Though Cephalas presents it as a preface (pröoimion, 'proem'), I think Agathias must instead have written the closing poem of Book 4 of the Anthology as his sphragis -- the poetic signet-seal that came at the end of his completed book to warrant its quality and assert his authorship. It's much shorter than his actual proem, thank heavens, and I present the whole of it here in my own translation. The two emboldened bits identify translation choices that I'd like to talk about afterwards.
The slabs of stone, the tools to carve the line,Line 1's 'tools to carve the line' is my best stab at the Greek graphis. It is close kin to the Greek verb graphō, meaning to write or mark, and the noun gramma, a thing written or marked. Very many words in English derive from these -- telegraph and telegram, autograph, seismograph, graphic, graphite (and thereby graphene), and so on, down to Instagram. Here is the entry for graphis in the great Greek-English Lexicon of Liddell and Scott, revised by Jones, LSJ for short.
The tablets with inscriptions — all are cause
Of hearty cheer to those who call them ‘mine’;
But only while they live. The empty vaunts
Of men in life yield little benefit
To souls in that which follows. Nonetheless
Excellent character and wisdom’s grace
Can percolate to there and still remain
In this our world, here to perpetuate
Remembrance. So did Plato never brag
In pigments and in graven monuments,
Nor Homer, but in wisdom all alone.
Happy are they whose memory endures
In libraries of volumes subtly wrought,
And not in emblems destitute of thought.
A. [select] = γραφεῖον 1, AP6.63 (Damoch.), 65 (Paul. Sil.), 67 (Jul.): esp. stilus for writing on waxen tablets, Pl.Prt.326d; paint-brush, APl.4.178 (Antip.); graving tool, LXX Ex.32.4; “σύμβολα . . γραφίδεσσι κατέξυσα” Hymn.Is.11, cf. AP4.3b.72 (Agath.); needle for embroidering, APl.4.324.II. [select] embroidery, AP5.275 (Agath.): but in pl., = paintings, Nonn.D.25.433.
The citation I've highlighted is annoying if you're working from the old Loeb, as I generally am, since it's (a) in the public domain and (b) lovely. The reference is to Agathias' proem, 4.3, parts of which we've considered in previous blog posts. At some point an editor decided to divide this block of verse into a pair of poems: 4.3a (lines 1-46), and 4.3b (everything that follows). There is obvious sense in this. Not only does the topic change, but the metre as well, from iambics to epic hexameter, apt for extolling the heroic conquests of Agathias' emperor. But it means you have to do sums if you are using an older edition and want to make sense of the citation, which in this case refers to the dative plural graphidessi in what used to be line 118.
I think LSJ is actually wrong about this instance. As I understand that line, Agathias is contrasting the use of a graphis to compose inscriptional or quasi-inscriptional epigrams, with the physical carving of such poems onto actual monuments. As my translation had it,
... Part the Second, though,
Collects the antique votive offering:
All that we graved with pens or had inscribed
Out in the world, on well-wrought statue’s base
Or on the many far-flung monuments
That witness to the breadth of human art.
'Percolate' in my line 8 is my playful best guess at what to do with the compound verb sunerpei in line 5 of the Greek. Herpō by itself means to move by slow gradations, as might the very young or very old. If I had to settle for a single English rendering, it would have to be 'creep'. Its close cousin in Latin is serpo, from which we get 'serpent', and our 'herpetology' is the study of creeping things (reptiles). But what is sun-erpō? It has no dictionary entry of its own, and a TLG search turns up just one other instance in the whole of Greek literature, in the philosopher Epictetus. There, it just means to 'herp' in another's company.
On the one hand, that's not terribly helpful; on the other, it frees me up to go for something quirky, as surely was Agathias' original word choice in the poet's own eyes.