In the whole long history of ancient Rome there is no more significant date than 13 January 27 BC. On that day, after fifteen years of civil war and emergency powers, Julius Caesar's adopted son restored to the Roman people the republic they had established 480 years earlier,Footnote 1 and unwittingly inaugurated the dynastic monarchy that would govern them for the next 504 years. He himself recorded the event in his bilingual statement of achievements (Res gestae 34.2):Footnote 2
For this service, I was named Augustus by senatorial decree, and the doorposts [Greek: forecourt] of my house were publicly clothed with laurels, and a civic crown [Greek: the oak crown given to me for saving citizens] was fastened above my doorway [Greek: porch].
In the next paragraph (35.1) he recorded another prized honour, which was granted on 5 February 2 BC: ‘The Senate and the equestrian order and the entire Roman people named me pater patriae, and voted that this be inscribed in the forecourt of my house.’Footnote 3
His house had featured in a comparably symbolic event on 28 April 12 BC, as two contemporary sources report. First, the public calendar (Fasti Praenestini, Degrassi, Reference Degrassi1963: 132–3):
Holiday by decree of the Senate because on that day the statue[?] and altar[?] of Vesta was dedicated in the house of Imperator Caesar Augustus, the pontifex maximus, in the consulship of Quirinius and Valgius.Footnote 4
And second, the ‘April’ book of Ovid's calendar poem (Fasti 4.949–54):
Take the day, Vesta! Vesta has been received at her kinsman's threshold, as the just Fathers have decreed. Phoebus has one part, a second has gone to Vesta, he himself as the third occupies what is left from them. Stand, you Palatine laurels! May the house stand, wreathed with oak! One [house] holds three eternal gods.Footnote 5
Ovid's phrase ‘at the threshold’ (limine) fits precisely with Augustus’ own references to his forecourt (uestibulum).Footnote 6 It was a location full of powerful symbolism, represented on coin-types and monumental relief sculpture.Footnote 7 Where was it situated?
1. ARCHAEOLOGICAL INFERENCE
Ovid's reference to Phoebus Apollo gives an approximate position. The temple of Palatine Apollo, promised to the Roman people in 36 BC, was dedicated by the future Augustus on 9 October 28 BC.Footnote 8 Its site is known: the concrete core of the podium it stood on was excavated by Pietro Rosa in 1865,Footnote 9 and the whole area was intensively explored by Gianfilippo Carettoni from 1958 to 1984.Footnote 10
Carettoni discovered the surviving rooms of a late-Republican house immediately adjacent to the temple site at a lower level on the northwest side (Fig. 1). He identified it as the house of Augustus, and that is how it is presented to visitors today.Footnote 11 Although Carettoni never gave his reasons for the identification,Footnote 12 he probably based it on the passage in Suetonius where Augustus’ most conspicuous public works are listed (Suet., Aug. 29.3):
He erected the temple of Apollo on the part of his Palatine domus that had been struck by lightning, which the haruspices had declared was desired by the god.Footnote 13
But since Suetonius also says that Augustus’ house was a modest one (Aug. 72.2),Footnote 14 it is obvious that one part of it could not have contained a grand temple over 26 m wide and over 48 m long (not counting frontal steps).Footnote 15
Fig. 1. The supposed ‘house of Augustus’: drawn by Seán Goddard after Carettoni, Reference Carettoni1969: 24. The site is terraced on two levels, with a vertical difference of 9 m; it is not clear what, if any, connection there was between them. The upper level was evidently entered from the street by the ‘house of Livia’, and the entrance to the lower level, from the narrow street leading down the hill (‘Scalae Caci’), is at the bottom left-hand corner.
Dio Cassius’ account of the event explains what happened (49.15.5, 36 BC): ‘he made public the place on the Palatine which he had bought in order to build something, and consecrated it to Apollo after a thunderbolt had fallen on it.’Footnote 16 Suetonius was evidently using domus in the extended sense of ‘a property consisting of multiple adjacent buildings’,Footnote 17 meaning in this case the ‘several houses’ around his own that had been bought up before 36 BC by people acting on his behalf.Footnote 18 However, it is easy to see how a casual reading of Suetonius could be taken as proof that the temple was literally part of the house Augustus lived in.
By the 1970s, Carettoni had established that the house was much more extensive and imposing than had first been thought, featuring two matching colonnaded courts and a frontage of about 140 m facing out over the valley of the Circus Maximus (Tomei, Reference Tomei2014: 241–58). The importance of Carettoni's identification was emphasized by Paul Zanker (Reference Zanker1983: 23–4; Reference Zanker1988: 51–2), who referred to it as ‘Haus und Tempel als Einheit’ and saw it as analogous to a Hellenistic palace. Similarly, Pierre Gros (Reference Gros and Steinby1993: 57) declared that ‘le sanctuaire apollinien du Palatin … reprenait sous une forme nouvelle le schéma des palais-sanctuaires hellénistiques’.
This new historical understanding, presented not as a hypothesis but as a certainty,Footnote 19 was based on one specific feature of the site: a ramp that was believed to have run directly from the western peristyle of the house to the terrace in front of the temple itself.Footnote 20 Although it later became clear that that was not the case (Iacopi and Tedone, Reference Iacopi and Tedone2006: 366–7), for more than twenty years it seemed to offer a profoundly significant insight into the nature of Augustus’ principate.Footnote 21 But even this exciting new idea did not help with the question of Augustus’ honorifically decorated doorway and uestibulum. Carettoni never mentioned it, and there was no place for it in his reconstruction of the house.Footnote 22
Though incompatible with Suetonius’ description, the newly established dimensions of the supposed ‘house of Augustus’ were very appropriate for a residence now interpreted as imitating the ‘palace-sanctuaries’ of Hellenistic kings. However, the house did not have a long life: it was soon abandoned and incorporated into the foundations of subsequent constructions at a higher level. According to Carettoni, that happened after Augustus’ death (Tomei, Reference Tomei2014: 296), but the next detailed examination of the site proposed a different order of events.
Investigating the supposed site of the ‘portico of the Danaids’, Irene Iacopi and Giovanna Tedone (Reference Iacopi and Tedone2006: 370–1) argued that the house Carettoni excavated had been put out of use by the construction of the Apollo complex itself soon after 31 BC: in order to make space for a grander concept (‘un piano di ampio respiro’), Octavian abandoned his previous house and constructed a new residence at a higher level.Footnote 23 The true ‘house of Augustus’, they proposed, was adjacent to the temple on the west side,Footnote 24 incorporating both the upper level of the ‘Carettoni house’ and also the so-called ‘house of Livia’ (Iacopi and Tedone, Reference Iacopi and Tedone2006: 376–7; Iacopi, Reference Iacopi2007: 13). But still no proposal was made for the position of the doorway and uestibulum.
The purpose of their investigation was to explain ‘the entire pavilion of the palace that developed around the temple of Apollo’ (Iacopi and Tedone, Reference Iacopi and Tedone2006: 351).Footnote 25 But what exactly was a ‘pavilion’ (padiglione)? The word had previously been used in a classic study by Eugenio La Rocca discussing the most luxurious of all imperial properties, the horti Lamiani. That was where in AD 40 a Jewish delegation from Alexandria tried in vain to get the attention of Gaius ‘Caligula’ as he toured the estate inspecting the various buildings.Footnote 26 The eyewitness who reported that event described the buildings as ἐπαύλεις, a word with a wide range of meanings; evidently they were not simply houses in the normal sense. As La Rocca pointed out (Reference La Rocca, Rocca and Cima1986: 29), the story reveals an extensive imperial residence consisting of a series of interconnected buildings and ‘pavilions’ (padiglioni), recalling ‘the most elaborate residential complexes of Hellenistic rulers’. To make that apply to Augustus’ residence, two assumptions are necessary: that horti on the Esquiline were directly comparable with town houses on the Palatine, and that Caligula, who certainly did behave like a Hellenistic ruler,Footnote 27 was directly comparable with Augustus 70 years earlier.
The new archaeological evidence and its impact soon resulted in major monographs on the historical development of the Palatine (Royo, Reference Royo1999; Cecamore, Reference Cecamore2002; Mar, Reference Mar2005), the last of which, Ricardo Mar's volume El Palatí, effectively defined a new consensus (Mar, Reference Mar2005: 339):Footnote 28
When Augustus decided to build his house at the top of the Palatine Hill, the civil war was still far from being definitively resolved and the design of the future imperial regime was, in many aspects, still unknown. We know that the political regime was shaped with the exercise of power, first by Augustus and then by his successors. The Palatine Palace soon proved to be an important propaganda instrument … We have only to look at the models that guided the governing actions of Augustus. Indeed, the prince's policy fluctuated between respect for the old Roman traditions and an attempt to establish a Hellenistic type monarchy. In the same way, his Palatine residence was conceived within the limits imposed by the Roman republic on the palaces of noblemen, but with the ultimate aim of creating the palace of an absolute monarch.
Since then it has been common ground that Augustus had a palace.Footnote 29 Some think of it as a ‘palace-sanctuary’,Footnote 30 others as a ‘pavilion-palace’,Footnote 31 but with no definition of either term. And the question remains: where was the uestibulum with its famous doorway?
2. TEXTUAL EVIDENCE
It is worth comparing the current consensus with what the ancient sources say about Augustus’ residence. Beyond the contemporary evidence, set out above in the introduction, by far the most useful information is provided by Suetonius (Aug. 72.2):
He lived first near the Roman Forum above the ‘Ringmakers’ Steps’, in a house which had belonged to the orator Calvus, and afterwards on the Palatine, but in the no less modest house of Hortensius, which was distinguished for neither space nor elegance, having as it did only short porticoes with columns of Alban stone, and rooms without any marble or luxurious pavements. And for more than forty years he remained in the same bedroom for winter and summer, although he found the city unfavourable to his health in winter and consistently wintered there.Footnote 32
Since Quintus Hortensius (son of the famous orator) fought for Brutus at Philippi,Footnote 33 it is likely that the future Augustus acquired his Palatine house in the proscriptions of 43–42 BC.Footnote 34 The ‘more than forty’ years he used the same bedroom must have come to an end in AD 3, when the house was burned down and had to be rebuilt.Footnote 35
Suetonius was well informed about the late Republic and the Augustan period,Footnote 36 and ‘unlike most ancient historians and some of their modern successors, understood the fundamental distinction between primary and secondary sources’ (Cornell, Reference Cornell and Cornell2013: 126). A learned author with very wide-ranging interests, holding the posts of a studiis, a bybliothecis and ab epistulis on the imperial staff under Trajan and Hadrian (AE 1953.73), he had access to a vast range of source material, both literary and documentary. He knew what he was talking about. After giving this circumstantial account of Augustus’ successive residences, Suetonius made the specific point that the princeps ‘disliked grand and elaborate palaces’ (Aug. 72.3, ampla et operosa praetoria grauabatur).
If the current consensus is right, Suetonius must be wrong. Attempts are regularly made to discredit his testimony a priori, but on no good grounds.Footnote 37 It is not enough to assert the inherently superior authority of archaeological data.Footnote 38 In their conference on the Palatine in 2016 Manuel de Souza and Olivier Devillers made a brave attempt to do justice to all the different types of evidence,Footnote 39 and their contributor on Suetonius was Pauline Duchêne, an expert on historiography. She rightly stressed the precision of Suetonius’ treatment of Augustus’ residences (Duchêne, Reference Duchêne, de Souza and Devillers2019: 345), but then went on to say that his description had been falsified by the mismatch with the ‘Carettoni house’.Footnote 40 Why not take the mismatch as showing that Carettoni's identification of the house was mistaken? As it is, in a classic circular argument, Duchêne's authority is now cited to justify the identification itself.Footnote 41
It seems likely that Carettoni misunderstood Suetonius’ description of where the thunderbolt fell that claimed the site for Apollo (Suet., Aug. 29.3, Footnote note 13 above). That is important, because the error undermines Zanker's notion of ‘Haus und Tempel als Einheit’, and all the confident assumptions about Hellenistic-style monarchy that followed from it. The true topographical relationship of the temple with Augustus’ house is not easy to establish, but any reconstruction must be compatible with the contemporary evidence of Ovid in Tristia 3.1 (AD 9 or 10).
In that poem, Ovid's book is imagined as making its way from the Forum to the Palatine library in the company of a friendly guide (Tr. 3.1.33–8):Footnote 42
While I'm admiring everything, I see a doorway conspicuous with gleaming arms,Footnote 43 and a building worthy of a god. ‘Is this too’, I say, ‘the house of Jupiter?’Footnote 44 The oak crown gave to my mind the augury to think so. When I'm told whose house it is, I say ‘I was right: it's true that this too is the house of Jupiter.’
The book goes on to enquire about the laurels and the crown for the saving of citizens (lines 39–48), leading into a prayer to the pater patriae to save one more citizen and recall the poet (49–58).Footnote 45 And then (59–60): ‘From there, in the same direction, I am led to the shining-white temple of the unshorn god, high on its lofty steps.’Footnote 46
Three things follow from this passage. Firstly, the building was ‘worthy of a god’; so however modest the house was inside, its public forecourt must have been impressive (we know it housed the shrine of Vesta, as celebrated by Ovid elsewhere).Footnote 47 Secondly, the book was ‘led’ from the house to the temple; so the house and the temple cannot have been an architectural unity, as Carettoni and Zanker believed. Thirdly, the book went from the Forum to the house and then ‘in the same direction’ to the steps of the temple; so the temple should face north, contrary to what has always been believed in 150 years of scholarship.
These consequences are rarely given proper attention by archaeologists. For example, Iacopi and Tedone (Reference Iacopi and Tedone2006: 370 n. 43) cite the poem as if it were evidence for how access was provided to the supposed Portico of the Danaids:
L'ingresso monumentale del santuario, cui si giungeva, come ricorda Ov. Trist. 3,1,1–59, attraverso ‘celsi gradus’, era sul pendio meridionale del colle; il dislivello veniva superato per mezzo di rampe, impostate anche su preesistenti strutture, convergenti in sommità al centro del recinto porticato.
But those ‘lofty steps’ (celsi gradus) were the steps of the temple itself. It is hard to see how any reader of Ovid could suppose that they formed an entrance to the portico from the valley of the Circus Maximus;Footnote 48 nevertheless, the misreading continues to be endorsed (Pensabene, Reference Pensabene2017: 96).
Building on Iacopi and Tedone's work,Footnote 49 in 2008 Andrea Carandini and Daniela Bruno proposed an elaborate and detailed reconstruction of the ‘casa-santuario’ (Fig. 2).Footnote 50 The most striking feature of this reconstruction was that it featured two separate ‘houses of Augustus’, one on each side of the Apollo temple (but facing the other way). What is the evidence for this duplication?
Fig. 2. The ‘palazzo-santuario’, first version: Studio Inklink (Carandini and Bruno, Reference Carandini and Bruno2008: tav. II).
The two houses were identified as ‘domus publica’ and ‘domus priuata’.Footnote 51 It is certainly true that the question of public or private ownership was very important, as Dio Cassius points out (54.27.3): in 12 BC, when Augustus was elected pontifex maximus, ‘he did not take any public residence, but since it was absolutely necessary that the pontifex maximus should reside in public, he made part of his own house public property’.Footnote 52
Similarly in AD 3, when the house was destroyed by fire (Dio Cass. 55.12a.5):Footnote 53
When Augustus (re)built his house he made it all public property, either because of the contribution made by the People or because he was pontifex maximus, so that he might live in premises that were simultaneously private and public.
Suetonius refers to the same event (Aug. 57.2):Footnote 54
For the reconstruction of his Palatine house after its destruction by fire, the veterans, the jury-panels, the tribes and even individuals from the rest of the community freely contributed money, each according to his means.
In each passage only one house is referred to, in the singular (τὴν οἰκίαν, Palatinae domus). But Suetonius’ phrase was misconstrued by Carandini and Bruno (Reference Carandini and Bruno2008: 55): ‘La casa di Augusto è definita da Svetonio al plurale: domus Palatinae.’
The textual evidence is clear and consistent. There was one house, one door, one forecourt; the house was modest, though the forecourt was imposing; it was called domus publica,Footnote 55 because it had been made public property; and it was separate from the temple of Apollo. What causes the problem is how this evidence has been used and misused in support of successive archaeological reconstructions.
3. THE SORRENTO BASE
The Carandini–Bruno hypothesis largely depends on the Ovid passage about Vesta (Fast. 4.949–54):Footnote 56
Take the day, Vesta! Vesta has been received at her kinsman's threshold, as the just Fathers have decreed. Phoebus has one part, a second has gone to Vesta, he himself as the third occupies what is left from them. Stand, you Palatine laurels! May the house stand, wreathed with oak! One [house] holds three eternal gods.
One domus, but in some sense tripartite, for Apollo, Vesta and Augustus himself.Footnote 57 Combining that with the misreading of Suetonius’ Palatinae domus as plural, Carandini and Bruno found an elegantly symmetrical solution:Footnote 58
La domus Publica, connessa al culto di Vesta e dei Penates, e la domus privata, connessa al Genius e ai Lares Augusti, poste simmetricamente ai lati del tempio di Apollo, fanno pensare a Ovidio, secondo il quale un terzo del complesso domestico-sacrale era abitato da Apollo, un terzo da Vesta e quanto rimaneva, cioè l'ultimo terzo, da Augusto stesso.
The two matching houses in their reconstruction were given more or less matching internal layouts, a six-stage sequence identified as uestibulum, atrium, tablinum, peristylium, oecus and finally ‘corte’.Footnote 59 The last and innermost recess was presented as housing respectively Augustus’ lararium (in the ‘private domus’ to the west of the temple) and the aedicula of Vesta (in the ‘domus publica’ to the east).Footnote 60
However, this position for the shrine of Vesta is incompatible with the passage in Ovid. The goddess was received at Augustus’ threshold (limine), and the reference to the laurels and the oak wreath puts Ovid's meaning beyond doubt: she was installed in Augustus’ forecourt.Footnote 61 But Carandini and Bruno's reconstruction did not have a forecourt: their two houses had doors opening directly on to the street behind the temple.Footnote 62 Although that position was quickly abandoned,Footnote 63 and a new plan produced for the supposed ‘private domus’, now turned through 90 degrees and equipped with an exterior porch opening on to a different street (Fig. 3),Footnote 64 there was still no place in it for Vesta, who remained, and still remains, assigned to the rear court of the ‘domus publica’.Footnote 65
Fig. 3. The ‘palazzo-santuario’, latest version: Studio Inklink (Carandini and Carafa, Reference Carandini and Carafa2021: frontispiece). Note the ‘vestibolo’, just visible at the top left opposite the side of the temple of Victoria.
Much depends on the interpretation of the reliefs on the ‘Sorrento base’:Footnote 66 the long side ‘C’ shows the Vestals with the goddess Vesta herself in front of her circular shrine, and the adjacent short side ‘A’ (half of which survives) shows Augustus’ doorway with the oak-leaf crown (Fig. 4). It is generally assumed that the Ionic colonnade in the background of both scenes links them together as a single topographical entity (Fig. 5).Footnote 67 That would fit precisely with Ovid's reference to Vesta being received at Augustus’ threshold. Carandini and Bruno, however, treat the two scenes as separate, with side ‘A’ illustrating the garden court at the back of the supposed ‘domus publica’ and side ‘C’ illustrating the doorway at the front of the supposed ‘private domus’.Footnote 68
Fig. 4. Base with relief sculpture, Museo Correale, Sorrento. Left: the surviving half of the short side C (D-DAI-ROM-65.1251, H Koppermann). Right: the surviving two-thirds of the long side A (D-DAI-ROM-65.1252, H Koppermann).
Fig. 5. The ‘Sorrento base’, combination of sides ‘C’ and ‘A’: drawn by Seán Goddard (Wiseman, Reference Wiseman2009: 532 fig. 1).
Despite its incorporation into the monumental Atlas of Ancient Rome (Carandini and Carafa, Reference Carandini and Carafa2012 and Reference Carandini and Carafa2017), the Carandini–Bruno reconstruction has not been received with any enthusiasm,Footnote 69 and is hardly mentioned in the proceedings of the 2016 Palatine conference: Patrizio Pensabene and Enrico Gallocchio refer to it (Reference Pensabene, Gallocchio, de Souza and Devillers2019, 59), but only to note its controversial nature. Their own interpretation takes sides ‘A’ and ‘C’ of the Sorrento base as united by the Ionic colonnade, which they identify as the portico of the Danaids.Footnote 70 That too is an idea that needs careful attention.
4. THE PORTICO
The two narrative sources that report the building of the Apollo temple also report the building of a portico around it.Footnote 71 That kind of portico is a familiar phenomenon; the best-known example in Rome is the porticus Metelli, which surrounded the temples of Jupiter and Juno in the Campus Martius.Footnote 72 Nevertheless, since 2005 it has been the general archaeological consensus that Apollo's portico, commonly called ‘the portico of the Danaids’,Footnote 73 was in front of his temple, built out beyond the natural slope of the hill on an artificial platform extending towards the Circus Maximus (Figs 2 and 3).Footnote 74 Evidence for its alleged outer limit at the site of S. Anastasia is about 110 m from the supposed frontage of the temple and at a level 33 m below it.Footnote 75
It is important to understand that this conjectural portico is very largely an imagined concept, supposedly on top of a huge building (‘corpo di fabbrica’), more than 100 feet high,Footnote 76 of which neither the construction date nor the architectural design has ever been decided.Footnote 77 In its latest form it is provided with an altar commemorating Augustus’ refounding of Rome in 7 BC (sic).Footnote 78
Rather than invent symbolic monuments, it makes better sense to test the portico hypothesis against the real evidence for Augustan ideology. This ‘palazzo-santuario’ model dispenses with the original idea of a grand staircase down to the Circus Maximus, as suggested by Mar and endorsed by Pensabene (Reference Pensabene2017: 96); what is proposed is a huge square as large as the Roman Forum, completely enclosed by the portico on three sides and by the temple and the supposed two houses on the other. The only entrance to the ‘sanctuary’ area is from the north,Footnote 79 and so the only access to the portico itself is by the passages that ran along the two sides of the temple.
With that in mind, let us consider one of the most iconic passages in the whole of Augustan literature, the culminating scene on the shield of Aeneas in Virgil's Aeneid (8.714–23):Footnote 80
But Caesar, who had entered the walls of Rome in a triple triumph, was consecrating an everlasting vow to the gods of Italy — three hundred great shrines throughout the whole city. The streets were loud with gladness and games and applause; at all the temples there were matrons dancing, and altars, and before the altars slain bullocks strewed the ground. He himself, seated at the snow-white threshold of gleaming Phoebus, is reviewing the gifts of nations and fixing them to the proud doors. The conquered peoples process in a long line, as varied in language as they are in costume and arms.
Real events are referred to: the triple triumph (13–15 August 29 BC),Footnote 81 the vow to restore the temples (February 28 BC) and the newly instituted quadrennial games (autumn 28 BC).Footnote 82 Caesar Augustus sits ‘at the marble threshold’ of Apollo's temple, dedicated on 9 October 28 BC, and the ‘gifts of nations’ are of course the spoils of war, brought as tribute by ‘the conquered peoples’ and now to be put on display.
Virgil was writing in the 20s BC, when he and his readers knew exactly how one approached the new temple.Footnote 83 If the post-2005 consensus is accepted, he must have imagined this long procession of peoples and booty approaching from behind the temple along one of the passages, turning round in the portico to face Augustus and present the ‘gifts of nations’, and then filing out again, presumably along the opposite passage. But since Virgil's scene clearly implies a temple facing on to a street or piazza accessible by wagonloads of booty, the hypothesis of an enclosed sanctuary is better abandoned.
The less elaborate ‘pavilion-palace’ model of Pensabene and Gallocchio equally presupposes the huge square portico in front of the temple, and is equally incompatible with what Virgil says; their main entrance is still on the north side at the rear of the temple, and the procession with its gifts would hardly have come up a staircase from the Circus Maximus. But they do at least recognize that the portico should be ‘around’ the temple, and they therefore extend the colonnade along its two sides (Pensabene, Fileri and Gallocchio, Reference Pensabene, Fileri and Gallocchio2021: 169 fig. III,14).
5. THE VESTIBULUM
In their reading of the reliefs on the Sorrento base, Pensabene and Gallocchio take it for granted that the Ionic colonnade in the background of sides ‘A’ and ‘C’ represents the ‘portico of the Danaids’, and therefore that both the doorway of Augustus’ house and the shrine of Vesta set up in 12 BC are to be found in the immediate vicinity of the Apollo temple.Footnote 84 They place the doorway to the west of the temple, identifying ‘the new pavilion of the house of Augustus’ with the so-called ‘House of Livia’, and suggesting a sort of ‘propyleum’ projecting into the portico;Footnote 85 on the other hand, they place the Vesta shrine to the east of the temple, in the supposed ‘public sector’ of the complex (Pensabene, Reference Pensabene2017: 56).
Pensabene and Gallocchio deliberately separate Vesta from Augustus’ residential quarters because they take Ovid's passage in Fasti 4 (Footnote note 5 above) as guaranteeing a tripartite arrangement (Augustus, Apollo, Vesta), without also noticing that Ovid places Vesta at Augustus’ threshold, therefore in the forecourt (uestibulum) of his house.Footnote 86 The same error was made by Carandini and Bruno (Reference Carandini and Bruno2008: 55); it became the basis of their symmetrical ‘palazzo-santuario’ layout, and continues to be repeated (Ippoliti, Reference Ippoliti, Carandini and Carafa2021: 183).
Since both reconstructions started from Carettoni's original identification of the house immediately west of (and below) the Apollo temple as the house of Augustus,Footnote 87 it is not surprising that the ‘new pavilion’ of Pensabene and Gallocchio should be sited in practically the same place as the ‘private domus’ of Carandini and Bruno. They differ, however, in their placing of its formal entrance, the famous uestibulum featuring the laurels, the oak-leaf crown and the pater patriae inscription, with which this article began.
To start with, as noted in Section 3 (Footnote note 62 above), Carandini and Bruno ignored the uestibulum and had both their houses opening directly on to the street. Their revised version, which still incorporated the so-called House of Livia into the ‘private domus’, now positioned a colonnaded porch at its western end, with a door opening directly into an imagined atrium on the lost upper storey (Footnote note 64 above).Footnote 88 That solution has been criticized (Wiseman, Reference Wiseman2013: 261):
Less than 3m out from the wall of the house, the imagined columns [of the porch] stand on a platform about 5m above the level of the street, facing the side wall of the temple of Victoria; a flight of steps leads down at an improbably steep angle, but even so the bottom step is only about 3–4m away from the temple wall opposite.Footnote 89 There is, of course, no room for the Vesta shrine.
This porch (Fig. 3) is hardly ‘a building worthy of a god’ (Footnote note 42 above), or a likely place to choose for the lying-in-state of Augustus’ coffin.Footnote 90
Pensabene and Gallocchio avoid that objection with their idea of a ‘pavilion’ opening on to the great portico,Footnote 91 but are equally mistaken in their separation of the doorway from the shrine of Vesta. If the Ionic colonnade really is continuous from side ‘A’ to side ‘C’ of the Sorrento base (Fig. 5), then the common site it represents should be the uestibulum of Augustus’ house, with Vesta's shrine ‘at the threshold’ within it.
The fact is that each of the competing archaeological reconstructions of the supposed ‘palace of Augustus’ depends on a combination of misreadings of the ancient sources:
(1) Suet., Aug. 29.3 (Footnote note 13 above). It is assumed that the thunderbolt fell on Octavian's living quarters, which were therefore at the exact site of the Apollo temple,Footnote 92 rather than ‘on the place he had bought in order to build something’, as Dio Cassius reports (Footnote note 16 above).
(2) Ov., Fast. 4.949–54 (Footnote note 5 above). It is assumed that the ‘house holding three eternal gods’ attests a tripartite complex that separated Vesta from Augustus himself, rather than received her ‘at his threshold’.
-
Other clear statements are simply ignored:
(3) Suet., Aug. 72.2 (Footnote note 32 above): Augustus’ house was modest in both size and decoration.
(4) Suet., Aug. 72.3: Augustus ‘disliked grand and elaborate palaces’.
(5) Ov., Tr. 3.1.59–60 (Footnote note 46 above): the house and the Apollo temple were in different places, and the columns of the temple pronaos were approached from the north.
(6) Prop. 2.31.9, Vell. Pat. 2.81.3, Dio Cass. 53.1.3 (Footnote note 71 above): the Apollo temple was in the middle of the portico, and the portico went round it.
(7) Virg., Aen. 8.714–23 (Footnote note 80 above): the Apollo temple looked out on a space accessible to a long procession bringing substantial gifts of tribute.
Too often, when faced with textual evidence, archaeologists have resorted to special pleading (Footnote note 37 above), mistranslation (Carandini and Bruno, Reference Carandini and Bruno2008: 55) or declarations that such sources are inadequate,Footnote 93 as if modern investigators necessarily know better than people who were there at the time.
6. THE TEMPLE
The Apollo temple is normally reconstructed as facing south, with six columns across the front. The hexastyle pronaos owes its existence to Giuseppe Lugli's interpretation of four spaces left by robbed-out masonry that had once been deeply embedded in the concrete core: his inference that the missing blocks had supported six columns (Lugli, Reference Lugli1965: 276) is taken as proof of the temple's orientation.Footnote 94
However, Amanda Claridge (Reference Claridge, Häuber, Schütz and Winder2014: 138–41) has shown that Lugli's hypothesis is inconsistent with the testimony of Vitruvius, who defined different types of temple by the respective distances between their columns. The third of his types, called ‘diastyle’, was ‘when we can insert the thickness of three columns in the intercolumniation, as in the temple of Apollo and Diana’ (Vitr. De arch. 3.3.4). He was certainly referring to the Palatine temple, which we know from contemporary sources — Virgil, Horace and the ludi saeculares inscription — belonged equally to Apollo and his sister.Footnote 95
But if the temple was diastyle, the known width of the podium makes a six-column pronaos impossible; Lugli's interpretation of the robbed-out masonry holes will have to be rejected, and the evidence for the temple's south-facing orientation disappears. Claridge argues persuasively that the temple faced the other way. If it did, then two of the problematic passages noted above are immediately solved: the long line of conquered peoples reviewed by Augustus in Virgil's imagined scene (Footnote note 80 above) bring their tributary offerings from the direction of the Forum; and Ovid's book, approaching the same way 40 years later (Footnote note 46 above), is led in the same direction from Augustus’ doorway to the frontal columns of the temple.
It is a particularly acute example of the mismatch between archaeological inference and textual evidence. Vitruvius was an architect, precisely contemporary with the construction of the temple; it is impossible to imagine a better-qualified authority. Though Claridge made the case for a north-facing temple in 2014, seven years later Pensabene makes no mention of it in the ‘sources’ section of his chapter on the archaeology of the temple (Pensabene, Fileri and Gallocchio, Reference Pensabene, Fileri and Gallocchio2021: 93–6), and a subsequent brief reference to Vitruvius simply states that the architect should not be taken literally.Footnote 96
My own book on the house of Augustus did take Vitruvius literally, and incorporated Claridge's argument into a wider discussion of the Augustan Palatine, based explicitly on primary sources. It was presented as a challenge to archaeological thinking (Wiseman, Reference Wiseman2019: 22–9), but whether there will be a meaningful response remains to be seen.Footnote 97 There has certainly been no movement on the orientation of the temple: Lugli's interpretation has become a dogma.Footnote 98
7. CONCLUSION
Sixty years have passed since Carettoni identified the house he was excavating as the house of Augustus. Since then, the ongoing efforts of distinguished archaeologists have still failed to find an agreed answer to the question posed at the beginning of this article: where was the historic forecourt in which Augustus’ doorway was flanked by laurels and decorated with the oak-leaf crown?
I think the reason for that failure is that archaeologists have repeatedly not taken proper account of the textual evidence in testing their own inferences from the surviving material remains. The first error, I suggest, was Carettoni's original identification of the late-Republican house 9 m below the level of the Apollo temple; building on that, other misconceptions have followed, notably the idea of Octavian as a Hellenistic monarch, with its corollary, the assumption of an Augustan palace.
Neither the ‘pavilion-palace’ model of Pensabene and Gallocchio nor the ‘palace-sanctuary’ model of Carandini and Bruno (now Carandini and Carafa) is compatible with the statement of Augustus’ biographer that he disliked grand and elaborate palaces and lived for over 40 years in the same modest quarters. Outside the door of his unpretentious house was the much more impressive forecourt, a building described by a contemporary author as ‘worthy of a god’, containing the shrine of Vesta that marked Augustus’ election as pontifex maximus in 12 BC. Where was it?
The best answer was given long ago by Ferdinando Castagnoli: the house of Augustus was where the Flavian palace was,Footnote 99 and the Flavian palace was called domus Augustana for that very reason.Footnote 100 In that case Augustus’ forecourt would correspond to the entrance to the Flavian palace, which looked out on to the summit plateau of the Palatine; and we happen to know that Augustus’ forecourt looked out on to a space that could accommodate thousands of people watching the ludi Palatini in AD 41.Footnote 101
That seems to be a good fit, but the evidence is not allowed to mean what it says. We are told that domus Augustana must be just ‘the house of the emperor’,Footnote 102 and that the ludi Palatini must have been held in front of the Magna Mater temple, like the ludi Megalenses,Footnote 103 or else at the imagined monumental steps leading up from the Circus Maximus to the ‘portico of the Danaids’.Footnote 104 ‘L'ipotesi di Castagnoli appare superata dagli scavi Carettoni’ (Sasso D'Elia, Reference Sasso d'Elia and Steinby1995: 41);Footnote 105 it is assumed that archaeological inference trumps textual evidence.
That is how it has been for more than half a century, and the outcome is not satisfactory. The main difficulty is the sheer complexity of the different types of evidence that must be taken into account, but that problem is compounded by a general reluctance to take seriously what the literary sources say. The result is that the experts have still not provided an agreed, empirically defensible description of that historically pivotal time and place, the Augustan Palatine, and the reasonable expectations of non-specialist colleagues and students continue to be frustrated. In the absence of any serious counter-argument, I believe the reconstruction presented in The House of Augustus (Wiseman, Reference Wiseman2019) offers the most reliable explanation of the Augustan Palatine, if only because it gives systematic priority to what the contemporary evidence implies.
