Aristotle’s is well-known, the first sentence of the Metaphysics:

pántes hoi ánthropoi toû eidénai orégontai phúsei.

All humans by nature desire knowledge.

Plato’s is less well-known, being tucked away in a complex argument in Book VII of the Republic (535e), where Socrates describes:

anáperon psukhén . . . hè àn tò mèn hekoúsion pseûdos misêi kaì khalepôs phérei auté te kaì hetéron pseudoménon huperaganaktêi, tò d’ akoúsion eukólos prosdékhetai kaì amathaínousá pou haliskoméne mè aganaktêi, all’ eukherôs hósper theríon húeion en amathíai molúnetai.

. . . the lame soul which hates the voluntary falsehood and not only cannot bear to lie itself but is greatly angered when others lie, yet cheerfully accepts the involuntary falsehood and is not distressed when caught in ignorance of something, but wallows in ignorance like a brutal hog.

In the thirty years since I first heard of them, I’ve had the vague impression that siculate lunate sigmas, like adscript iotas for the traditional subscripts and use of capital V and small u for both vowels and consonants in Latin texts, were the coming thing, and that more and more editions were using them. When I went to look for some examples a few nights ago to show my students, I was surprised how difficult it was to find any. Of a dozen or two texts from the last half-century, a mixture of Oxford, Teubner, and Budé editions, only Sandbach’s Oxford Classical Text of Menander uses them. Have they gone back out of style? Is this a fad convention that never really caught on? Or was my selection unrepresentative? If so, can someone give some other examples? I don’t doubt that siculate sigmas are abundant in the pages of ZPE and the Oxyrhynchus Papyri and similar collections, but they are far rarer in standard reading texts than I had thought.

Finding one text was easy, even if a second has proved elusive. I began my sigma-hunting with Menander because of something that happened to me in graduate school. A neighbor in my apartment building was a third-world immigrant who had gone to the best prep school in his home country, where he had taken (I think it was) 8 years of Latin and 4 of Greek. Though studying other subjects in the U.S., he had decided to brush up both languages, and sent away for the Oxford texts of Vergil and Menander. When they arrived, he consulted me in a panic, wondering if he should send them both back as defective, since the Menander had a lot of Latin Cs where he expected Greek sigmas, and the Vergil had vowels for consonants and consonants for vowels. Vrbs antiqua fuit on the first page of the Aeneid particularly distressed him: “Verbs ahn-tee-kwa foo-it ? Verbs ? What is this Verbs ?”

Update: (10/2, 8:15am)

I’ve changed ’siculate’ to ‘lunate’ above, since it seems to be more familiar. In fact, I’m starting to wonder whether I remembered it wrong, since ’siculate’, “sickle-shaped”, might well describe the traditional end-of-word small sigma. The few dictionaries at hand don’t seem to recognize the usage.

As for examples, the only other texts I’ve found so far that use lunate sigmas are Diggle’s Oxford Classical Text of the Tragicorum Graecorum Fragmenta Selecta (1998) and his Cambridge editions of Euripides’ Phaethon (1970) and Theophrastus’ Characters (2004). There are certainly plenty of texts of authors surviving only (or primarily) in papyri that do not use them, e.g. Cunningham’s Teubner Herondas. Time for a more thorough check? I’ve just about finished unpacking my books, so it wouldn’t be too strenuous.

Some purely verbal jokes work equally well in many languages. Here is a paragraph of Chekhov’s one-page squib, “From a Retired Teacher’s Notebook”:

The words ‘proposition’ and ‘conjunction’ make schoolgirls modestly lower their eyes and blush, but the terms ‘organic’ and ‘copulative’ enable schoolboys to face the future hopefully.

(The Oxford Chekhov, tr. Ronald Hingley, Volume VI, Stories, 1892-1893, p. 260)

Here is the next paragraph, the only other one (of six) that struck me as particularly interesting:

As the vocative case and certain rare letters of the Russian alphabet are practically obsolete, teachers of Russian should in all fairness have their salaries reduced, inasmuch as this decline in cases and letters has reduced their work load.

Now you don’t have to read the whole story.

I wonder if any Latin teachers in the Middle Ages thought to ask for a pay raise when W and J and the distinction between U and V were added to the original 23-letter Latin alphabet to make the modern English set of 26 — not that that all happened at once. Actually, the development of the English alphabet was a bit more complicated than simple accretion, since the two th’s, eth (ð) and thorn (þ), were added at some point and later subtracted, though they survive in Icelandic.

Silius

Update: (9/5, 4:15pm)

Since no one has ‘gotten’ it yet, here’s another version of the joke with the same answer:

Baebius

And here are two more, non-Classical this time, with a different, but parallel, answer:

Philip
Charles