3. Lemmas: ‘ἀγεωμέτρητος μηδεὶς εἰσίτω’1
Even in Homeric studies, it would seem that the motto in the legend inscribed above the entrance to Plato’s classroom, is apposite: ‘let no one unmeasured in geometry, enter here.’
Let us consider the Venetus A manuscript as an artefact. Gregory Nagy’s treatment of this document will serve as an object lesson in the importance of applying the new theory of the Greek accent, and its many implications, so as to avoid debilitating errors in the editing of texts and the fostering of some very bad habits of mind indeed. How ought one to look at this document? I propose that the text framed within it should be viewed as a proposition. I say this because of the use of the word ‘lemma’ to refer to the piece of text upon the which the main or ‘frame’ scholia make an annotation.2
Nagy thus glosses ‘lemma’:
… an ancient technical term referring to whatever wording is literally ‘taken’ (the corresponding Greek word is λαμβάνειν / λαβεῖν) from the overall wording of a scriptio continua that is being quoted. In the case of the Venetus A, what happens is that the wording of any given lemma is notionally being ‘taken’ out of the overall wording of a Homeric verse and then transferred into the scholia, where it serves to lead off the wording of the relevant commentary. Literally, the string of letters that is the lemma is being ‘taken’ out of the longer string of letters that is the overall verse, which had formerly been written in scriptio continua.3
His interpretation of the usage here is wrong on its face. Nagy is evidently unaware of the use of lemma as a term of art among the geometers who practiced in Alexandria. A lemma is a proposition that has to be assumed in the course of the proof of another proposition, although the lemma is not itself proved in the course of the demonstration. Often in Euclid the proofs of any required lemmas are provided after the formal demonstration is over—whether by Euclid himself or someone else. (The work of geometers is genuinely communal and anonymous, unlike the composition of original poetry.)
The use of λῆψις or ὑπόληψις as ‘supposition’ acquired a terminological authority under Aristotle. The influence of Aristotle is evident among Alexandrian scholars generally. Dionysius of Thrace, for example, defines the grammatical art (γραμματική τέχνη) as a kind of ἐμπειρία4—‘experience’, perhaps a ‘descriptive’ or ‘empirical’ art, so to distinguish it, pointedly in Aristotelian epistemology, from ἐπιστημή—the latter being the kind of knowledge best characterised by the deductive-synthetic demonstrations of geometry. It is to be expected, however, that a lemma will be supported by an argument or demonstration of some kind, even though the descriptive, comparative principles of grammar do not carry the metaphysical weight of the first principles in arithmetic and geometry. The recourse to descriptive argument is in fact characteristic of the scholia in Venetus A, as we shall see. Lemmas are typically justified by the scholiast, just as they implicitly or explicitly serve to justify the chosen reading in the main text.
So what sort of proposition does a manuscript like Venetus A make? It is that the text itself—the thing that is centred in the frame, not the apparatus located within the frame and margins attracting scholars like bees to pollen—is a genuine text of Homer’s composition.
This despite the fact that unlike the Alexandrian texts cited in the lemmas, it is written with accentual units (‘words’) separated, and with the terminal prosody for each and every such unit indicated with an economy of signs. Acutes, circumflexes and graves mark where in the pronunciation of a word the voice rose in pitch; the rest, including the barytonic nature of barytone words, literally follows. It is readily evident that economy characterises everything about the ancient scholarly approach to presenting the prosodic data in this framed text. In geometry one speaks of ‘elegance’. It is, to be sure, unfortunate that the descriptive economy of marking only the mora where the voice rose in pitch has up to now obscured the crucial role of the subsequent fall in pitch of the contonation in Greek prosody. But the system of signs in its economy is nevertheless an elegant solution.
The lemmas are snippets of text from an authoritative source, which, as Nagy points out, was itself likely written in the scriptio continua. They each constitute the basis—the authoritative assumption—upon which the relevant segment of the framed text is proposed. In the proposed text the words are separated, and the accent marks are filled in for each prosodic word—not necessarily a feature of the lemma quoted in scriptio continua—but the lemma nevertheless is invoked to inform (not dictate) the choices in these prosodic matters that are made in the proposed main text. We are not in the realm of geometrical demonstration, but we are in the realm of assumptions or givens, in relation to a proposition. In the background are sources for the scholia who have commented, among other things, on the authenticity of lines, on the correctness of accentuation, and on punctuation.
The last refers to places where the text is made to instruct a performer to pause; as we have seen, such pausing can have the effect of ‘releasing’ suppressed accents, and modern editors punctuate and re-accent according to the graphic rules. The original punctuators among Homeric scholars, one assumes, were aware, unlike some of their successors, of the aural and audible prosodic consequences of prescribed pauses in performance. It is not just a matter of changing the direction of accent marks on a page, when an editor inserts a comma.
An immediate consequence of this understanding of the use of ‘lemma’ is the realisation that most of the main proposed text does not require them. In other words, most of the correctness of the corrected text is, in the language of geometers, self-evident—not in need of a supporting lemma. Here again we meet the remarkable fact, the mystery, that the Homeric compositions have had an astounding coherence and fixity throughout their transmission. This quality is implicit in the original mandate of Solon, prior to any Pisistratean recension, that rhapsodes be required to begin where their predecessor left off. This mandate bespeaks an audience (or an elite segment of the audience) who would not tolerate the liberties a performer might take to suit an occasion or his own aesthetic impulse, excerpting a palate of ‘greatest hits’ and stitching together a memorable Homeric programme. This audience knew the original and insisted that they hear the album complete—or at the very least, in its proper order.
This is not to say, however, that the texts cited in the lemma are windows to the original composition. Their authority in Venetus A, to judge by the contents of the scholia, seems to come from earlier editions, sometimes with readings conflicting, of the seminal Alexandrian scholars. It is not the case, therefore, that the texts cited as lemmas are of necessity transparent windows to the original composition, in either diction or prosody.
Confidence in the text of Homer cannot come from the mere survival of late artefacts like the East Roman manuscripts. Something close to everything depends on the coherence of the method and philosophy of the Alexandrian sophists whose editions of Homer are quoted by the scholiasts in support. But the fact that these scholiasts appear to frame their efforts in the language of lemmas to a proposition, suggests that at the very least they are aware of their situation with respect to the historical truth, which is often something more fluid in the eyes of musicians and performers in the moment of enactment; and it does indeed inspire confidence in these scholiasts, that they know that they live in the frame, as it were, of a view beyond the window.
But most persuasive of all will be a demonstration of the internal coherence and musicality of the text they have so preserved, on its own terms, when it is allowed to speak for itself. Unlike in geometry, there will be proof in the pudding.
Originally from A. P. David, ‘Singing Homer’s Spell: The Disyllabic Contonation and the Proposition Made by East Roman Manuscripts’, Dramaturgias: Revista do Laboratório de Dramaturgia da Universidade de Brasília, 19:7 (2022), 805-66; since revised.
For an excellent analysis of the different types of scholia collated and formatted in this manuscript, see Lara Pagani, ‘The Iliad “Textscholien” in the Venetus A”, in Marco Ercoles, Lara Pagani, Filippomaria Pontani and Giuseppe Ucciardello, eds., Approaches to Greek Poetry, Berlin: De Gruyter, 2018, 83-106.
Gregory Nagy, ‘Traces of an ancient system of reading Homeric verse in the Venetus A’, in Casey Dué (ed.), Recapturing a Homeric Legacy, Cambridge MA: Center for Hellenic Studies, 2009, 135-6.
ἐμπειρία was also the term used earlier by Eratosthenes; see Lara Pagani, ‘Pioneers of Grammar. Hellenistic Scholarship and the Study of Language’, in Franco Montanari and Lara Pagani, eds., From Scholars to Scholia, Chapters in the History of Ancient Greek Scholarship, Berlin: De Gruyter, 2011, 17-18.