is not known.

References

  • Anselm of Canterbury. (1998). Proslogion. In Brian Davies & G. R. Evans (Eds.), The major works (pp. 82–104). Oxford: Oxford University Press.

  • Biro, J. I. (1977). Rescuing ‘begging the question’. Metaphilosophy, 8, 257–271.

    Article  Google Scholar 

  • Burgess-Jackson, K. (1994). Anselm, Gaunilo, and Lost Island. Philosophy and Theology, 8, 243–249.

    Article  Google Scholar 

  • Curd, M. (1992). Argument and analysis: An introduction to philosophy. St. Paul: West Publishing Company.

    Google Scholar 

  • Davis, S. T. (1976a). Does the Ontological Argument beg the question? International Journal for Philosophy of Religion, 7, 433–442. This is a reply to Rowe 1976a. It is criticized by Rowe 1976b.

    Google Scholar 

  • Davis, S. T. (1976b). Anselm and question-begging: A reply to William Rowe. International Journal for Philosophy of Religion, 7, 448–457. This is a reply to Rowe 1976b.

    Google Scholar 

  • Devine, P. E. (1975). Does St Anselm beg the question? Philosophy, 50, 271–281.

    Article  Google Scholar 

  • Dicker, G. (1988a). A refutation of Rowe’s critique of Anselm’s Ontological Argument. Faith and Philosophy, 5, 193–202.

    Google Scholar 

  • Dicker, G. (1988b). A note on Rowe’s ‘Response to Dicker’. Faith and Philosophy, 5, 206.

    Article  Google Scholar 

  • Hick, J. (2006). Ontological Argument for the existence of God. In D. M. Borchert (Ed), Encyclopedia of philosophy, 2d ed., 7 (pp. 15–20). Farmington Hills, MI: Macmillan Reference USA.

  • King, P. (2006). Anselm, St. In D. M. Borchert (Ed), Encyclopedia of philosophy, 2d ed., 1, (pp. 214–219). Farmington Hills, MI: Macmillan Reference USA.

  • Oppy, G. (1995). Ontological arguments and belief in God. Cambridge: Cambridge University Press.

    Google Scholar 

  • Plantinga, A. (1974). The nature of necessity. Oxford: Clarendon Press.

    Google Scholar 

  • Rowe, W. L. (1971). The Cosmological Argument. Noûs, 5, 49–61.

    Google Scholar 

  • Rowe, W. L. (1975). The Ontological Argument. In J. Feinberg (Ed), Reason and responsibility: Readings in some basic problems of philosophy, 3d ed., (pp. 8–17). Encino, CA: Dickenson Publishing Company. This essay appears, unchanged, in all subsequent editions.

  • Rowe, W. L. (1976a). The Ontological Argument and question-begging. International Journal for Philosophy of Religion, 7, 425–432. Rowe calls this a “condensation” of Rowe 1975. It is criticized by Davis 1976a.

    Google Scholar 

  • Rowe, W. L. (1976b). Comments on Professor Davis’ ‘Does the Ontological Argument beg the question?’ International Journal for Philosophy of Religion, 7, 443–447. This is a reply to Davis 1976a. It is criticized by Davis 1976b.

    Google Scholar 

  • Rowe, W. L. (1979). The problem of evil and some varieties of atheism. American Philosophical Quarterly, 16, 335–341.

    Google Scholar 

  • Rowe, W. L. (1988). Response to Dicker. Faith and Philosophy, 5, 203–205.

    Article  Google Scholar 

  • Rowe, W. L. (2007). Philosophy of religion: An introduction. 4th ed. Belmont, CA: Thomson Higher Education. Chapter 3 of this book, entitled “The Ontological Argument,” is virtually identical to Rowe 1975.

  • Rowe, W. L. (2009). Alvin Plantinga on the Ontological Argument. International Journal for Philosophy of Religion, 65, 87–92.

    Google Scholar 

  • Rowe, W. L. (2010). Friendly atheism revisited. International Journal for Philosophy of Religion, 68, 7–13.

    Article  Google Scholar 

  • Swinburne, R. (1993). The coherence of theism (Rev ed.). Oxford: Clarendon Press.

    Book  Google Scholar 

  • Van Inwagen, P. (2006). Ontological Argument for the existence of God (Addendum). In D. M. Borchert (Ed), Encyclopedia of philosophy, 2d ed., 7, (pp. 20–21). Farmington Hills, MI: Macmillan Reference USA.

Download references

Acknowledgments

This essay is dedicated to William L. Rowe, whose work in philosophy of religion has edified, entertained, inspired, and challenged philosophers for more than half a century. I thank two anonymous reviewers for their constructive criticisms of earlier versions of this essay. Neither is responsible for what remains. I thank the editor in chief, Ronald Hall, for his courtesy and professionalism.

Author information

Authors and Affiliations

  1. Department of Philosophy and Humanities, The University of Texas at Arlington, Box 19527, Arlington, TX, 76019-0527, USA

    Keith Burgess-Jackson

Authors
  1. Keith Burgess-Jackson
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to Keith Burgess-Jackson.

Appendix

Appendix

In his first essay on Anselm’s ontological argument, published in 1975, Rowe claimed that the argument “fails as a proof of the existence of God” (1975, p. 21). He was still making this claim as late as 2009. What exactly is the relationship between (1) a proof, (2) a sound argument, and (3) a question-begging argument? Unfortunately, Rowe never tells us what he means by “proof,” but it’s reasonably clear from other things he says that he conceives of proof in the manner of his longtime Purdue colleague Martin Curd (1992, p. 553):

A proof is a noncircular, deductively valid argument in which we know that all the premises are true. An argument is noncircular when it is possible to know that all the premises are true without first having to know that the conclusion is true.

A proof, we might say (following Curd), is a sound argument the premises of which are known independently of the conclusion. Let us isolate the elements of a proof, starting with the most general and becoming increasingly specific:

  1. 1.

    It is an argument.

  2. 2.

    It is a deductive (as opposed to an inductive) argument.

  3. 3.

    It is a valid (as opposed to an invalid) deductive argument.

  4. 4.

    It is a sound argument (meaning that, in addition to being valid, its premises are true).

  5. 5.

    Its premises are known (meaning, roughly, that they are believed and that there is good reason to believe them).

  6. 6.

    Its premises are known independently of its conclusion (meaning that the good reasons for believing the premises are not also good reasons for believing the conclusion).

As we saw in Part 5 of this essay, Rowe (1971, p. 60) has said, of the cosmological argument, that it “may” be sound, the implication being that it is valid. What he means by this, evidently, is that, for all we know, its premises are true; but since we don’t know that they’re true, the argument fails to incorporate element 5. A fortiori, it fails to incorporate element 6. Hence, the cosmological argument fails as a proof; it incorporates at most four of the six required elements.

Let us apply this analysis to Anselm’s ontological argument. Rowe concedes that the argument is valid, so it incorporates elements 1 through 3. By claiming that the argument begs the question, Rowe claims that it fails to incorporate element 6. So Rowe could concede that Anselm’s argument incorporates elements 4 and 5. In other words, he could say, as he does of the cosmological argument, that, for all we know, the premises of Anselm’s argument are true. He could even concede that these premises are known. What he wishes to deny is the further claim that the premises are known independently of the conclusion. In other words, he wishes to assert that the argument begs the question.

It appears, therefore, that Rowe has different criticisms of the two arguments, though he believes that both arguments fail as proofs.Footnote 18 The cosmological argument fails as a proof because its premises are not known. Anselm’s ontological argument fails as a proof because, while its premises are known (recall that this is a concession Rowe could make, even if he in fact doesn’t make itFootnote 19), they are not known independently of its conclusion. Anselm’s argument begs the question; the cosmological argument does not. Or so Rowe maintains.

Rights and permissions

Reprints and permissions

About this article

Cite this article

Burgess-Jackson, K. Does Anselm beg the question?. Int J Philos Relig 76, 5–18 (2014). https://doi.org/10.1007/s11153-014-9450-9

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s11153-014-9450-9

Keywords

Search

Navigation