Gaunilo’s Island

Gaunilo’s famous objec­tion to Anselm’s Onto­log­i­cal argu­ment is known as ‘Gaunilo’s Island,’ it fol­lows as such from his On Behalf of the Fool:

For exam­ple: it is said that some­where in the ocean is an island, which, because of the dif­fi­culty, or rather the impos­si­bil­ity, of dis­cov­er­ing what does not exist, is called the lost island. And they say that this island has an ines­timable wealth of all man­ner of riches and del­i­ca­cies in greater abun­dance than is told of the Islands of the Blest; and that hav­ing no owner or inhab­i­tant, it is more excel­lent than all other coun­tries, which are inhab­ited by mankind, in the abun­dance with which it is stored.

Now if some one should tell me that there is such an island, I should eas­ily under­stand his words, in which there is no dif­fi­culty. But sup­pose that he went on to say, as if by a log­i­cal infer­ence: “You can no longer doubt that this island which is more excel­lent than all lands exists some­where, since you have no doubt that it is in your under­stand­ing. And since it is more excel­lent not to be in the under­stand­ing alone, but to exist both in the under­stand­ing and in real­ity, for this rea­son it must exist. For if it does not exist, any land which really exists will be more excel­lent than it; and so the island already under­stood by you to be more excel­lent will not be more excellent”

If a man should try to prove to me by such rea­son­ing that this island truly exists, and that its exis­tence should no longer be doubted, either I should believe that he was jest­ing, or I know not which I ought to regard as the greater fool: myself, sup­pos­ing that I should allow this proof; or him, if he should sup­pose that he had estab­lished with any cer­tainty the exis­tence of this island.¹

There have been a mul­ti­plic­ity of responses in this fash­ion, attempt­ing to prove the exis­tence of uni­corns, lep­rechauns, fly­ing spaghetti mon­sters or “a high­est pos­si­ble mount­ing, a great­est pos­si­ble line­backer, a mean­est pos­si­ble man and the like”.² These argu­ments, regard­less of what they set out to prove, fol­low the same basic pat­tern: I can think of it, exis­tence is greater than nonex­is­tence, there­fore it exists. This, how­ever, is a mis­un­der­stand­ing of Anselm’s argu­ment, “Anselm is not with­out a reply. He points out, first, that Gau­nilo mis­quotes him. What is under con­sid­er­a­tion is not a being that is in fact greater than any other, but one such that a greater can­not be con­ceived; a being than which it’s not pos­si­ble that there be a greater. Gau­nilo seems to over­look this. And thus his famous lost island argu­ment isn’t strictly par­al­lel to Anselm’s argu­ment; his con­clu­sion should be only that there is an island such that no other island is greater than it–which, if there are any islands at all, is a fairly innocu­ous con­clu­sion”.³

Fur­ther­more, Gaunilo’s Island is not anal­o­gous to Anselm’s Onto­log­i­cal proof in that the qual­i­ties which make for great­ness in islands hold no intrin­sic max­i­mum. “The idea of an island than which it’s not pos­si­ble that there be a greater is like the idea of a nat­ural num­ber than which it’s not pos­si­ble that there be a greater, or the idea of a line which none is more crooked is pos­si­ble. There nei­ther is nor there could be a great­est pos­si­ble nat­ural num­ber; indeed, there isn’t a great­est actual num­ber, let alone a great­est pos­si­ble. And the same goes for islands. No mat­ter how great an island is, no mat­ter how many Nubian maid­ens and danc­ing girls adorn it, there could always be greater […] There is no degree of pro­duc­tiv­ity or num­ber of danc­ing girls such that it is impos­si­ble that an island dis­play more of that qual­ity […] The idea of a great­est pos­si­ble island is an incon­sis­tent or inco­her­ent idea; it’s not pos­si­ble that there be such a thing”.⁴

Fol­low­ing this cri­tique of Gaunilo’s island is Philoso­pher William L. Rowe who, in his William L. Rowe on Phi­los­o­phy of Reli­gion: Selected Writ­ings, has replied to Gaunilo:

A […] dif­fi­culty in apply­ing Anselm’s rea­son­ing to Gaunilo’s island is that we must accept the premise that Gaunilo’s island is a pos­si­ble thing. But this seems to require us to believe that some finite, lim­ited thing (an island) might have unlim­ited per­fec­tions. It is not at all clear that this is pos­si­ble. Try to think, for exam­ple, of a hockey player than which none greater is pos­si­ble. How fast would he have to skate? How many goals would he have to score in a game? How fast would he have to shoot the puck? Could he ever fall down, be checked, or receive a penalty? Although the phrase, “the hockey player than which none greater is pos­si­ble,” seems mean­ing­ful, as soon as we try to get a clear idea of what such a being would be like we dis­cover that we can’t form a coher­ent idea of it at all. For we are being invited to think of some lim­ited, finite thing–a hockey player or an island–and then to think of it as exhibit­ing unlim­ited, infi­nite per­fec­tions. )⁵

To quickly note, Gaunilo”s objec­tion dif­fers from Anselm in the sense that the qual­i­ties Anselm speaks of do have intrin­sic max­i­mums (omni­science, omnipo­tence, omnipres­ence) so whereas one could keep pos­tu­lat­ing a bet­ter and bet­ter island, one could not pos­tu­late a more omni­science God than God — omni­science is the maximum.

In reply to Planti­nga above (as well as Rowe, though not explic­itly), philoso­pher Nick Everitt commented:

But this objec­tion to Gau­nilo seems less than com­pelling. Given an island with a cer­tain degree of F, where F is some desir­able fea­ture, there is no rea­son to accept that an island with twice as much would be any bet­ter. You can have too much of a good thing — and Plantinga’s own exam­ples neatly illus­trate this. Even if we grant that an abun­dance of coconuts and redou­bling repeat­edly the num­ber of coconuts would keep improv­ing the island, clearly there would come a point where the super­abun­dance of coconuts became a pos­i­tive nui­sance. The same point surely goes for palm trees — and pre­sum­ably at some point even for Nubian maid­ens and danc­ing girls⁶

While I would acknowl­edge the slight chal­lenge intrin­sic maximum’s and qual­i­ties pose to Anselms proof, I would not con­sider Everitt’s response to Planti­nga to be very com­pelling. It seems to me to be much like Rowe’s hockey player. I don’t believe Gaunilo’s refu­ta­tion of Anselm’s nec­es­sar­ily requires a greater quan­tity of a thing to be con­sid­ered ‘than which noth­ing greater can be con­ceived’. Applied to God this would be inco­her­ent (a god with lesser qual­i­ties is not desir­able to a god with greater qual­i­ties), applied to a thing (an island) this seems to me per­fectly coher­ent. An island of P pro­por­tions might only accom­mo­date, as best it can, Q qual­i­ties before there is ‘too much of a good thing’ — thus an island with less is more desir­able than an island with more (rel­a­tive to P and Q; why are we assum­ing desir­abil­ity?). This also does not remove the pos­si­bil­ity (as noted by Rowe) that as a thing an unsur­pass­able island might be sur­passed by a dif­fer­ent thing (a man, for instance). Thus show­ing the inco­herency that results from apply­ing Anselm’s proof to any­thing but God and only fur­ther demon­strat­ing the insuf­fi­ciency of Guanilo’s objec­tion to Anselm and Everitt’s objec­tion to Plantinga.

Thus, though not with­out it’s prob­lems Anselms proof seems to me to be a coher­ent and valid proof for the exis­tence of God. Gaunilo’s island fails as a dis­proof and Everitt, sid­ing with Gau­nilo against Planti­nga and Rowe fur­ther fails in pro­vid­ing a con­vinc­ing refu­ta­tion to Plantinga’s defense of Anselm. That will be that for Gaunilo’s island.

Per­haps I should next con­sider Kant and Hume’s objec­tion to Anselms proof?

Related posts:

1. And you thought Anselm was absurd?