In these well-known lines Horace describes his three books of Odes in metaphorical terms as a monumentum or ‘grave-monument’.Footnote ¹ The description is well chosen because, in addition to its meanings of ‘grave-monument’ or ‘memorial’, monumentum is frequently used tout court to refer to works of literature.Footnote ² This duality is conveniently illustrated by the very first line. Since in the eyes of the Romans bronze typified durability (cf. Plin. HN 34.99 usus aeris ad perpetuitatem monumentorum iam pridem tralatus est tabulis aereis, in quibus publicae constitutiones inciduntur),Footnote ³ ‘more enduring than bronze’ makes good sense twice over: epitaphs sometimes claim that a grave-monument will last for ever (e.g. CLE 467.1 Aeternam tibi sedem … dicaui),Footnote ⁴ while Catullus had used the very word perennis when praying for the durability of his poetry (1.10 plus uno maneat perenne saeclo). Thus aere perennius is appropriate both to the metaphor and to the Odes; it functions both in terms of the comparison and as a claim to poetic immortality.

Horace's second line, however, is not so straightforward. Grave-monuments frequently adopted or incorporated the shape of a pyramid: a decade after the publication of the Odes, for example, a substantial pyramid, faced in marble, was constructed at Rome for the former praetor C. Cestius,Footnote ⁵ while at Ostia there is a miniature pyramidal tomb for one C. Annaeus Atticus.Footnote ⁶ But regali at the start of line 2 indicates that Horace is thinking not of domestic examples but of the famous pyramids in Egypt, the resting places of the pharaohs (Strabo 17.1.33 τάφοι τῶν βασιλέων). Since the pyramids were famous above all for their height (e.g. Prop. 3.2.19 pyramidum sumptus ad sidera ducti, Plin. HN 36.75–82, Tac. Ann. 2.61.1 instar montium eductae pyramides),Footnote ⁷ a reference to the superior height (altius) of Horace's monumentum seems at first sight natural enough; yet the text does not read ‘higher than the pyramids’ but ‘higher than the royal situs of the pyramids’, and the word situs has caused a great deal of scholarly trouble.

Some think that situ means ‘decay’ (OLD s.v. situs ³) and that Horace is alluding to Simonides’ famous poem on the dead at Thermopylae (531.4–5/261P), in which a burial (ἐντάφιον), because metaphorical, will not be destroyed by physical decay (εὐρώς):Footnote ⁸

But it is no compliment to say that a monument is higher than a decayed ruin. One might attempt to get round this difficulty by adopting the interpretation favoured by D. West in his commentary, who says that regali situ pyramidum is an example of the ‘genetiuus inuersus’, as it is sometimes called: ‘“the royal decay of Pyramids” is “the decaying Pyramids of kings”’.Footnote ⁹ Yet, while it is self-evidently true that all monuments are inherently subject to decay (cf. Prop. 3.2.22), the pyramids were known to be constructed of a hard stone of ‘everlasting durability’ (Diod. 1.63.5 διαμονὴν αἰώνιον) and to be ‘undecayed’ in Horace's day (ἄσηπτον): it would be very odd to draw attention to a feature that was conspicuously absent. Alternatively we might think of understanding situ in a proleptic sense (‘higher than the royal pyramids which one day will themselves decay’), but such a meaning is extremely difficult to extract from the Latin. Besides, any interpretation along these lines destroys the parallel with line 1, where the whole point of monumentum aere perennius is that bronze resists decay.

Others think that situ must mean ‘site’ (OLD s.v. situs ² 1).Footnote ¹⁰ A reference to ‘the royal site of the pyramids’ would make excellent sense in itself: not only were the Giza pyramids, for example, sited with impressive precision according to astronomical principles,Footnote ¹¹ but well-to-do Romans took an intense interest in the arrangements for their resting place and its layout. A lengthy inscription, thought to date from Trajan's reign, records the instructions left by an anonymous individual from the region around modern Langres in France (ILS 8379): ‘I want it completed according to the plan which I have provided, in such a way that there is a recess in the place where a seated statue, no less than 5 feet in height, can be placed, either of marble from the best possible overseas stone or of bronze …’ The monument itself was to be situated in an orchard and alongside a lake.Footnote ¹² Scholars have drawn attention to the similarity between this inscription and the instructions which are issued by Trimalchio in Petronius: ‘Are you building my monumentum exactly as I have ordered you?’, he begins (71.5), before saying that it must have a frontage of 100 feet and must stretch back 200 feet, thereby incorporating a memorial garden.Footnote ¹³ Yet, although it is therefore attractive to draw attention to the site which the pyramids occupied, it makes little (if any) sense to say ‘higher than the royal site of the pyramids’: the point about the pyramids is not that they are sited on conspicuously high ground but that they themselves are famously high:Footnote ¹⁴ the Great Pyramid at Giza is in fact two-and-a-half times the height of the plateau on which it stands. In other words, no matter whether situ is interpreted to mean ‘decay’ or ‘site’, its construction with altius is problematic.Footnote ¹⁵

And this is not the only problem with altius. Whereas perennius was equally applicable both to a grave-monument and to Horace's poetry, as we saw, in what sense can height (altius) be ascribed to the Odes?Footnote ¹⁶ It might be thought that altus is the Latin equivalent of ὑψηλός and that Horace is claiming for his Odes the ‘sublimity’ (ὕψος) discussed by Longinus.Footnote ¹⁷ But the equivalence of the two terms is denied by Brink, who adds correctly that ‘altus in Latin critical terminology is surprisingly rare’.Footnote ¹⁸ There are a handful of examples of altus applied to ‘style’,Footnote ¹⁹ one of which is Seneca, Agam. 332–3 modis … altis; but, since Tarrant explains this as a reference to the hexameter and explicitly draws a contrast with lyric, this line of approach seems unpromising.Footnote ²⁰ There are more numerous examples of altus as applied to ‘thought, studies, enterprises, qualities, etc.’ in the senses of ‘elevated, noble, lofty’;Footnote ²¹ but this seems scarcely an appropriate description for a collection which includes repeated invitations to sex and drink and which the poet himself elsewhere characterises as ‘playthings’, the product of a ‘light-hearted lyre’ (Hor. Carm. 3.3.69 iocosae … lyrae; cf. Epist. 1.1.10 ludicra).Footnote ²² altius, in short, is no more applicable to the Odes than it is to the metaphorical monumentum.

Although the manuscripts at Lucr. 6.357 are unanimous in transmitting alta, the poet's evident allusion to an Ennian phrase (Ann. 27, 145, 348) means that Turnebus was almost certainly right to emend alta to apta. At Aen. 1.429 the manuscripts again transmit alta, but, since the same word appears in the same sedes two lines earlier, one of them is likely to be an error; Bentley emended the second alta to apta, which is printed by Mynors and Austin.Footnote ²³ That the one adjective can be written mistakenly for the other is shown by the fact that they are variant readings at Prop. 3.22.42 and several times in Ovid (Am. 3.1.14, Fast. 2.216, Ibis 212). Is it possible that similar considerations apply in the case of Horace's ode and that he wrote aptius rather than altius? It is generally agreed that, in addition to an allusion to Simonides, Horace is here alluding also to Pindar, Pyth. 6.5–14:

Both poets describe their poetry in terms of a building metaphor, and in both there are a relative clause, double negative, and the pairing of rain and wind. Since Horace's allusion is so clear, it is obviously relevant to note that Pindar describes his ‘treasure house’ as ἑτοῖμος, with which aptus overlaps in meaning. But, whereas in Pindar ἑτοῖμος is constructed with a dative (‘ready for ~’), a construction to which aptus also commonly lends itself (as at Val. Max. 4.8. ext. 2 publicis usibus apta monumenta exstruebantur), the meaning of the adjective in the ode would be ‘better adapted to its purpose, more appropriate, more apt’ (i.e. as a memorial).Footnote ²⁴ Moreover aptus, unlike altus, is frequently used in literary criticism and has a wide variety of applications: in particular it is one of the terms used to express the key literary virtue of τὸ πρέπον (‘appropriateness’).Footnote ²⁵

Although ‘more apt than the royal site of the pyramids’ may seem rather a come-down to readers used to finding ‘higher’ or ‘loftier’ in their texts,Footnote ²⁶ this is perhaps to underestimate the significance of the words regali … situ. The necropolis at Giza covered almost a square mile in area and displayed the three vast pyramids for which it was famous (cf. e.g. Strabo 17.1.33, Plin. HN 36.76, Mela 1.55).Footnote ²⁷ It is beyond question that the site of these pyramids was the ultimate lieu de mémoire in the ancient world, but – the ultimate paradox –there was no agreement about which kings the pyramids commemorated, ‘since by the most just of fates the authors of such uanitas have been forgotten’ (Plin. HN 36.79 inter omnes non constat a quibus factae sint, iustissimo casu obliteratis tantae uanitatis auctoribus). uanitas means both ‘foolish pride’ and ‘futility, pointlessness’:Footnote ²⁸ Horace's monumentum is thus even more appropriate than the world's grandest and most famous site of memory, since, despite the challenge constituted by the omission of his name from his ‘epitaph’,Footnote ²⁹ the three books of Odes will achieve their purpose in ensuring that his fame lives on for ever.