Hostname: page-component-cd9895bd7-hc48f Total loading time: 0 Render date: 2024-12-25T15:53:21.051Z Has data issue: false hasContentIssue false
  • English
  • Français

Constructing reconstruction: claim and reality of Roman rebuilding inscriptions from the Latin west*

Published online by Cambridge University Press:  09 August 2013

Edmund Thomas  and
Christian Witschel
Get access

Costruendo ricostruzioni: rivendicazioni e realtà delle iscrizioni da riedificazioni romane provenienti dal latinum occidentale

In quest'articolo gli Autori analizzano l'atteggiamento che in epoca romana veniva riservato alia riedificazione e al restauro di edifici, utilizzando in particolare le iscrizioni di riedificazione rinvenute nella parte occidentale dell'Impero (Roma esclusa); inoltre, dove possibile, le iscrizioni sono confrontate con le evidenze archeologiche. Viene dimostrato qui che non sempre tali iscrizioni descrivono in maniera accurata i danni precedenti alla ricostruzione o il tipo di lavori eseguiti. Il loro linguaggio è spesso metaforico e vengono stabilite nozioni di distruzione e ricostruzione che non necessariamente hanno una diretta relazione con il reale stato dell'edificio o con i lavori di restauro eseguiti. La parola ‘ricostruzione’ era in generale considerata sotto un punto di vista idealistico espresso in modi diversi, colleganti fatti architettonici locali, spesso complessi, ad un'idea simbolica di rinnovamento. Quindi, a meno che una singola causa reale non sia menzionata, le iscrizioni di riedificazione non riportano nessun dato definitivo circa le circostanze reali della distruzione o della ricostruzione di un edificio. Queste devono essere studiate individualmente e non devono influenzare l'analisi indipendente dei resti archeologici.

Type
Articles
Information
Papers of the British School at Rome , Volume 60 , November 1992 , pp. 135 - 177
Copyright
Copyright © British School at Rome 1992

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

Footnotes

*

The idea for this paper arose after a presentation by Dr Arthur Segal of the University of Haifa at a seminar in Oxford in March 1990. Talking about the Roman buildings of Beth She'an (Scythopolis) in Israel, he observed the incongruity of the inscription belonging to the Nymphaeum (cf. Excavations and Surveys in Israel 1987/8, 6 (Jerusalem, 1988), 27–8, giving only a translation), which commemorates a fourth-century construction ‘from the foundations’ of a building archaeologically datable almost entirely to the second century; the fourth century saw just the addition of pools and water-supply systems. An earlier form of the present paper was subsequently delivered at the same seminar. We should like to thank all those who attended the seminar, especially Jim Coulton, Janet DeLaine, John Lloyd and Dr Segal himself for their helpful comments and suggestions, as well as others who have offered kind advice to us in the course of preparation, including Gunnar Brands, Amanda Claridge, Manfred Clauss, Howard Colvin, Nicholas Purcell, Frank Sear, Margareta Steinby and Bryan Ward-Perkins. Finally, we should both like to thank the staff of the British School at Rome for their hospitality during a short stay in September and October 1990, during which the final researches were carried out. Edmund Thomas would also like to thank the Craven Committee of Oxford University and St John's College, Oxford for their generous support to him in financing this period of travel.

References

1 For periodicals, the abbrevations are those given in L'Année Philologique, except where there is no abbreviation given there, in which case the abbreviation given by Archäologische Bibliographie is used. For epigraphic or legal collections, the abbreviations given by Berard, F. et al. (eds), Guide de l'Epigraphiste2 (Paris, 1989), 16–7Google Scholar are used. For literary sources (authors and texts), we use the abbreviations given by the Oxford Classical Dictionary.

2 CIL 6.896 = 31196 = ILS 129 (M. Agrippa Lf. cos. tertium fecit).

3 Terenzio, A., ‘La restauration du Panthéon de Rome’, Museion 20 (1932), 52–7Google Scholar and La conservation des monuments d'art et d'histoire (Paris, 1933), 280ffGoogle Scholar (cited by Licht, cit. (n. 4).

4 Guey, J., ‘Devrait-on dire: Le Panthéon de Septime-Sévère ?’, MEFRA 53 (1936), 198–249CrossRefGoogle Scholar; de Fine Licht, K., The Rotunda in Rome. A Study of Hadrian's Pantheon (Arhus, 1966), 190, 290Google Scholar n. 46 (with literature).

5 Licht, cit. (n. 4), 202. The iconoclastic argument of Cozzo, G., ‘La corporazione dei figuli ed i bolli doliari’, MAL6 5.4 (1936), 233–366Google Scholar that ‘il Pantheon, castruito da Agrippa, fu due secoli dopo, restaurato in modo assolutamente radicale, dai due Severi’ (343), has been convincingly refuted by Guey.

6 Licht, cit. (n. 4). For an uncritical list of their building activity in Rome, see Benario, H. W., ‘Rome of the Severi’, Latomus 17 (1958), 712–22Google Scholar.

7 Most recent surveys of building activity including restoration:Italy and Africa: Jouffroy, H., La construction publique en Italie et dans l'Afrique romaine (Strasbourg, 1986)Google Scholar; Spain: Gurchin, L. A., ‘Personal wealth in Roman Spain’, Historia 32 (1983), 227–44Google Scholar; Gaul and Germany: Frézouls, E., ‘Evergétisme et construction urbaine dans les Trois Gaules et les Germanies’, Revue du Nord 66, no. 260 (1984), 27–54CrossRefGoogle Scholar; Britain: Blagg, T. F. C., ‘Architectural munificence in Britain. The evidence of inscriptions’, Britannia 21 (1990), 13–31CrossRefGoogle Scholar; Africa in Late Antiquity: Lepelley, C., Les Cités de l'Afrique Romaine au bas-Empire, Tome 1, La permanence d'une civilisation municipale (Paris, 1979)Google Scholar, Tome 2, Notices d'histoire municipale (Paris, 1981)Google Scholar, Waldherr, G., Kaiserliche Baupolitik in Nordafrika. Studien zu den Bauinschriften der diokletianischen Zeit und ihre räumlichen Verteilung in den römischen Provinzen Nordafrikas (Frankfurt, 1989)Google Scholar; general: Duncan-Jones, R. P., Structure and Scale in the Roman Economy (Cambridge, 1990), ch. 11, ‘Who paid for public buildings?’, 174–84CrossRefGoogle Scholar.

For misconceptions about reconstruction and, particularly, vetustas: Curchin, 241; Frézouls, 46–8. Blagg, 27 is best, but refuses to concede the key point of non-denotative reference (‘a phrase [vetustate collapsum] which should not always be taken too literally, since it may tactfully conceal the results of structural defects or enemy action’—our italics).

8 Cf. Waldherr, cit. (n. 7), 94: ‘Die Wendung vetustate conlapsas … könnte an eine gewisse wirtschaftliche Stagnation oder sogar Regression… denken lassen’.

9 S.H.A. Ant. Pi. 8.2 (instauratum … templum Agrippae); cf. Bloch, H., I bolli lalerizi e la storia edilizia romana (Rome, 1947), 116Google Scholar (two brickstamps), pace Guey, cit. (n. 4), 239 n. 1 (three).

10 S.H.A. Hadr. 19.10 (Romae instauravil Pantheum).

11 CIL 8.18260.

12 CIL 8.18328 = ILS 5520, AD 379/83 (Aureis temporibus … non solum labsa reparantur sed et nova pro felicitate construuntur: curia igitur ordinis quam maiores nostri merito templum eiusdem ordinis vocitari voluerunt vetustate immo incuria ve(te)rum in odiu (sic) feda[ta] iacuisse [v]idebatur, qua(e) nunc ex novo opere in eodem solo egregiae (sic) cognoscitur…).

13 Janon, , ‘Recherches à Lambèse III: essais sur le temple d'Esculape’, AntAfr 21 (1985), 36CrossRefGoogle Scholar n. 6. He adds that Lepelley (above (n. 7), 420–1) ‘me semble accorder un crédit un peu rapide à une témoignage épigraphique hors-contexte’.

14 Inscriptions ‘constituted the best means by which priests and pious alike could impress their beliefs on the public entering the shrines’ (MacMullen, R., Paganism in the Roman Empire (New Haven, 1981), 31)Google Scholar; for secular epigraphic visibility, cf. Beard, M., ‘Writing and ritual: a study of diversity and expansion in the Arval Acta’, PBSR 53 (1985), 114–62Google Scholar, and Corbier, M., ‘L'écriture dans l'espace public romain’, in L'Urbs: espace urbain et histoire (Rome, 1987), 27–60Google Scholar.

15 This polar alternative is a marked feature of previous debate, for example on the rebuilding inscriptions from Roman Britain, where Grace Simpson's reading of them as governed by stereotype determines her interpretation that they hide damage caused by enemy attack (‘Caerleon and the Roman forts in Wales in the second century AD’, ArchCam 111 (1962), 103–66Google Scholar; eadem, Britons and the Roman Army (London, 1964), 48–52)Google Scholar, denied by Welsby, D., ‘Roman building inscriptions, recording buildings collapsed through age or destroyed by the enemy?’, ArchAel5 8 (1980), 89–94Google Scholar, on the grounds that they should be taken literally; debate on fourth-century, north African examples, is similarly misguidedly framed, between Lepelley, C., ‘L'Afrique du nord et le prétendu séisme du 21 juillet 365’, MEFRA 96(1) (1984), 463–91CrossRefGoogle Scholar and di Vita, A., ‘Sismi, urbanistica e cronologia assoluta. Terremoti e urbanistica nelle città di Tripolitania fra il I secolo a.C. ed il IV d.C.’, in L'Afrique dans l'Occident romain (Rome, 1990), 425–94Google Scholar.

16 As (most recently) by Waldherr, cit. (n. 7); earlier, Stuart, D. R., ‘Imperial methods of inscription on restored buildings: Augustus and Hadrian’, AJA 9 (1905), 427–49CrossRefGoogle Scholar, was too ready to assume that the highest possible claim must have been the claim of new construction. He suggested that the message of reconstruction was ignored by Augustus and Hadrian out of political tact: ‘an emperor laid himself open to ridicule or criticism by appending his titulus when repairs had been insignificant’ (434). It is hoped that this paper will suggest the inaptness of that statement.

17 Nemo dubitat quin multo maius sit novam (columnam) facere (‘No one doubts that it is a much bigger thing to make (a column) new’: Cic. Verr. 2.1.147).

18 For the nature of aedificatio as random congerere, unless controlled by dispositio, cf. Quint, Inst. 7 pr. 1.

19 On monumenta dilapsa, result of various natural processes manifesting time, cf. Lucr. 5.306–17, with Kinsey, T. E., ‘Again Lucretius V.312’, Latomus 25 (1966), 553–4Google Scholar on the textual corruption, and Berns, G., ‘Time and nature in Lucretius' “De rerum nature”’, Hermes, 104 (1976), 477–92Google Scholar for the general ideas.

20 For: Cic. De Or. 2.60–4; (ironically) Lucian Hist. Conscr., passim; against: Tac. Ann, 13.31.1, with Syme, R., Tacitus (Oxford, 1958), vol. 1, 226Google Scholar.

21 See Malissard, A., ‘Incendium et ruinae. A propos des villes et des monuments dans les Histoires et les Annales de Tacite’ in Chevallier, R. (ed.) Présence de l'architecture et de l'urbanisme romains. Hommage à Paul Dufournet (Paris, 1983), 45–55Google Scholar for the negation of architecture in the works of Tacitus.

22 FIRA I2 (ed. Riccobono, S.; Florence, 1941), 18Google Scholar (Lex Municipii Tarentini: (4), 11.32–8), 21 (Lex Coloniae Genetivae Iuliae seu Ursonensis: LXXV, Tab. II, col. 2, 11.17–23) and 24 (Lex Malacitana: LXII, col. 3, 11.59–72). For the last, see Spitzl, T., Lex Municipii Malacitani (München, 1984), with commentary at 79–82Google Scholar. For the Lex Irnitana, see Gonzalez, J., ‘The Lex Irnitana: a new copy of the Flavian municipal law’, JRS 76 (1980), 147–243Google Scholar (Tab. VIIA, LXII, 11.38ff, on 166).

23 On these rebuilding laws in general, see especially Sargenti, M., ‘La disciplina urbanistica a Roma nella normativa di età tardo-repubblicana e imperiale’, in La città antica come fatto di cultura (Como, 1983), 265–84Google Scholar and Garnsey, P., ‘Urban property investment’, in Finley, M. I. (ed.) Studies in Roman Property (Cambridge, 1976), 133–6Google Scholar.

24 Res Gestae Divi Augusti 20.2 (note the modestly vague compluribus locis).

25 FIRA, no. 45, 11.6–10. The metaphor in cruentissimo and inimicissiman pace faciem assimilates ordinary commercial trade in building land and materials to a threat of military conquest, and so ruina suggests specifically the debris left after a hostile assault in war.

26 Note the use of the verb custodi[re].

27 FIRA, no. 45, 11.36–7, 42–5 (suppl. Mommsen).

28 CJ 8.10.2 (AD 222).

29 Vespasian: Suet. Vesp. 8.5; reconstruction statements: idemAug. 18.2, 30.2; Calig. 21; Claud. 25.5. Cf. Gros, P., Aurea templa. L'architecture réligieuse de Rome à l'époque d'Auguste (Paris, 1976), 18–9Google Scholar on Dio's history of rebuilding in the Augustan period, in which fires and natural disasters occur only at significant political moments. For Tacitus, this shows the vanity of architecture: cf. Malissard, cit. (n. 21).

30 SIG 3 837 = IGRR 4.1156a: ‘so as [the house] should not be brought down by time and neglect’ (ὡς μὴ χρόνωι [καί α]μελ(ε)ίαι καταριϕθείη).

31 Dig. 50.10.7.pr.; cf. 50.8.7.

32 Details in Janvier, Y., La législation sur les édifices publiques dans le bas-Empire (Paris, 1974)Google Scholar.

33 Tacitus's aedificiorum et hominum strages (Ann. 1.76.1).

34 Cf. Lepelley, cit. (n. 15), 482–5, interpreting the correspondence between Code and inscriptions too literally and thus taking the number of inscriptions per period as an index of actual building activity in the fourth century. Distaste for age together with admiration for the old is a theme of later history, on which see Lowenthal, D., The Past is a Foreign Country (Cambridge, 1985), especially 145Google Scholar: ‘Shabbiness seldom brings history to life; the only way the past can seem real is if its relics are in their prime‘. Cf. Plutarch on the Periclean monuments of Athens (172, below).

35 Amm. Marc. 27.3.7, of C. Caeionius Rufius Volusianus Lampadius, praefectus urbi in 365/6 (an example of his use of his own name with a verb of new construction may be the verse inscription AE 1986.109, which records baths that he had built (δειμαμένου)); compare contemporary inscriptions, mostly from north Africa, where, on the one hand, long, exaggerated descriptions of strages invoke vetustas ideology, and, on the other, these end with verbs suggesting new work, e.g. perfecit, institut, dedicavit, etc.

36 Cassiod. Var. 4.51, proposing restoration of Pompey's Theatre.

37 E.g. AE 1984.450, probably from Gallia Narbonensis, and his activity at the mons Aeflanus (CIL 14.3530 = ILS 3512 = InscrIt 4.1.611).

38 E.g. his reconstruction of amphitheatra at Lambaesis (AE 1955.135) and Mesarfelta (CIL 8.2488), both vetustate corrupta and a solo.

39 See now Waldherr, cit. (n. 7), passim.

40 Lucr. 1.225 (quaecumque vetustate amovet aetas), 2.69–70 with Berns, cit. (n. 19) and Segal, C., Lucretius on Death and Anxiety. Poetry and Philosophy in De rerum natura (Princeton, 1990), esp. ch. 5 ‘The world's body and the human body: walls, boundaries and mortality’, 94–114Google Scholar.

41 Plaut, . Mostell. 146–8Google Scholar: ‘I don't seem to be able to patch up my house to stop it falling down all at one go …’ (non videor mihi sarcire posse aedes meas, quin totae perpetuae ruant…, with explanation of the force of perpetuus here by Ramsay, W., The Mostellaria of Plautus, ed. Ramsay, G. G. (London, 1889), 123)Google Scholar. Cf. the human analogy of Pliny Ep. 6.30.2: ‘The Camillian villa … is indeed harassed with old age; … so let's wait for (your more precious possessions) to be restored to the best possible health’ (Villa Camilliana… est quidem vetustate vexata… Attendimus ergo ut quam saluberrime reficiantur; see below for an additional nuance of vexatus).

42 The phrase appears at AE 1984.151 (Lavinium, 313/24) ([?v]i temporis deformatas), by analogy with vi ignis (e.g. CIL 14.376, Ostia, c. 160/90), vii (sic) tempestatis (CIL 14.113, Ostia, 196), vi maris (CIL 10.1640–1, Puteoli, 139), vi terrae motus (e.g. CIL 9.2338, Allifae, c. 352/7), perhaps [vi flu]mi[nis] (AE 1969/70.135 = Chelotti, M. et al. (edd.), Le epigrafi romane di Canosa (Bari, 1985)Google Scholar, no. 253; Canusium, 210), violentia torrentis (AE 1958.269, Portus Curensis, ? third/fourth century), and maris et fluminum violentia (Bastianelli, S., Centumcellae (Rome, 1954), 87Google Scholar, Castrum Novum, 205/6). For the military metaphor incursio applied to fire in combination with the definite suggestion of burning in concrematus, cf. CIL 8.8457 (Sitifis, stressed by the annus provinciae of 288).

43 Mart. 1.12.7 (Subito collapsa ruit porticus Reguli), Tac. Ann. 2.47 (urbes collapsae nocturno motu terrae). For the image, Lyne, R. O. A. M., Words and the Poet. Characteristic Techniques of Style in Vergil's Aeneid (Oxford, 1989), 39–43Google Scholar.

44 E.g. Vitr. De Arch. 5.10.3 (‘ … then water from steam will not be able to corrupt the timber of the woodwork’, non enim a vapore umor corrumpere poterit materiem contignationis), Tac. Hist. 1.86 (‘the islands’ foundations corrupted, then collapsed from the intrusion of the stream', corrupta … insularum fundamenta, dein remeante flumine collapsa).

45 Thus, the record of, for instance, ‘towers consumed by old age’ (CIL 3.2907, Iader, turris vetustate consumptas) is drawn directly from a perceived equivalence to real military danger, for example, Caes. BCiv. 2.14.2 (‘so that towers and siege-engines nurture flame and they all be consumed’, uti… turris, tormenta flammam conciperent et haec… omnia consumerentur).

46 Vetustate vexatus: CIL 14.2804 = ILS 6218 (Gabii, 138/61), CIL 11.3801 (Veii, Tiberian); vetustate quassatus: AE 1987.307 (Canusium?, second/third century), cf. Frontin. Aq. 7.

47 Gell. NA 2.6.

48 CIL 13.6592 = ILS 9184; Baatz, D., ‘Das Badegebäude des Limeskastells Walldüm (Odenwaldkreis)’, SJ 35 (1978), 61–95Google Scholar.

49 CIL 2.3270 = ILS 5513 =Vives, J., Inscripciones latinas de la Espana romana (Barcelona, 1971–1972)Google Scholar, no. 1417.

50 Details and all literature given by Humphrey, J. H., Roman Circuses. Arenas for Chariot Racing (London, 1986), 668 n. 27Google Scholar (who prefers the third century).

51 Livy 24.41.7. Blázquez, J. M., ‘Die Stadt Castulo (Hispanien) in der römischen Kaiserzeit’, in Wirth, G. (ed.), Romanitas-Christianitas. Untersuchungen zur Geschichte und Literatur der römischen Kaiserzeit Johannes Straub zum 70. Geburtslag am 18. Oktober 1982 gewidmet (Berlin-New York, 1982), 727–48Google Scholar, esp. 743, and, with Molina, F., ‘La muralla de Castulo’, in Blázquez, J. M. (ed.), Castulo II (Madrid, 1979), 268–82Google Scholar. G. López Pinto (cited by Blázquez and Molina, 269 n. 160) claimed in 1657 to have seen the walls in their complete extent, but some sections of the circuit now no longer survive. It is thus possible that those parts may have been repaired during the interval.

52 The situation is complicated by the suggestion that the real cause of rebuilding might have been something other than vetustas, namely an earthquake, or, more likely, the Frankish invasion of c. AD 260 (cf. Oros. 7.41.2). In this case, the human violence might have been perceived as the culmination of the ravages of time.

53 For a similar gloss on the process of vetustas through the locally specific events of torrents, cf. Carcopino, J., ‘Deux inscriptions du département de Constantine récemment publiées: I. Inscription d'Ain-Aziz-ben-Tellis’, BCTH 1914, 562–6Google Scholar, on an inscription found near Aziz Ben Tellis, in Numidia (‘the fount of Caputamsaga fallen apart from old age and demolished by persistent torrents’, … fontem Caputamsagae vetustate dilapsum et torrentibus adsiduis dim ⟨m⟩[o]litum …).

54 Longa incuria et vetustate: AE1939.151 (Serino, 323/4), CIL 12.4355 = ILS 5904 (Narbonne, probably fifth century); multorum incuria dilapsam et per longam annorum seriem neglectam: CIL 8.2572 = ILS 5786 (Lambaesis, 290/3); neglegentia temporum: BCTH 1907, 274 (Timgad, 303); vetustate et terrae motu; AE 1912.216 (Samos, 47), AE 1913.227 (Reggio di Calabria, 374); in praeteritum igne: CIL 11.4781 = ILS 739 (Spoletium, probably 356); ante plurima tempora rebellium incursione: CIL 8.20836 = ILS 638 (Rapidum, probably c. 300/305); vetusta[te dilapsum… ] … bell(o) dissi[patum]: IRT 896 (Gheriat el-Garbia, probably 239); ante annos plurimos… vi torrentis: CIL 8.2661 = ILS 5788 (Lambaesis, 276/82).

55 Longa incuria: CIL 10.5917 = ILS 1909 (Anagnia, late Antonine); longi temporis incuria: CIL 14.135 (Ostia, 340/50); velustatis incuria: AE 1955.287b (Ostia, Constantinian?: Zevi, F., ‘Miscellanea Ostiense: I-La carriera di Gavio Massimo e i restauri tardi alle terme del foro’, RAL 26 (1971), 467Google Scholar with n. 51); perenni incuria: CIL 10.4860 (Venafrum, 198/209).

56 See the full catalogue in Guidoboni, E. (ed.) I Terremoti prima del mille in Italia e nell'area mediterranea (Bologna, 1989), 139–67Google Scholar; for the East, Robert, L., ‘Documents d'Asie Mineure: V. Stèle funéraire de Nicomédie et séismes dans les inscriptions’, BCH 102 (1978), 395–408CrossRefGoogle Scholar.

57 Terremoti (n. 56), passim. The same reconstruction of thermae Sabinianae at Telesia is recorded by two inscriptions, of which one (AE 1972.150) mentions an earthquake and the other (CIL 9.3018 = ILS 5761) does not. The precise reasons for this selectivity remain elusive, but it is made more likely by the non-denotative value attached to such expressions of damage. In the case of early Flavian Campania, the assumption that all inscriptions mentioning destruction by a single ‘earth movement’ (terrae motu) must date to before the AD 79 eruption, and all those with the vague plural terrae motibus to after 79, is unjustified.

58 AE 1913.227 = Supplementa Italica 5 (1989), 54 ([t]hermas vetustate et terrae motu conlapsas in meliorem cultum formamque auspiciis felicioribus reddiderunt).

59 The same combination appeared 327 years earlier in the Emperor Claudius's record of restoration of the temple of Liber Pater on Samos: AE 1912.216 + 1987.931.

60 Cit. (n. 58): … quae [n]umquam habuerat.

61 CIL 13.11757 (quod aqua non esset, induxit).

62 CIL 13.11758 (aquam [[Alexandrianam]] perduxit).

63 CIL 13.11759 = ILS 9179b (aquam Gordianam multo tempore [inter]missam n[ovo aq]uaed[u]ctu pe[rd]uxit).

64 For instance, the rulers Gallienus, Salonina and Saloninus could name a bath-building thermas Gal[lienas], yet also claim that this was the building [r]eformatas et excultas (ILAfr 506, on which see Merlin, A. and Poinssot, L., ‘Une nouvelle inscription de Teboursouk’, Mémoires de la Société Nationale de la France 72 (1913), 109–58)Google Scholar.

65 … aquas Curtiam et Caeruleam perductas a divo Claudio et postea a divo Vespasiano … restitutas cum a capite aquarum a solo vetustate dilapsae essent nova forma reducendas… curavit (CIL 6.1258, AD 81).

66 Summed up by Salama, P., ‘Nouveaux témoignages de l'oeuvre des Sévères dans la Maurétanie Césarienne’, LibycaBServAnt 3 (1955), 351Google Scholar: ‘ce phénomène tient … beaucoup moins à la fragilité ou à la malfaçon des constructions romaines qu'à une question de terminologie latine’.

67 Temple remains: Cagnat, R. and Gauckler, P., Les monuments historiques de la Tunisie—les temples païens (Paris, 1898), 37Google Scholar, with plates XXXIV-XXXV. The well preserved parts of the temple visible in the plates show no clear signs of a rebuilding, but we could not expect any, of course, if the claim of full restitutio were correct.

68 CIL 8.26400 ( … quod C. Arafrius pater fecerat vetustate ad solum corruptum ab solo restituit).

69 Suggestions of destruction a fundamentis are used by ancient historians, in particular, for emphasis (e.g. Nep. Hann. 7.7, Timol. 3.3; Livy 26.13.16, 42.63.11). They are the reverse of the historical statements of reconstruction mentioned above (n. 29).

70 In the same way that mentioning the destruction by fire of the Forum and Basilica Julia (Res Gestae Divi Augusti 20.3) increases Augustus's pietas in reconstruction.

71 E.g. several at Madauros in Africa Proconsularis, cf. ILAlg 1. 2100–8.

72 IRT 55 (antiqua ruina cum labe). Cf. Kenrick, P. M., Excavations at Sabratha 1948–1951 (London, 1986), 1–67Google Scholar, esp. 29–32. The connection of this disaster with the hypothetical earthquake of 306/10 (proposed by di Vita (n. 15), but not mentioned in any source) remains highly problematic. On the difficulty in identifying an earthquake by archaeology in general, see Ward-Perkins, B. R., ‘Archeologia e terremoti: le tracce perdute’ in Terremoti (n. 56), 409–13Google Scholar, and Russell, K. W., ‘The earthquake of May 19, AD 365’, BASO 238 (1982), 47–64Google Scholar.

73 For another case where destruction mentioned in inscriptions is almost certainly confirmed by archaeology, cf. the fabri[ca], [portic]us and taber[nae] at Martigny, vi ign[is consu]mpta and probably restored in the Severan period (AE 1945.124): a thick burnt layer on the floors of Phase I of the porticoes with a coin of Faustina Minor clearly indicates a big fire in the later second century ad (Simonett, C., ‘Octodurus. Kurzer Bericht über die Ausgrabungen 1938/39 in Martigny (Wallis)’, ZSchwArch 3 (1941), 77–94)Google Scholar.

74 Against Simpson (above, n. 15), note the readiness of some inscriptions to mention previous destruction by a military enemy, e.g. AE 1974.285 (Sipontum, perhaps late Republican), IRT 896 (Gheriat-el-Garbia, probably 239), CIL 8.9041 = ILS 627 (Auzia, 290), CIL 8.20836 = ILS 638 (Rapidum, probably 300/1), perhaps CIL 10.1693 (Puteoli, 393/4); Welsby (n. 15) is misguided to argue that the invocation of vetustas necessarily rules out the possibility of destruction by the enemy: it simply says nothing about it. For a discussion of the same problem elsewhere, see Baatz, D., ‘Zum archäologischen Nachweis eines Alemanneneinfalls am obergermanischen Limes unter Elagabal’, BJ 171 (1971), 377–85Google Scholar.

75 E.g. CIL 5.7376 (Dertona, 22 BC: porticum vet[ustate corruplam]); NSA 1953, 299ff. no. 67 (Ostia, Augustan: quot vetus[tate deficiebat]); CIL 9.2983 (Cluviae, Tiberian: vetustate conlaps[---]); perhaps also CIL 10.6526 (Cora, first century BC?: longo tempore) and 9.3885 (Lucus Angitiae, late Republican?), but right-hand margin broken off: vet[ust.] consumtum).

76 AE 1902.40 (Surrentum, 80: terrae motibus conlapsum).

77 CIL 13.7800 = ILS 9363 (Remagen, 218: ab horis intermissum et vetustate conlabsum).

78 Olim: CIL 8.8507 (Sitifis), 13.939 = ILS 4638 (Périgueux), 14.126 = ILS 608 (Ostia, 283/4), RIB 1988 = ILS 4788 (Castlesteads); dudum: CIL 5.7270 = ILS 5701 (Segusio, 375/8); iam: RIB 927 = EphEp 9.1226 (Old Penrith); nuper: CIL 8.8393 + 20266 (Satafis, 379/83), RIB 979 (Netherby, 222). Compare the vital donec of the Augustan reconstruction of Rome (Hor. Carm. 3.6.2; Livy pr. 9).

79 E.g. AE 1937.119 (Amiternum, tabula patronatus with consular date of 325 in line 1): aquas Arentani quas iam dilaps(a)e fuerant civitati n(ostrae)… restituit, thermas quas iam olim disperierant antiquitus… at pulcridinem (sic) restauravit (11.19–22).

80 AE 1905.35 (Macomades, 265, again a consular date), cf. CIL. 11.3780 (Veii: Victoriae Aug(ustae) sacrum restitutae post antiquissimam vetustatem).

81 RIB 141 (Bath): the excess of age is reiterated, not much less vaguely, in the fragment --- Jlonga seria[---, which also seems to have belonged to the same frieze of the ‘Four Seasons’ façade; ILAlg 1.2048 (Madauros, 290/4): nimia vetus[tate] dilabsam et per annos plurimos intermissam.

82 Juv. 10.106–7: excelsae turris… unde altior esset casus et inpulsae praeceps inmane ruinae.

83 Cic. Verr. 2.1.56.146–8: Si quid operis causa rescideris, reficito… Rediviva sibi habeto.

84 Cf. Deichmann, F. W., Die Spolien in der spätantiken Architektur, SBMünchen Phil. hist. Klasse (München, 1975)Google Scholar. For the question of historical restoration of antique styles of architecture, which is separate from that of reconstruction as conceived here, see the discussion of the theatre at Mérida, below (Appendix 2).

85 CIL 6.1246 (… disturbatos per C. Caesarem a fundamentis novos fecit ac restituit). Cf. Ramage, E. S., ‘Denigration of predecessor under Claudius, Galba, and Vespasian’, Historia 32 (1983), 205Google Scholar.

86 Nep. Att. 20.3; Livy 4.20.7. For the preservation of the (τò) ὰρχαῖον ῖχνος or vetus forma (cf. Tac. Hist. 4.53.2 on the Capitolium in 69), cf. Dion. Hal. Ant. Rom. 2.34.4, in an antiquarian context.

87 Harrer, G. A., review of Jean Gagé. Res Gestae Divi Augusti. Les Belles Lettres, Paris, 1935Google Scholar, AJPh 58 (1937), 249Google Scholar. Cf. Sablayrolles, R., ‘Espace urbain et propagande politique: l'organisation du centre de Rome par Auguste (Res Gestae, 19 à 21)’, Pallas 28 (1981), 59–77CrossRefGoogle Scholar.

88 Harrison, S.J., ‘Augustus, the poets, and the spolia opima’, CQ 39 (1989), 409CrossRefGoogle Scholar.

89 Zanker, P., Augustus und die Machl der Bilder (München, 1987)Google Scholar, esp. ch. IV.

90 Croke, B. and Crow, R., ‘Procopius and Dara’, JRS 73 (1983), 143–59Google Scholar.

91 Downey, G., ‘Imperial building records in Malalas’, ByzZ 38.1 (1938), 6Google Scholar. (We owe this reference to Judith McKenzie, Oxford.)

92 AE 1939.151 (… aquaeductum longa incuria et vetustate … refici iusserunt et usui civitatum infra scriptarum reddiderunt…).

93 Liber Pontificalis 34 (Silvester), ed. Duchesne, L., 1.186Google Scholar, xxxiiii (per VIII milia passuum fecit), giving the date 323/4.

94 Sgobbo, I., ‘Serino — L'acquedotto romano della Campania: “Fontis Augustei Aquaeductus”’, NSA 1938, 75–97Google Scholar, with figure.

95 As there seem to have been numerous smaller repairs between the Augustan construction and the big Constantinian reconstruction and afterwards, some of the alterations described may very well belong to other periods. Restoration was normally a continual process. This certainly belies the simplistic images of single, sudden events of reconstruction envisaged by both the Liber Pontificalis and Constantine's inscription.

96 CIL 10.829 = 12.1635 = ILS 5706 = ILLRP 648 (… laconicum et destrictarium faciund(a) et porticus et palaestram reficiunda locarunt).

97 Eschebach, H., Die Stabianer Thermen in Pompeji (Berlin, 1979), esp. 68–9Google Scholar with Taf. 37c; idem, ‘… laconicum et destrictarium faciund… locarunt… Untersuchungen in den Stabianer Thermen zu Pompeji’, MDAI(R) 80 (1973), 235–42 (pace Richardson, L., Pompeii. An Architectural History (Baltimore, 1988), 103–4)Google Scholar.

98 Laconicum: cf. Cic. Att. 4.10.2, Vitr. De Arch. 5.10.5 (with gloss/translation, sudationes); destrictarium: e.g. CIL 8.24106 = ILLRP 1275 (Carpi).

99 ‘Replacement’: CIL 10.5159 = ILS 3784 = ILLRP 562a (Casinum, 40 BC), which is the earliest dated reconstruction inscription collected to use restituere; ‘restoration’: CIL 11.3801 (Veii, under Tiberius: statuam ex ruina templi vexatam … refecit et in publicum restituit), which indicates the distinction between restituere and reficere.

100 CIL 13.6592 = ILS 9184 (Walldürn, … balineu[m] vetustate conla[p]sum restituerunt); CIL 13.6562 (Jagsthausen, … balneum vetustate conlapsum restituerunt).

101 Baatz, cit. (n. 48).

102 Mettler, A., in Fabricius, E., Hettner, F. and von Sarwey, O. (eds), Der Obergermanisch-Raetische Limes des Römerreiches (Heidelberg, 1894–1937)Google Scholar, Lieferung, 32: Jagsthausen, Mainhardt (Heidelberg, 1905), no. 41 (‘Das Kastell Jagsthausen’), 14–7Google Scholar; Heinz, W., Römische Bäder in Baden- Württemberg (Tübingen, 1979), 69–70Google Scholar.

103 On the construction of military barracks, Davidson, D. P., The Barracks of the Roman Army from the First to the Third Centuries AD (Oxford, 1989), 209–30Google Scholar. That the rebuilding even of the stone walls of a fort from the foundations may have been quite a regular practice, especially after a hard winter, is shown by the example of the modern walls of the Saalburg; cf. D. Baatz (n. 74), 378. For antiquity, note the numerous repairs to the walls at Vindolanda, some of them down to the lowest courses: Bidwell, P. T., The Roman Fort of Vindolanda (London, 1985), 38–46Google Scholar.

104 RIB 334 = ILS 537.

105 A rebuilding inscription for a temple of Diana may be contemporary (RIB 316, found in a meadow apparently near the fort).

106 RIB 1706 (… por[tam cum tu]rribus [a] fundamen[tis restitu]erunt, 223/5). Cf. Bidwell, cit. (n. 103), esp. 79.

107 The possibility that there was originally an inscription that commemorated the reconstruction of the whole fort as such is reduced by the fact that none survives from any other site.

108 RIB 1234 = ILS 2618 (portam cum muris vetustate dilapsis a solo rest(ituit), 205/7). Cf. Richmond, I. A., ‘Excavations at High Rochester and Risingham, 1935’, ArchAel4 13 (1936), 188–91Google Scholar; Bruce, J. C., Handbook to the Roman Wall13 (ed. Daniels, C.; Newcastle-upon-Tyne, 1978), 289–94Google Scholar.

109 As was done at Feurs (CIL 13.1642 = ILS 5639: theatrum quod … ligneum posuerat lapideum restituit, Claudius) and Satafis (CIL 8.20266 = 8393: ductum thermarum nuper lignis putrib. constitutum at [nunc] mirabili opere ac pe(t)ra [constr]uctum instituit, perfecit, 379/83).

110 The plan is a novelty of the later second century: cf. Bechert, T., ‘Römische Lagertore und ihre Bauinschriften. Ein Beitrag zur Entwicklung und Datierung kaiserzeitlicher Lagertorgrundrisse von Claudius bis Severus Alexander’, BJ 171 (1971), 260–74Google Scholar.

111 … portam vetustate conlabsam lapidi quadrato arco curvato restituit, referred to by Rebuffat, R., ‘Les centurions de Gholaia’, in L'Africa Romania 2 (Sassari, 1985), 227Google Scholar. For the archaeology, Rebuffat, R., ‘Nouvelles recherches dans le Sud de la Tripolitaine’, CRAI 1972, 330–1Google Scholar; idem, ‘Bu Njem 1971’, LibyaAnt 11/2 (1974/5), 214–5; idem, ‘Bu Njem 1972’, LibyaAnt 13/4 (1976/7), 47–50; cf. Welsby, D., ‘The defences of the Roman forts at Bu Ngem and Gheriat el-Garbia’, in Bidwell, P., Miket, R., and Ford, B. (eds) Portae cum turribus. Studies of Roman Fort Gates (Oxford, 1988), 63–82Google Scholar.

112 Rebuffat, R., ‘Les inscriptions des portes du camp de Bu Njem’, LibyaAnt 9/10 (1972/1973), 99–120Google Scholar.

113 There is no justification for the interpretation of a phrase with vetustate such that, ‘even if it is a stock military phrase, either a break in occupation had occurred or a fairly considerable time must have elapsed’ (Mattingly, D. J., ‘The Roman road-station at Thenadassa (Ain Wif)’, LibStud 13 (1982), 78)Google Scholar.

114 For another acrostic on a rebuilding inscription, this time with Virgilian echoes, cf. the public fountain at Thibilis restored by a Christian called, probably, Felix (ILAlg 2.2.4724, with Marrou, H.-I., ‘Deux inscriptions Chrétiennes’, BAA 3 (1968), 343–8)Google Scholar. This, of course, is a significant name in Reconstruction.

115 Rebuffat, R., ‘Bu Njem 1970’, LibyaAnt 6/7 (1969/1970), 107–21Google Scholar; idem, ‘Bu Njem 1971’, LibyaAnt 11/12 (1974/5), 189–209. Summed up by Rebuffat, , ‘Notes sur le camp romain de Gholaia (Bu Njem)’, LibStud 20 (1989), 157Google Scholar: ‘il [the fort] n'a pas été remanié dans sa structure, s'il a au contraire subi des réparations et des modifications de détail’.

116 Inscription: Rebuffat, R., ‘Bu Njem 1970’, LibyaAnt 6/7 (1969/1970), 141Google Scholar, no. 70–46; archaeology: ibid., 121–33; idem, ‘Bu Njem 1971’, LibyaAnt 11/12 (1974–5), 211–4; idem, ‘Bu Njem 1972’, LibyaAnt 13/4 (1976/7), 44–7.

117 As made by Thornton, M. K. and Thornton, R. L., Julio-Claudian Building Programs: a Quantitative Study in Political Management (New York, 1989), 22Google Scholar, based on understanding the real nature of Latin terms for different acts of construction as percentages of the entirely new facere.

118 CIL 13.6541 = ILS 7098: … signum Minervae restituit.

119 Strocka, V. M., ‘Weihedenkmäler aus Öhringen. II. Die Skulpturen’, Fundberichte aus Schwaben, n.f. 18/1 (1967), 115–31Google Scholar, esp. 130.

120 CIL 14.376 (aedem Veneris restituit); on this inscription and its date, Meiggs, R., Roman Ostia2 (Oxford, 1973), 493–502Google Scholar.

121 Pavolini, C., Ostia (Guida archeologica Laterza, 8; Rome-Bari, 1983), 72Google Scholar; Paribeni, R., ‘I quattro tempietti di Ostia’, MonAL 23 (1914), 441–84Google Scholar. Cf. Zevi, F., ‘Publio Lucilio Gamala “senior” e i “quattro tempietti” di Ostia’, MEFRA 85 (1973), 555–73CrossRefGoogle Scholar.

122 CIL 14.375 (aedem Veneris constituit). For the date (disputed), Zevi. cit. (n. 121); contra Meiggs, cit. (n. 120) (early Augustan). For the verbs here, cf. the case of the Temple of Demeter at Cumae (see below).

123 E.g. the porticoes of the forum at Thuburbo Maius (ILAfr 275: ad melio[rem] cul[tum] … res[titui]t, excoluit, dedicavit, 376/7) were ‘l'objet d'une modification’ (Lézine, A., Architecture romaine d'Afrique (Tunis-Paris, 1963), 131Google Scholar; cf. Merlin, A., Le Forum de Thuburbo Maius, Notes et documents VII (Tunis, 1922))Google Scholar, rather than a substantial alteration.

124 AE 1948.27; for the date, Calza, G., ‘Il santuario della Magna Mater a Ostia’, MPAA 6.2 (1974), 201Google Scholar.

125 Calza, cit. (n. 124), 198–202; Pavolini, cit. (n. 121), 201–2.

126 AE 1948.31 (… edem (sic)vetustatis (sic) collasa (sic) fecerunt), which, given the find-spot, must also belong to the temple, pace Calza, cit. (n. 124), 201.

127 AE 1978.303 (restituendam el ampliandam novis capitibus et rivis).

128 Bartoccini, R., ‘Il rifornimento idrico della Colonia Iulia Felix L. F.’, Autostrade 7/8 (July/ August 1903), 1–16Google Scholar; Jones, G. D. B., ‘Capena and the Ager Campanus’, PBSR 30 (1962), 197–9Google Scholar.

129 Cf. Simoncini, G., ‘Il foro di Lucus Feroniae’, Quaderni del Istituto della Storia dell'Architettura 52/3 (1962), 1–7Google Scholar.

130 For the range of reconstruction terminology, cf. Augustus's laws on the repair of aqueducts (CIL 10.4842, Venafrum copy, line 13: reficere reponere restituere resarcire); for the rare recurare, CIL 9.5804 ([…]rior Vici Cluentensis vetustate dilapsum […]tus Rufinus et Iustus inpendio suo recuraverunt).

131 Cf. the frequency of re- compounds in works of the Augustan age, cumulatively suggesting the reversal of a catastrophic process, esp. Hor. Carm. 4.15.4–12: tua Caesar, aetas … rettulit … restituit … revocavit (‘les mots-clés de le littérature officielle du Principat’, Gros, cit. (n. 29), 25–6).

132 AE 1954.169c, with a consular date (236), but not the official record (cf. Guarducci, M., ‘Alba Fucens. Graffiti nell'antico tempio sul colle di S. Pietro’, NSA 1953, 117–25Google Scholar). Lacunar as a sundial: Vitr. De Arch. 9.8.1.

133 Suet. Dom. 20.

134 CIL 10.6656 = ILS 5702 (… exclusa totius {s}carie vetustatis … r[e]paravi in meliorem civitatis effigiem).

135 CIL 14.4719 (… [r]eparatu[m ad ornatum] urbis).

136 de Ruyt, C., Macellum: marché alimentaire des romains (Louvain, 1983), 115–24Google Scholar; Heres, T. L., Paries—a Proposal for a Dating System of Late Antique Masonry Structures in Rome and Ostia, AD 235–600 (Amsterdam, 1982), 516–8Google Scholar, cat. no. 75.

137 Larrieu, J.-F., Paradoxes archéologiques sur l'évolution de l'architecture religieuse du moyen-âge au xviiie siècle (Paris, 1938), ch. 2, 17–24Google Scholar.

138 E.g. CIL 5.6357 (Laus Pompeia).

139 E.g. Cicero, De Re Publica 5.2.

140 E.g. CIL 2.2420 ( = Vives (n. 49), no. 2079) (Bracara); 8.7957 (Rusicada), 28046 ( = ILAlg 1.2963) (Hr. El-Hamacha); ILGN 15 (Aime-en-Tarantaise).

141 IRT 543.

142 AE 1985.876b (curiam renovatam et exornatam, 152).

143 E.g. CIL 8.17733 = ILAlg 2.2.6309 (Hr. el-Guesseria): [oper]e signin[o] renovari; 25909 (Thignica): renovatis cameris.

144 Boeswillwald, E., Cagnat, R. and Ballu, A., Timgad, une cité africaine sous l'empire romain (1902), 32ffGoogle Scholar. Cf. now Zimmer, G., Locus dato decreto decurionum. Zur Statuenaufstellung zweier Forumsanglagen im römischen Afrika (München, 1989), 50–1Google Scholar.

145 Above (n. 142).

146 On this later change of inscriptional habit, Mitchell, J., ‘Literacy displayed: the use of inscriptions at the monastery of San Vincenzo al Volturno in the early ninth century’ in McKitterick, R. (ed.), The Uses of Literacy in Early Medieval Europe (Cambridge, 1990), 186–225Google Scholar, esp. 214.

147 ILAlg 1.2101 (364), 2102 (366–7), 2108 (407/8).

148 Gsell, S. and Joly, C. A., Mdaourouch (Algiers-Paris, 1922), 93–101Google Scholar.

149 CIL 13.3653 = ILS 4562. An almost nonsensical example would be the ‘ocean begun from the foundations’ (oceanum a fundamentis coeptum) said to have been completed at Abbir Maius in AD 368/70 (AE 1975.873, where the phrase may be a symbolic, polar opposite to soliarem ruina conlapsum).

150 CIL 8.20836 = ILS 638 (ante plurima tempora rebellium incursione captum ac dirutum).

151 Laporte, J.-P., ‘Rapidum: le camp et la ville’, BAntFr 1983, 253–66Google Scholar (who cannot be followed on all points as he mixes inscription and archaeology in an unacceptable manner); Seston, W., ‘Le secteur de Rapidum sur le times de Maurétanie Césarienne après les fouilles de 1927’, MEFRA 45 (1928), 150–83CrossRefGoogle Scholar; Benseddik, N., Les troupes auxiliaries de l'armée romaine en Mauretanie Césarienne sous le Haut Empire (Algiers, 1979), 169–71Google Scholar (sceptical about the truth of a fundamentis, yet still ultimately interpreting it literally). [We have been unable to consult Laporte, J.-P., Rapidum. Le camp de la cohorte des Sardes en Maurétanie Césarienne (Sassari, 1989)Google Scholar.]

152 CIL 10.846 = ILS 6367.

153 Maiuri, A., L'ultima fase edilizia di Pompeii (Spoleto, 1942), 68–70Google Scholar; de Vos, A. and , M., Pompeii, Ercolano, Stabia (Guida archeologica Laterza, 11; Rome-Bari, 1982), 72–8Google Scholar; Richardson, L., Pompeii: an Architectural History (Baltimore, 1988), 281–5Google Scholar; Coarelli, F., la Rocca, E., de Vos, A. and de Vos, M., Pompeji-archäologischer Führer (Bergisch Gladbach, 1990), 209–15Google Scholar.

154 Overbeck, J. and Mau, A., Pompeji in seinen Gebäuden, Alterthümern und Kunstwerken4 (Leipzig, 1884), 104Google Scholar.

155 Chatelain, L., ‘Travaux et recherches du service des antiquités du Maroc’, CRAI 1922, 29Google Scholar.

156 IAM 2.404 = ILAfr 614 (… domum cum balineo vetustate conlapsam a solo restituit, curante M. Ulpio Victore v.e. proc. prolegato).

157 Thouvenot, R., Maisons de Volubilis: le Palais dit de Gordien et la Maison à la mosaïque de Vénus (Rabat, 1958), 9–47Google Scholar.

158 From pottery evidence found in the excavations of the palace, which suggests that the previous house was still occupied at least in the Severan period (Thouvenot, cit. (n. 157), 47 n. 1), Rebuffat, René has argued (‘Le développement urbain de Volubilis au second siècle de notre ère’, BCTH n.s. 1–2 (1965–1966), 235)Google Scholar that the period of its decay must be a minimum of five and a maximum of 50 years and, therefore, would not be consistent with vetustas. However, the latter has here, as in all rebuilding inscriptions which deploy the term, a notional value, which asserts the historical significance of this building project, implies nothing at all about a period of real decay after the construction of the building and cannot be tied even to a minimal reality of just five or ten years.

159 E.g. CIL 8.2546, 2548 (Lambaesis, 172/4 and probably 178/80, mu[nime]nta et turns a solo [refecta]), which seems to have referred to the strengthening of the original Hadrianic walls and the addition of new towers in front of the old gates: cf. Bechert, cit. (n. 110), 262–9, esp. no. 208.

160 E.g. RIB 1234 (Risingham, 205/7), 1706 (Vindolanda, 223).

161 E.g. RIB 334 (Caerleon, c. 255/60).

162 E.g. AE 1971.213 (Caerleon, ? 176/80), RIB 587 (Ribchester, probably 225/35).

163 E.g. AE 1966.606 = IAM 2.824 (Ain-Schkour, 222/32).

164 E.g. RIB 1738 (Great Chesters, 225).

165 E.g. RIB 1281 (High Rochester, 225/35); another, or the same (?), ballistarium was recorded by the terms a solo fecit only about ten years earlier (RIB 1280).

166 IRT 396, with additions (cellam f[rigi]darii et [--- c]ry[ptas? --- ] rui[na con]labsas … a fundamentis [---] marmoribus et co[l]umnis exornavit stat[u]am Aesculapii novam [--- res]tituit ceter[as] refe[c]it).

167 Bartoccini, R., Le terme di Lepcis (Leptis Magna) (Bergamo, 1929)Google Scholar.

168 Bartoccini, cit. (n. 167), 76.

169 IRT 263 (cur(atores) refectionis thermarum tert(iae)).

170 Fant, J. C., ‘IRT 794b and the building history of the Hadrianic baths at Lepcis Magna’, ZPE 75 (1988), 291–4Google Scholar (quotation on 292). The history of the building continued in this vein, so that at an unknown point perhaps in the later third century a reconstruction of the baths was signalled by the grand statement that the builder ‘with foresight and exceptional [diligence(e.g.)] renovated the baths … that had been pulled down to the ground [because of?] old age’ (IRT 315); the archaeological evidence for work in the third or fourth century is extremely limited.

171 Cf. Duval, N., ‘Observations sur l'urbanisme tardif de Sufetula (Tunisie)’, CT 12 (1964), 94Google Scholar; for instance, the baths at Thamusida, on which Rebuffat, R., Hallier, G. and Manon, J., Thamusida II (Paris, 1970), 161–9Google Scholar, esp. table at 168 and figs 26, 27.

172 B. and Galsterer, H., Die römischen Steininschriften aus Köln (Köln, 1975), 47–8Google Scholar no. 188 ( = AE 1953.271 = CIL 13.8262 = ILS 790): [vetustat]e conlabsam [a fundamenti]s ex integro opere faciun[dum --- cura]vil (AD 392/4). [Note, however, that this does not certainly refer to the praetorium and that the building affected by this work was not called praetorium, but rather described by a feminine term, e.g. aedes or domus.]

173 Precht, G., Baugeschichtliche Untersuchungen zum römischen Praetorium in Köln (RA 14; Köln-Bonn, 1973)Google Scholar.

174 CIL II.4809.

175 Pietrangeli, C., Spoletium (Spoleto, 1939), 42–6Google Scholar; Gaggiotti, M. et al. , Umbria, Marche (Guida archeologica Laterza, 4; Rome-Bari, 1980), 105–9Google Scholar; Brands, G., Republikanische Stadttore in Italien (Oxford, 1988), 202–3Google Scholar.

176 CIL 9.2448 ( … a fundamentis tribunal columnatum fecit), AE 1930.120, with Gaggiotti, M., ‘Le iscrizioni della basilica di Saepinum e i Rectores della Provincia del Samnium’, Athenaeum 56 (1978), 145–69Google Scholar.

177 Gaggiotti, cit. (n. 176), 159–60; Sepino. Archeologia e Continuità (Campobasso, 1979), 80–2Google Scholar. In this case the inscription is doubly misleading, since a fundamentis also is not literally true, and thus parallel to the Beth She'an example given above (title note).

178 Durliat, J., Les dedicaces d'ouvrages de défense dans l'Afrique Byzantine (Rome, 1981), 109–11Google Scholar with nos. 4, 6, 8, 12, 13, 14, 16, 17, 19, 20, 21; Pringle, D., The Defence of Byzantine Africa from Justinian to the Arab Conquest (Oxford, 1981), 109–20Google Scholar, esp. 111.

179 Maier, F. G., Griechische Mauerbauinschriften (Heidelberg, 1959), 1, 206–8Google Scholar, no. 58.

180 Oppermann, H., Zeus Panamaros (Giessen, 1924), 20–1Google Scholar, after Holleaux, M., ‘Inscriptions du sanctuaire de Zeus Panamaros’, BCH 28 (1904), 346–8Google Scholar, nos. 2 and 3 (συνσεισθέ[ν] των τών τειχέωνὑπὸ τοῦ σεισμοῦ… ẻπεσκεύασεν πάντα…).

181 Cf. Gast, K., Die zensorischen Bauberichte bei Livius und die römischen Bauinschriften (Göttingen, 1965), 79–80Google Scholar.

182 Cf. Zanker, cit. (n. 89), ch. IV, ‘Das Programm der kulturellen Erneuerung’; Gros, cit. (n. 29), ch. 1, ‘Buts et modalités de la restauration religieuse d'Auguste, envisagée sous son aspect architectural et édilitaire’.

183 For the tetrarchs, see now Waldherr, cit. (n. 7).

184 MacMullen, R., Roman Government's Response to Crisis, AD 235–337 (New Haven, 1976), 43Google Scholar, referring to Amm. Marc. 21.10.8 on Julian's criticism of Constantine; cf. Licinius in Euseb. Vita Constantini 2.5. His view is restricted to the third and fourth centuries, but such an attitude went further back.

185 Cf. Cameron, A., Procopius (London, 1985), 245Google Scholar, for suggestive parallels between restoration of virtues and restoration of buildings.

186 Cic. Sen. 20 (res publicas… labefactatas… restitutas).

187 Vespasian: CIL 6.934; Caracalla: CIL 14.4387 (Ostia, in 207), 2596 (at Tusculum, in 216).

188 Similarly, milestones with the language of repair are used to assert a new emperor's authority (di Vita Evrard, G., ‘Inscriptions routières de Nerva et de Trajan sur l'Appia Pontine’, La Via Appia. Decimo incontro di studio del comitato per l'archeologia Laziale, ed. Gigli, S. Quilici (Rome, 1990), 73–93, esp. 86–7)Google Scholar. For a modern parallel, compare the scheme to rebuild Babylon with thousands of bricks, each inscribed with the Arabic legend ‘rebuilt during the era of the leader, President Saddam Hussein’ (The Independent, Saturday 11 August 1990, 9).

189 AE 1987.839. The combination of Imperial epithet and vetustas is similarly used on a stone found at Aïoun-Sbiba (AE 1957.180), where, likewise, the renaming is made in combination with a restatement of reconstruction.

190 AE 1984.151 (313/24); cf. Cecere, M. G. Granino, ‘Iscrizioni senatorie di Roma e dintorni: Lavinium’, in Epigrafia e ordine senatorio I (Rome, 1982), 662–6Google Scholar. Unfortunately, the baths themselves seem to be virtually unpublished.

191 Compare the placements of the rebuilding inscriptions from the temple of Isis at Pompeii (over the entrance to the court, see n. 153) and the Palace of Gordian at Volubilis (at the entrance to the palace, see n. 154). For epigraphic visibility, Susini, G., The Roman Stonecutter. An Introduction to Latin Epigraphy (ed. Badian, E.; Oxford, 1973), 55–6Google Scholar.

192 CIL 11.2122 = 12.3360 = ILLRP 571: revised as AE 1987.371.

193 AE 1968.86 (375/6); cf. Chastagnol, A., ‘La restauration du temple d'Isis au Portus Romae sous la règne de Gratien’, in Hommages à M. Renard II (Brussels, 1969), 135–44Google Scholar.

194 RIB 141a-c; cf. Cunliffe, B. W., Roman Bath (Oxford, 1969), 29–33Google Scholar with fig. 9.

195 CIL 8.2658 (AD 226) and 2657 + 18105 = ILS 5626 (AD 246/8); cf. Janon, M., ‘Recherches à Lambèse’, AntAfr 7 (1973), 238–9CrossRefGoogle Scholar with fig. 17.

196 Like the aqueducts, whose ‘historic’ names are an interesting parallel to those at Rome and Öhringen, especially the reconstruction in 290/3 after longa vetustate, in reality only eight/seventeen years (CIL 8.2660 = ILS 5787; cf. CIL 8.2661 = ILS 5788): Janon, cit. (n. 195), 241–51.

197 Janon, cit. (n. 195), 239; idem, ‘Lambaesis – Ein Überblick’, AW 8.2 (1977), 15; idem, ‘Recherches à Lambèse III: essais sur le temple d'Esculape’, AntAfr 21 (1985), 36 n. 4 (although perhaps based on the inscriptional language, marmorib. musaeo et omni cultu).

198 S. H. A. Sev. 19.5, 24.3, and on fragments 8a + b of the marble plan, probably as Sept[izo]dium (Almeida, E. Rodriguez, Forma Urbis Marmorea. Aggiornamento generale 1980 (Rome, 1981), 74–5Google Scholar with Tav. 5); the same name is also used of a building at Hr. Bedd (CIL 8.14372 = ILS 5076).

199 Mertens, J., Alba Fucens, 1 (Brussels-Rome, 1969), 69–72Google Scholar, but no longer visible. Cf. the reconstruction inscription in a mosaic belonging to baths in Djemila (AE 1920.33, said to be still in place).

200 AE 1955.137. Leschi, L., ‘Autour de l'amphithéâtre de Lambèse’, LibycaBServAnt 2 (1954), 185–6Google Scholar, with fig.

201 Rebuffat, R., cit., LibyaAnt 6/7 (1969/1970), 141 no. 70–46Google Scholar with Pl. XXXVI.

202 CIL 11.4095 (= ILS 5696); 4097 (= ILS 5697) and 4096.

203 Specific to the character of the building in the case of the building inscription (‘entrusted the pleasure of the winter baths … to better beauty … operating with an increase of new work’, volumptatem [sic] thermarum hiemalium … at meliorem pulcritudinem … cum augmento operi(s) novi ex[er]cientes adsi[g]naverunt et dedicaverunt); more general in the case of the statue-bases (‘to the restorer of the winter baths’, restauratori thermarum hiemalium).

204 Nymphaeum: e.g. Lambaesis, where a number of inscriptions were placed at the Septizonium; cf. Janon, cit. (n. 195), 235; arch: famously, on the ‘Porta Maggiore’ in Rome; town-centure: e.g. Lucus Feroniae (see nn. 128–9).

205 Hence the inclusion of reconstruction statements in Vidman, L. (ed.), Fasti Ostienses2 (Prague, 1982), 45 (AD 94) and 48 (112)Google Scholar.

206 Wiseman, T. P., ‘Monuments and the Roman annalists’, in Moxon, I. S., Smart, J. D. and Woodman, A. J. (ed), Past Perspectives. Studies in Greek and Roman Historical Writing (Cambridge, 1986), 89Google Scholar.

207 Petrucci, A., ‘La scrittura tra ideologia e rappresentazione’ in Zeri, F. (ed.), Storia dell'arte italiana 9.1 (Turin, 1980), 5–123Google Scholar, esp. section 4. ‘Epigrafia urbana per città-monumento’, 32–7.

208 For Suetonius, see n. 29 above; cf. the pseudo-epigraphic form of Tacitus's extravagant parody of Nero's legacy for one of the prime elements of Rome's heritage: Nero frumentum plebis vetustate corruptum in Tiberim iecit, quo securitatem annonae sustentaret (Ann. 15.18).

209 Petron., Sat. 135.3Google Scholar (camellam vetustate ruptam refecit); Apul., Met. 1.14Google Scholar (fores ad pristinum statum integrae resurgunt).

210 CIL 14.4707, with Meiggs, cit. (n. 120), 35–6 and Pavolini, cit. (n. 121), 44–5.

211 E.g. CIL 8.210 = ILS5570) (Cillium, 312/4), 12.1375 (Vaison). 14.4057 (Fidenae).

212 E.g. CIL 11.625 = ILS 5679, Fano: balineum a L. Rufellio Severo factum … quod res publica a novo refecerat, incendio ex maxima parte consumptum restituit; AE 1987.307, Canusium: res publica balineum publicum a Caesidio Proculo refectum vetustate quassatum restituit.

213 As probably with the south gate at Bu Ngem, where the original building inscription of 201 must have been left in place when the reconstruction of 222 was commemorated by another plaque (for literature see nn. 115–6).

214 CIL 10.5159 = ILS3784 = ILLRP 562a.

215 ILAlg 2.496, fragment (b) (Romulus limen posuit iussus et templ[um vetustate] dilabsum restituit dedicavitq. Idib.Oct. Macri[no et] Celso cos.).

216 Tac., Ann. 4.43.4Google Scholar: et Segestani aedem Veneris montem apud Erycum, vetustate dilapsam, restaurari postulavere, nota memorantes de origine eius et laeta Tiberio (‘The people of Segesta too requested that the temple of Venus on Mount Eryx, which had collapsed from age, be restored, mentioning famous facts about its original history and things that pleased Tiberius’).

217 Fronto, Ep. 4.4.

218 CIL 10.5918 = CIL 406 (another record of this project without the ‘historicist’ accent is CIL 10.5917 = ILS 1909). No remains of the baths are preserved: cf. Mazzolani, M., Anagnia (Forma Italiae 1.6) (Rome, 1969), 83–5Google Scholar no. 51.

219 For the literary record of ruinae, e.g. Hor., Epist. 1.11.7–8Google Scholar, Juv. 3.192, 6.56, 7.4, 10.100–7 (pairing Gabii with Fidenae, which also has reconstruction recorded, in CIL 14.4057); Prop. 4.1.34; Luc. 7.392 joins it to Veii, which has produced CIL 11.3780 (a marble altar of 249, with a dedication on the back Victoriae August, sacrum restitutae post antiquissimam vetustatem), 3801 (a statue-base with legend statuam ex ruina templi Martis vexatam … refecit et in publicum restituit), 3810 (in the name of a priestess of Fortuna Redux, the reconstruction a solo of a schola with a consular date) and 7746 (perhaps a Republican reconstruction), and Cora (cf. CIL 10.6526 = ILS 5772, a possibly Augustan repair ex S.C. of a local aqueduct).

220 The inscriptions from Gabii are AE 1982.142 (118/38: [r]uinis vetu[state prostratum restituit]), CIL 14.2804 = ILS 6218 (138/61: vetustate vexatum), and, less dramatically, a statue repair commemorated by CIL 14.2805 = ILS 6221.

221 Almagro-Garbea, M. (ed.), El Santuario de Juno en Gabii (Rome, 1982), 226–7Google Scholar, no. 16 ( = AE 1982.142). The restorations seem mainly to have concerned, not the temple itself, but the surrounding porticoes and the tabernae, as indicated by the brickstamps (Almagro-Garbea (ed.), 197–221, 599–624).

222 For a survey, see Boatwright, M. T., ‘Hadrian and Italian cities’, Chiron 19 (1989), 235–71Google Scholar.

223 CIL 10.5649; cf. Boatwright, cit. (n. 222), 266.

224 Compare, for example, the distinctively early (opus quadratum or reticulatum) appearance of the city-walls at Saepinum (Sepino. Archeologia e continuità (Campobasso, 1979), 44–56)Google Scholar, Telesia (Quilici, L., ‘Telesia’, Quaderni di Topografia Antica 2 (1966), 85–101Google Scholar; Brands, cit. (n. 175), 206–8) and Aesernia (Drago, C., ‘Archeologia iserniana’, Samnium 6 (1953), 59ffGoogle Scholar.; Valente, F., Isernia: origine e crescità di una città (Campobasso, 1982), 65ffGoogle Scholar; Pasqualini, A., ‘Isernia’, Quaderni di Topografia Antica 2 (1966), 79–84Google Scholar: badly preserved but visible parts are early) with the inscriptions of Fabius Maximus in AD 352/7 (CIL 9.2449, Saepinum; 2639 = ILS 1248, Aesernia: both moenium publicorum restauratori) or T. Fabius Saeverus, probably third century (CIL 9.2238 = ILS 5507, Telesia: omnia opera publica restaurata sunt). Moenia, of course, does not automatically refer to walls per se in this period, but rather to public monuments generally; however, the character of the city-walls may reasonably be taken as an indication of building activity generally.

225 Gaggiotti, cit. (n. 176), 167–8 lists his reconstruction inscriptions. On his political importance, cf. Camodeca, G., ‘Fabius Maximus e la creazione della provincia del Samnium’, AAN 82 (1971), 249–64Google Scholar. and Russi, A., ‘L'amministrazione del Samnium nel IV e V sec. d.C.’, in Terza Miscellanea Greca e Romana (Rome, 1971), 306–46Google Scholar. At around the same time, note the similar records of rebuilding by the governor Autonius Iustinianus (CIL 9.2638, Aesernia: macellum terrae motibus lapsum … fieri curavit; 4858, Venafrum: statum iam conlapsum… reparavit).

226 Hieron, . Chron. (ed. Helm, ), 236Google Scholar; CIL 9.2338 (Allifae), 2638 (Aesernia), AE 1972.150 (Telesia).

227 Saepinum: theatre, minor changes to scaenae frons (Sepino (n. 224), 67), south side of forum, remodelling of one room (ibid., 97–100), basilica, larger restorations, but note the exaggeration of fabricavit and a fundamentis in the inscriptions (Saepinum. Museo documentario dell' Altilia (1982), 127–41; Sepino, 76–82); Iuvanum: traces of undated earthquake (E. Fabricotti, ‘Montenerodomo (Chieti) – Relazione preliminare sulle campagne di scavo 1980 e 1981 in località S. Maria di Palazzo (antica Juvanum)’, NSA 1981, 149), but no clear signs of fourth-century work in the centre of the town, although pottery found there suggests continued occupation in this period (Candeloro, A., ‘Alcuni aspetti della zona sud del foro’, in Fabricotti, E. (ed.), Iuvanum. Atti del Convegno di Studi (Chieti, 1990), 95–9Google Scholar; M. Aromatorio, ‘I materiali rinvenuti nel fognolo a sud del foro’, in ibid., 105–18).

228 See nn. 223 and 226 for literature, and, for Iuvanum, Pellegrino, A., ‘Il Sannio Carricino dall'età Sannitica alla romanizzazione’, ArchClass 36 (1984), 171–91Google Scholar and Fabricotti, E. et al. , ‘Iuvanum: scavi e ricerche 1980–83’, in Malone, C. and Stoddart, S. (eds), Papers in Italian Archaeology IV, 4 (Oxford, 1985), 119–29, 142–9Google Scholar.

229 CIL 9.2212 = ILS 5690. (A second rebuilding inscription, AE 1972.150 actually reads [a fundamen]t[is]: cf. Camodeca, cit. (n. 225), 249–51, with the photograph, Tav. I.)

230 Quilici, cit. (n. 224), 98 with n. 44; de Caro, S. and Greco, A., Campania (Guida archeologica Laterza, 10; Rome-Bari, 1981), 198Google Scholar.

231 Mielsch, H., Römische Stuckreliefs (Heidelberg, 1975), 63, 147Google Scholar, K. 56.

232 CIL 9.2238 = ILS 5507 (cura et sollicitudine omnia oepra publica restaurata sunt).

233 Quilici, cit. (n. 224), 85–101.

234 Plutarch, , Pericles 13.5Google Scholar (κάλλει μὲν γὰρ ἕκαστον εὐθὺς ἦν τοτ᾿ ἀρχαῖον, ἀκμῇ δὲ μέχρι νῦν πρόσφατόν ὲστι καὶ νεουργόν· οὕτως ὲπανθεῖ καινότης ὰεί τις, ἄθυστον ὑπὸ χρόνου διατηροῦσα τὴν ὄψιν, ὥσπερ ἀειθαλὲς πνεῦμα καὶ ψυχὴν ἀγήρω καταμεμειγμένην τῶν ἔργων ἐχόντων).

235 Pristinam: e.g. AE 1968.157 (Pinna, 213), CIL 8.2656 (Lambaesis, 364/7), 24044 (Hr. Ben Hassen, 383/92), EphEp 8.456 = ILS 5693 (Liternum, fourth century); novam: AE 1920.15 (Djemila, 295); cf. ad pulchriorem faciem: CIL 8.26472 (Dougga, 293/305), in novitatis [faciem]: ILAlg 1.2107 (Madauros, 399/400).

236 Despite the claim of diligens instantia being applied to tabulari[a a]ntiquissima, the curator might finally say that he ad omnem [s]plendorem ins[tit]uit el dedicavit (CIL 8.27817 = ILS 5557, SidiAhmed-el-Hachemi, 368/70).

237 E.g. AE 1928.39 (Setif, 298, inchoatam perfectamque feliciter), 1987.848 (Tibiscum, Dacia, c. 202, restituit fel[iciter]).

238 Ballu, A., ‘Rapport sur les travaux de fouilles exécutés en 1906 par le service des monuments historiques en Algérie. Deuxième partie: fouilles de Timgad’, BCTH 1907, 262–3Google Scholar (probably fourth century, ruinis iam diu informib(us) tristem … felicius quam condita est restituit). The house is badly published, but the most substantial sign of actual reconstruction is the addition of lavatories! (Ballu, A., Les ruines de Timgad, antique Thamugadi: sept années de découvertes, 1903–10 (Paris, 1911), 72–6.Google Scholar) Similar expression at CIL 12.138 (Sitten, 377): augustas aedis … restituit longe praestantius illis quae priscae steterant.

239 Cassiod, . Var. 4.51.2Google Scholar, on the ideal of being both antiquorum diligentissimus imitator and modernorum nobilissimus institutor. Although the vocabulary shows a slight conceptual shift, this composite ideal is contrived to stand squarely within the classical Roman tradition.

240 Cf. CIL 9.3885 = ILS 4024 (murum vet[ustate] consumtum a solo resti[tuer.] ex p. p. Angitiae), perhaps the ceremonious expression of what was probably only the detailed repair of the circuit-wall of a sanctuary of the goddess Angitia (cf. several remains of polygonal circuit-walls with small reticulate additions: Grossi, G., La città di Angizia, il ‘Lucus Angitiae’ (n.p., 1981)Google Scholar, who, however, on little evidence, imagines that this was the site of a whole city, rather than a sanctuary).

241 CIL 10.3685 = ILS 4040; but the identification of the aedes et porticus of the inscription with the ‘tempio con portico’ excavated in 1970/1 is unsure.

242 Bertoldi, M. E., ‘Recenti scavi e scoperte a Cuma’, BA5 58 (1973), 38–42Google Scholar.

243 The inscriptions recording reconstruction from the foundations do not refer to any such real operations, but rather ‘reflect the ongoing process of adaptation and renovation from existing domus ecclesiae’ (White, L. M., Building God's House in the Roman World. Architectural Adaptation among Pagans, Jews and Christians (Baltimore, 1990), 128Google Scholar, with examples).

244 La Regina, A., ‘Ricerche sugli insediamenti vestini’, MAL8 13 (1968), 388–91Google Scholar: [--- aed]em Quirini vetu[state dilapsam ---]. Only one period is identifiable from the partially preserved structure, the podium of perhaps Flavian date. Hence the statement that ‘il rifacimento appare totale’. The inscription can only be read upside down around a pilaster of the present church.

245 CIL 14.3543 = ILS 3452 = InscrIt 4.1.48 (dated by the day and the suffect consuls of 79, with reinforcement a solo restituit). The cult is otherwise restricted to the military quarries of the Rhine and Moselle region, making its first appearances there too in the Flavian period; cf. Robert, P.-Ch., ‘Inscriptions laissées dans une carrière de la haute Moselle par des légions romaines’ in Mélanges Graux (Paris, 1884), 329–40Google Scholar; also PW 20, 266.54–307.39 [Keune] s.v. Saxanus.

246 Cf. North, J., ‘Conservatism and change in Roman religion’, PBSR 44 (1976), 1–12Google Scholar.

247 E.g. Julian, the reparator Romani orbis (CIL 9.417), restoring, for instance, a colossus in the names of Honos and Virtus (ILAlg 1.1229); also, the last Pagan revival of Eugenius, to which the reconstruction of the cella Herculis probably claimed by the fragment AE 1941.66 = 1948.127 (393/4) must belong: cf. Bloch, H., ‘A new document of the last Pagan revival in the West, 393–394 AD’, HThR 38 (1945), 199–244Google Scholar and ‘The Pagan revival in the West at the end of the fourth century’, in Momigliano, A. (ed.), The Conflict between Paganism and Christianity in the Fourth Century (Oxford, 1963), 193–218Google Scholar. The reconstruction consisted of the partial rebuilding in brick of the southern wall of the cella: Heres, cit. (n. 136), 422–4 n. 57; Pavolini, cit (n. 121), 115–6.

248 E.g. IRT 869 (Thenadassa, second or third century), with Mattingly (n. 113), 73–80: balneum v[etustate corrup]tum restituendum [curavit], eidem assam cellam a so[lo] fecit et cylisterium instituit; ILAfr 285 ( Thuburbo Maius, late Empire), with Merlin, A., ‘Fouilles à Thuburbo Maius en 1916’, CRAI 1917, 71–4Google Scholar, and further additions: [cellam s]oliarem … refec[it atque pra]esolidavit, piscinam novam nomine cochleam … adiecit [atque] dedicavit.

249 … montem Vaticanum vetustate conlabsum restituerunt (CIL 13.7281 = ILS 3805).

250 Palmer, R. E. A., ‘The neighborhood of Sullan Bellona at the Colline Gate’, MEFRA 87.2 (1975), 662–3Google Scholar. Fishwick, D., ‘Hastiferi’, JRS 57 (1967), 144–5Google Scholar, esp. no. 25, suggests that it was a temple, not an earthmound. Bellona: e.g. the temple in Ostia (above, 157); cf. the Claudii's responsibility for the reconstruction of the temple in Rome (Pliny, HN 35.12Google Scholar).

251 de Buruaga, J. Alvarez Saenz, ‘Observaciones sobre el teatro Romano de Mérida’, in Actas del Simposio: El teatro en la Hispania romana (Badajoz, 1982), 303ffGoogle Scholar; de la Barrera Anton, J. L., Los capitales romanos de Mérida (Badajoz, 1984), 82–90Google Scholar; Vallois, R., ‘Observations sur le théâtre romain de Mérida’, RFA 21 (1919), 193–209Google Scholar.

252 Alvarez, J. Menéndez-Pidal, ‘Algunas notas sobre la restauracion y atencion prestadas a los monumentos Emeritenses: il teatro romano’, in Augusta Emerita. Actas del bimilenario de Mérida (Madrid, 1976), 207–11Google Scholar.

253 ILER 2057 = AE 1915.33 = AE 1935.4 (… [indignam arbitrati ruinam operis tam an]tiqui … o[rnatu me]liori quam fuerat [adiecto restitui iusserunt]).

254 E.g. CIL 10.5918 = ILS 406, in historic Anagnia; also, in pristinum statum: e.g. CIL 8.21665 = ILS 4501, at Albulae. With Ulpian's definition of the plain reficere as quod corruptum est in pristinum statum restaurare (Dig. 43.21.1.6), compare the professed intention of the Camden Society in the first issue of its journal, The Ecclesiologist, ‘to recover the original appearance which has been lost by decay, accident or ill-judged attention’.

255 Kähler, H., Römische Gebälke II, 1-Die Gebälke des Konstantinbogens (Heidelberg, 1953)Google Scholar.

256 E.g. a Hellenistic painter's attempt to recreate an ‘archaic’ looking metope for the reconstructed temple at Thermos after its destruction by Philip V of Macedon in 218/206 BC (Stucky, R. A., ‘Die Tonmetope mit den drei sitzenden Frauen von Thermos: ein Dokument hellenisticher Denkmalpflege’, AK 31 (1988), 71–8)Google Scholar; or the big sixth-century altar of Hera at Samos reconstructed from the foundations, perhaps in the first century BC when already under Roman domination, but in marble instead of Poros limestone and adding some new parts in contemporary style (Schleif, H., ‘Der grosse Altar der Hera von Samos’, MIDAI(A) 58 (1933), 174–210)Google Scholar. (We owe these references to Ulf Buchert, Berlin.)

257 Dellheim, C., The Face of the Past. The Preservation of the Medieval Inheritance in Victorian England (Cambridge, 1982), 83Google Scholar.