Reference edition:

D’Alessandro, P.  2004.  Rufini Antiochensis. Commentaria in metra Terentiana et de compositione et de numeris oratorum.  Edizione critica.  Hildesheim-Zurich-New York.

This brief commentarium, which collects citations claiming that the prose discourse of orators utilizes clausulae, is found in the manuscripts following the author’s In metra Terentiana.  The division between the texts is marked by a passage that is more properly a subscriptio to the preceding work, rather than a dedicatory inscriptio to the De compositione et metris oratorum (thus Kaster 1988: 351, n. 130). This section on metrical clausulae is introduced by sixty-eight hexameters by Rufinus himself; these nevertheless appear in distinct groups, most likely through the intervention of a subsequent excerptor who cites them restating the work’s complete title (Versus Rufini v.c. litteratoris de compositione et de metris oratorum).  From this we can observe that the conjectural title assigned to the anonymous commentarium is arbitrary.  We may speculate that the verses derive from another work of Rufinus on the same theme or, more likely, that the author followed the practice—then fashionable—of pairing his treatise in prose with a version in meter (this is also attested in the technical-scientific and artigraphical works of late antiquity).

Cicero as a rhetor represents the chief auctoritas in the work (from whom is also derived the list of Greek authors who wrote on this theme); after him come Quintilian, Charisius, Diomedes, and others, cited without regard for their relative chronology or diverse interests, with the classical auctor and the late grammaticus being placed on the same level.  Standing out among these last names is the otherwise unknown Pompeius Messalinus, author of a similar work De numeris et pedibus oratorum, from whom Rufinus transcribes a peculiar metrical analysis of the incipit of Sallust’s Histories (frg. 1 La Penna = p. 34.22 d’A.). The conditions under which the treatise comes down to us, most likely through the work of an inept excerptor, help to explain the many oddities and uncertainties littering the text; despite this, the information and citations it reports are no doubt good.  It happens, for instance, to preserve what has been judged to be a fragment of the De re publica (on which see most recently d’Alessandro 2002). Rufinus thus remains a precious and useful source.

Rufinus’ text is transmitted by 28 manuscripts, of which at least seven can be dated to the 9^th c.; these also contain the Institutiones of Priscian and/or his three short works Ad Symmachum that treat similar themes (de figuris numerum, de arte metrica, and de rhetorica).  Seven other mss., meanwhile, are codices descripti.  Halm’s partial and “provisional” 1863 edition (treating only the second part of the work = 575-85: Versus Rufini V. C. litteratoris de compositione et de metris oratorum) was based on three codices: Paris. Lat. 7496 and 7501 (collated for him by Keil), and Monac. Lat. 18375.  Keil produced a complete edition in 1874 (GL 6: 547-78) that was based on four testimonia: the two BnF manuscripts mentioned above, as well as Paris. Lat. 7498 and Vat. Reg. Lat. 733.  In contrast with these, Paolo d’Alessandro’s edition in the CGL is based on a complete review of the manuscript tradition.  He identifies two subarchetypes, α and β.  The first of these is represented directly by four older testimonia and two later ones; the other, which better preserves traces of Greek passages, is divided between Paris. Lat. 7501 (9^th c.) and the majority of manuscripts, collected as family γ, from which the humanistic editions derive.  One text stands apart from the two subarchetypes: ms. Einsiedlensis 339 (2^nd half of the 9^th c.); this is considered the testimony closest to the archetype, and contains lectiones singulares.  A different manuscript—Palat. Lat. 1741 (probably 15^th c.)—includes some corrected readings and a different recension that may also go back to an older codex that is no longer extant.  The archetype must have contained various errors, some of which were corrected in the humanistic period.  There are at least three loci desperati, one of which has invited numerous interventions by philologists (p. 20.3 d’A. = 565 K.: †sacerdos qui et domatus†); this is located in the final list of auctores who argued dramatic works were written in verse, with a majority of scholars inserting the names Plotius Sacerdos and Aelius Donatus into the corrupted text.

Rufinus’ work was published for the first time in 1470, in the editio princeps of Priscian that Vindelino da Spira produced at Venice.  [A. Lagioia; tr. C. L. Caterine].