Did he bring his crocodile?

Gifford calls this a “famous conundrum”, but offers no origin for it nor any indication of where and when it is or was “famous.”¹ Other annotators (e.g., Anderson, Johnson, Riquelme) have simply neglected the passage, instead eagerly identifying the Shakespeare quotation in the next line.

Joyce’s source for this might be the second-century Greek satirist Lucian. Richard Ellmann casually ascribes to Joyce a familiarity with Lucian and notes that Joyce’s Triestine library included an 1885 Italian translation of Lucian’s works.² In Lucian’s Philosophies for Sale (Βίων Πρ σις, also translated as Sale of Creeds and known in Latin as Vitarum auctio), Zeus and Hermes oversee a sale of various philosophers, including Socrates, Aristotle, Democritus, Heraclitus, Epicurus, and so on, allowing each to give an account of his teachings and value as a commodity. Representing the Stoics in this satire is Chrysippus (ca. 280-207 BC).³ Diogenes Laertius credits him with over seven hundred books, among these, according to another work by Lucian, Icaromenippus (καρομένιππος), a book of Syllogisms. In Philosophies for Sale, the Cynic offers an example of his syllogistic prowess (here in the 1905 translation by H. W. Fowler and F. G. Fowler):⁴

Chrys. Take a case. You have a child?

Seventh [Dealer]. Well, and what if I have?

Chrys. A crocodile catches him as he wanders along the bank of a river, and promises to restore him to you, if you will first guess correctly whether he means to restore him or not. Which are you going to say?

Seventh D. A difficult question. I don’t know which way I should get him back soonest. In Heaven’s name, answer for me, and save the child before he is eaten up.

With the exception of Marc A. Mamigonian and John Turner, who have likewise pointed to this text of Lucian's, Portrait annotations have followed Gifford’s lead in suggesting that the winning answer is to say that the crocodile intends to eat the child; by logic the crocodile cannot then eat the child without having proved the parent’s guess correct. Classical commentators, on the other hand, enjoy no such consensus.⁵ Francis Greenleaf Allison, for example, writes:⁶

If the mother says "You will not," he gives up the child, but as her words are false the child is lost; but if she says "You will," the crocodile cries “False!” and devours the child. No solution for the sophism! The humane grammarians, however, advise the mother to give the first answer, get temporary possession of the child, and make off with it. There were other such [puzzles] in the common stock.

The “logical” answer (saying that the crocodile means to eat the child) depends on two questionable assumptions. The first is that the crocodile is honest. The specificity of the animal is important – it’s not a lion or a wolf – for the notion of “crocodile tears” dates back at least as far as Plutarch, whose readers knew the expression, and Lepidus refers to those tears in the same passage from Antony and Cleopatra that Joyce quotes.⁷ The second assumption is that what the crocodile will do and what he means to do are the same thing: the intentional fallacy. This dimension of the problem might well have been of Lucian’s making, and it is interesting that Cranly seems to tell the traditional Stoic version (“what he was going to do”) rather than the ironic, mocking variation from Lucian (“whether he means to”).

The Stoics in fact called this riddle the krokodeilites (κρoκoδειλίτης), by definition a kind of aporia or unsolvable problem. The aesthetic questions Stephen recounts to Lynch in A Portrait and the riddle of the fox in “Nestor” might be read as kindred problems, “verbal involutions” or “word-snares” as different translators have rendered Lucian’s characterization of Stoic philosophy.⁸ Joyce’s introduction of the Nile into the question smooths the way for his subsequent Antony and Cleopatra allusion and connects with the prospect of “terrible queer creatures at the latter end of the world” (P 212), confirmed by the worldly sailor Murphy’s example of a crocodile as one of the “queer things” he has seen (16.465-66), and of course that “harping on the mother” theme is significant, not least because in Stephen’s version of the krokodeilites, the mother is implied to be to blame for the child’s misfortune (she “let her child fall”).

There are a few (frustrating but instructive) reasons for the necessary caution in saying that Lucian “might be” the author who provided Joyce with this riddle. For one thing, the volume of selected works by Lucian included in Joyce’s Trieste library, referred to above, does not include Philosophies for Sale.⁹ Another reason is that variations on any lost “original” Stoic puzzle make the positive identification of a fixed “source” a significant problem. A number of nineteenth-century renderings appear in the popular press, such as this one from Frank Leslie’s Pleasant Hours (1867):¹⁰

An infant, while playing on the bank of a river, was seized by a crocodile. The mother, hearing its cries, rushed to its assistance and by her fearful entreaties obtained a promise from the crocodile (who was obviously of the highest intelligence) that he would give it her back if she would tell him truly what would happen to it. On this the mother (perhaps rashly) asserted:

"You will not give it back."

The crocodile answers to this: "If you have spoken truly, I cannot give back the child without destroying the truth of your assertion; if you have spoken falsely, I cannot give back the child because you have not fulfilled the agreement; therefore, I cannot give it back whether you have spoken truly or falsely."

The mother retorted: "If I have spoken truly, you must give back the child, by virtue of your agreement; if I have spoken falsely, that can only be when you have given back the child; so that whether I have spoken truly or falsely, the child must be given back."

History is silent as to the issue of this remarkable dispute.

As unlikely as it is that Joyce read this New York periodical, this version does include the character of the mother, perhaps itself a modern addition.

The conspicuously clipped syntax of Stephen’s paraphrase of Cranly’s apparently winning version probably demonstrates some impatience and more jealousy, and we cannot disallow the possibility that the “source” for this borrowing is not textual at all but perhaps a performance by J. F. Byrne, the model for Cranly, as observed by Joyce. The crocodile problem that we read in A Portrait is thus a muddy problem for annotation and attribution. Is Cranly’s version of the krokodeilites drawn from Lucian, or some other report of Stoic problems of logic? Is Joyce’s version a version of Lucian’s version, distorted or truncated by a clearly nettled Stephen and thus eliding the question of intention? This crocodile is not easily tamed.

Tim Conley

Brock University

Search by keyword (within this site): Greece Logic Puzzles Crocodile Trieste Library