² Attempts to make sense of the descriptions run into considerable difficulties: see Maiuri, A., La Cena di Trimalchione di Petronio Arbitro (Napoli 1945) 153–8Google Scholar; Bagnani, G., ‘The house of Trimalchio’, AJP 75 (1954) 16–39Google Scholar; McKay, A. G., Houses, Villas and Palaces in the Roman World (1975) 113–14Google Scholar.

³ Well discussed by Wiseman, T. P., ‘Conspicui postes tectaque digna deo; the public image of aristocratic and imperial houses in the late Republic and early Empire’ in L'Urbs. Espace urbain et histoire (Collection de l'école Française de Rome vol. 98, 1987) 393–413Google Scholar. For a good discussion of the role of the patron in dictating styles of decoration, Leach, E. W., ‘Patrons, painters, and patterns: the anonymity of Romano-Campanian painting and the transition from the second to the third style’, in Literary and Artistic Patronage in Ancient Rome, ed. Gold, B. K. (1982) 135–73Google Scholar.

⁴ Standard treatments, notably Friedländer, L., Sittengeschichte Roms¹⁰ (1922) ii, 330–49Google Scholar [= Roman Life and Manners ii, 185–202]Google Scholar persist in using these passages as sources of information rather than windows on Roman ideology.

⁵ Cato, agr. 3. 1.Google Scholar; cf. ORF 174 and 185; Nepos, Atticus 13. 1Google Scholar; Pliny, Pan. 51. 1Google Scholar; and in similar vein Varro, RR i. 13. 6Google Scholar; Columella i. 4. 8; Cicero, Pis. 48Google Scholar; Tac., Ann. 3. 37Google Scholar; Juvenal 14. 66.

⁶ Cf. Cato, in ORF 185Google Scholar; Varro, RR i. 59. 2Google Scholar; cf. i. 2. 10 for Lucullus; Pliny, NH xxxv. 118Google Scholar; cf. my remarks at JRS 73 (1983) 182Google Scholar.

⁷ Ibid. The same condemnation in Varro, RR i. 13. 7Google Scholar, ‘villis pessimo publico aedificatis’.

⁸ Illuminatingly discussed by Zanker, P., ‘Die Villa als Vorbild des späten pompejanischen Wohngeschmacks’, JdI 94 (1979) 480–523Google Scholar.

⁹ Pliny, NH xxxvi. 110Google Scholar, clearly drawing on Varro, or, more probably, Cornelius Nepos, both of whom were much concerned with the phenomenon of ‘luxury’.

¹⁰ Tac., Ann. 3. 55Google Scholar: “ut quisque opibus, domo, paratu speciosus per nomen et clientelas inlustrior habebatur’.

¹¹ Firmly grasped by Thébert, Y., ‘Private life and domestic architecture in Roman Africa’, in A History of Private Life. 1. From Pagan Rome to Byzantium (ed. Veyne, P., trans. 1987), 313–409Google Scholar, an excellent discussion which reached me only after drafting the present text.

¹² For the closure of the doors as an exceptional gesture of mourning see Valerius Maximus v. 7. ext. 1; Seneca, , Cons, ad Liv. 183Google Scholar; Vit. Beat. 28, 1; Brev. Vit. 20. 3; Cons. ad Polyb. 14. 2; Lucan 2. 22; Tacitus, Ann. 2. 82Google Scholar; Hist. i. 62. (I am grateful to Richard Saller for these references.) On the Gallic sack, Livy v. 41. 7, ‘plebis aedificiis obseratis, patentibus atriis principum’. On Livius Drusus' murder, Vell. Pat. ii. 14. 1.

¹³ Cf. Vitruvius vi. 5. 3, with the comments of Carandini, in Settefınestre … 1* La villa nel suo insieme 119Google Scholar and in Carandini, A., Ricci, A., de Vos, M., Filosofiana. The villa of Piazza Armerina (1982), 30 and 58 fGoogle Scholar.

¹⁴ Cf. Carandini, , Settefinestre 1* p. 107Google Scholar.

¹⁵ Ad Att. v. 2. 2; see D'Arms, J. H., The Romans on the Bay of Naples (1970) 48 ffGoogle Scholar.

¹⁶ See the evidence collected by Friedländer, , Sittengeschichte¹⁰ i. 343 ffGoogle Scholar. [ = Life and Manners i, 287 ffGoogle Scholar.], and D'Arms loc. cit.

¹⁷ Veblen, Thorstein, The Theory of the Leisure Class (1899)Google Scholar. Against Veblen's ‘conspicuous consumption’, see Elias, N., The Court Society (1983) 66 ffGoogle Scholar; Douglas, M. and Isherwood, B., The World of Goods. Towards an anthropology of consumption (1980) 3 ffGoogle Scholar.

¹⁸ The poverty of classical Athenian domestic (as opposed to public) building is striking: cf. Walker, S. in Images of Women in Antiquity ed. Cameron, A. and Kuhrt, A. (1983), 82–3Google Scholar; and for classical Olynthos, , Excavations at Olynthos VIII. The Hellenic House (D. M. Robinson and J. W. Graham 1938)Google Scholar. Houses of the hellenistic period were somewhat more impressive: for those of Delos see the publication of the Ilôt de la Maison des Comédiens in Explorations à Délos XXVII (1970)Google Scholar; for Pergamum, , Altertümer von Pergamon XIV. Peristylhäuser westlich der unteren Agora (1984)Google Scholar; for Priene, T. Wiegand and Schrader, H., Priene (1904) 285–300Google Scholar. Greek houses of the imperial period are another matter: see esp. Forschungen in Ephesos VIII. 1.2 (1977)Google Scholar on the Hanghäuser of Ephesos and Levi, Doro, Antioch Mosaic Pavements (1945)Google Scholar for the rich suburb of Daphne. A synthesis on Greek domestic architecture is badly needed. For the Roman material, see McKay, A. G., Houses, Villas and Palaces in the Roman World (1975)Google Scholar: he does not discuss Greek Houses.

¹⁹ For a later period, see the fine use by Thébert (op. cit. n. 11) of African literature to illuminate African domestic architecture.

²⁰ There are signs of attempts to ‘recontextualise’ the objects studied, particularly sculpture, notably by Dwyer, E. J., Pompeian Domestic Sculpture. A study of five Pompeian houses and their contents (1982)Google Scholar, and Zanker, P. in Pompei 79, Raccolta di studi per il deamonono centenario dell'eruzione vesuviana a cura di Fausto Zevi (1979, reissued 1984) 201–10Google Scholar. For mural decoration, see the works of Strocka and Barbet cited below.

²¹ The classic modern studies of chronological development are: Beyen, H. G., Die Pompejanische Wanddekoration vom zweiten bis zum vierten Stil vol. I (1939), II (1960)Google Scholar on the ‘second style’; Bastet, F. L. and de Vos, M., Proposta per una classificazione del terzo stile pompeiano (Arch. Stud. Nederlands Instituut te Rome, IV, 1979)Google Scholar for the ‘third style’; and now Laidlaw, A., The First Style in Pompeii: Painting and Architecture (1985)Google Scholar. But even those apart, the volume of discussion on the chronology of the styles is immense: see the surveys of Ling, R., ‘Pompeii and Herculaneum: recent research and future prospects’ in Papers in Italian Archaeology I (1978) part ii, 153–74Google Scholar; Mielsch, H., in Aufstieg und Niedergang der römischen Welt part 2, vol. 12 pt 2 (1981), 157–264, esp. 170–83Google Scholar; and recently Barbet, A., La peinture murale romaine (1985)Google Scholar.

²² This criticism does not apply to the pioneering work of Mau, A., Geschichte der decorativen Wandmalerei in Pompeji (1882)Google Scholar which was solidly based on archaeological and structural relationships. For similar criticism, see F. Coarelli's avant-propos to Barbet, A., La peinture murale romaine (1985) 7–8Google Scholar.

²³ Cf. Bastet and de Vos. op. cit. who observe (p. 100) that a typology of the fourth style can and should be very different, and suggest analysis by room types.

²⁴ Häuser in Pompeji (ed. Strocka, V. M.) band I. Casa del Principe di Napoli (VI 15, 7. 8) (1984)Google Scholar; cf. my review in Antiquaries Journal 66. 2 (1986) 433–4CrossRefGoogle Scholar.

²⁵ Strocka op. cit. esp. 39–48.

²⁶ Barbet, A., La peinture murale romaine. Les styles décoratifs pompéiens (1985)Google Scholar.

²⁷ Esp. pp. 57–77, 123–39, 193–214 on the theme of ‘adéquation du décor aux locaux’ in the successive styles.

²⁸ I have found particular value in studies concerned with the distinctions of public and private space: Bourdieu, P., ‘The Berber house’, in Rules and Meanings, ed Douglas, M. (1973) 98–110Google Scholar is a classic example of anthropological method. Girouard, M., Life in the English Country House (1978)Google Scholar is a justly famous essay on the sociology of the English house. Elias, N., The Court Society (1983Google Scholar, translation of Die höfische Gesellschaft 1969) 41–65Google Scholar is illuminating on the sociology of early modern French architecture. Daunton, M. J., ‘Public space and private space. The Victorian city and the working-class household’, in Fraser, D. and Sutcliffe, A., The Pursuit of Urban History (1983) 212–33Google Scholar has excellent observations on the redefinition of the boundaries of public and private in the nineteenth century.

²⁹ See the excellent discussion of Susan Walker, op.cit. (n. 18). This and Bourdieu's essay on male and female in the Berber house (above, n. 28) provide a model of where to look for gender distinctions of space; in the light of their work, attempts to seek similar distinctions in the Roman house (below) look unconvincing.

³⁰ Nepos, , praefatio 6–8Google Scholar. The context gives the passage peculiar weight: it is at the outset of a series of biographies of Greeks to explain to a Roman audience the fundamental differences between Greek and Roman society. Cf. Cicero, In Verrem ii. 66Google Scholar for an incident where Roman insensitivity to Greek segregation at table causes outrage in Sicily.

³¹ Vitruvius vi. 7. 2–4. Note that Vitruvius appears to think that segregation served to protect the men from the women, not vice versa, so strange is segregation to him. His remarks on the tranference of the Greek loan-word andron are also significant (4–5): in Latin the word applies to a corridor because there are no exclusive ‘men's rooms’.

³² Maiuri, A., ‘Ginecco e “hospitium” nella casa pompeiana’, Mem. Acc. Lincei ser. VIII, vol. 5 (1954) 451–61Google Scholar identifies various secluded areas as gynaecea, but without any cogent argument. Gynaeceum∣-ium∣-onitis are scarcely used before the fourth century A.D. when they come to refer to an imperial weaving room. Uses of the word in the context of comedy, particularly Plautus, Mostellaria 755 ffGoogle Scholar. (cited in support by e.g. Carandini, , Settefinestre 1* p. 120Google Scholar), are shown by their isolation to be a case of translation from Greek originals. The only passage in classical Latin that suggests a gynaeceum in a Roman house is Cicero, Phil. 2.95Google Scholar, where Cicero decries a corrupt deal fixed between Antony and Deiotarus ‘in gynaeceo’ i.e. through Fulvia. The usage is explained by the desire to defame Antony. Against Lawrence Richardson's attempts to identify separate women's dining-rooms, cf. below n. 147.

³³ Ariès, P., Centuries of Childhood (trans 1962) esp. 385 ffGoogle Scholar. on the eighteenth century emergence of the family as a private and differentiated household unit. Girouard, M., Life in the English Country House, 286 fGoogle Scholar. on the Victorian novelty of children's rooms.

³⁴ E.g., vividly, Tacitus, Dialogus 28–9Google Scholar on sleeping; on play, Lucretius 4. 400 f. (a reference owed to Peter Wiseman) and Virgil, , Aeneid 7. 378 ffGoogle Scholar.

³⁵ Tac., Ann. 15. 54 etc.Google Scholar

³⁶ ‘Forensibus et disertis’. ‘Professors of rhetoric’ (Granger in the Loeb translation) is surely wrong.

³⁷ Wiseman, T. P., ‘Pete nobiles amicos: poets and patrons in late republican Rome’, in Gold, B. K. (ed), Literary and Artistic Patronage in Ancient Rome (1982) 28–49Google Scholar.

³⁸ Kroll, W., Die Kultur der ciceronischen Zeit (1933) 187–90Google Scholar; Mau, A. (tr. Kelsey, F. W.), Pompeii. Its Life and Art (1902) 248–58Google Scholar, etc.

³⁹ My account is dependent on N. Elias (above, n. 28). On the importance of clientèle in French society, see now Kettering, S., Patrons, Brokers, and Clients in Seventeenth Century France (1986)Google Scholar.

⁴⁰ Study of such houses is still inadequate: see Packer, J., ‘Middle and lower class housing in Pompeii and Herculaneum: a preliminary survey, in Neue Forschungen in Pompeji (ed. Andreae, B. and Kyrieleis, H., 1975) 133–46Google Scholar; also ‘Inns at Pompeii: a short survey’, Cronache Pompeiane 4 (1978) 5–53Google Scholar, noting the lack of regular pattern of inns (e.g. p. 30); Hoffmann, A., ‘L'architettura privata’, in Pompei 79. Raccolta di studi … a cura di Fausto Zevi (1979) 105–18Google Scholar; also important is Maiuri, A., Ercolano (1958) vol. i, 407 ffGoogle Scholar. because of its ranking of houses according to inferred social standing.

⁴¹ Notably P. Zanker, ‘Die Villa als Vorbild’ (op. cit. n. 8); also note Maiuri, , Ercolano vol. i, 243–79Google Scholar ‘Case del ceto medio nello schema della domus’.

⁴² On the importance of social mobility in this respect, compare the remarks of U. T. Bezerra de Meneses on Delian society, Dialoghi di Archeologia n.s. 2 (1984) 86Google Scholar.

⁴³ Carettoni, G., Das Haus des Augustus auf dem Palatin (1983), 9Google Scholar; cf. Coarelli, F., Roma (Guide Laterze 1980), 132–3Google Scholar; cf. my criticisms in JRS 75 (1985) 247–8Google Scholar.

⁴⁴ Reception of friends in cubiculo: e.g. Tacitus, Dialogus 3. 1CrossRefGoogle Scholar and 14. 1, Maternus conducts whole dialogue in bedroom; Seneca, , de ira iii. 8. 6Google Scholar, Caelius dines with client; Pliny, ep. v. 3. 11Google Scholar, Pliny recites verse. Suet., Vesp. 21Google Scholar describing Vespasian's daily routine (sees secretaries then receives amici while putting on shoes) implies reception in cubiculo. Conducting of business in cubiculo: Cicero., Verr. iii. 133Google Scholar etc. Verres conducts trials; ad.Q.f. i. 1.25, brother as governor praised for accessibility of cubiculum; Pliny, ep. v. 1. 5Google Scholar, Pliny summons private consilium. For imperial trials intra cubiculum, e.g. Seneca, de clem. i. 9Google Scholar (Augustus), Tac., Ann. 11.2Google Scholar (Claudius) etc.; abolished by Nero, , Tac., Ann. 13. 4Google Scholar; Pliny, , Pan. 49, 1Google Scholar and 83, 1 contrasting Domitian's ‘lair’ with Trajan's open cubiculum. Cf. Tamm, B., Auditorium und Palatium 113–9Google Scholar; Crook, J. A., Consilium Principis (1955) 106–9Google Scholar.

⁴⁵ I hope to pursue this question elsewhere.

⁴⁶ Cf. Coarelli, F., ‘Architettura sacra e architettura privata nella tarda repubblica’, in Architecture et Société (Coll. Ec. Franc. Rome 66, 1983) 191–217Google Scholar.

⁴⁷ Maiuri, , Ercolano 286–90Google Scholar, repeating ‘Oecus aegyptius’ in Studies presented to D. M. Robinson ed. Mylonas, G. E. (1951) 423–9Google Scholar. The basilica-form and its significance are recognised by Tamm, B., Auditorium und Palatium 145Google Scholar; but the label ‘dining-room’ persists, e.g. McKay, , Houses, Villas and Palaces 51Google Scholar.

⁴⁸ See the good discussion by Tamm, , Auditorium und Palatium 132–47Google Scholar.

⁴⁹ The subsidiary pediment above the room in the Casa del Menandro may (Roger Ling warns me) be falsely reconstructed; there is also a major pediment above the tablinum. Note too the pediment of the ‘Rhodian’ peristyle in the C. degli Amorini Dorati (VI 16. 7). Other large rooms with imposing black-ground decoration like that in the Casa dei Cervi should be compared: the ‘Salone Nero’ in the house of that name at Herculaneum (VI. 13) or the central room of the C. di Fabio Rufo at Pompeii (Ins. Oca).

⁵⁰ T.L.L. VI, 320 s.v. fastigium; esp. Vitruvius v. 6. 9, ‘columnis et fastigiis et signis reliquisque regalibus rebus’Google Scholar; Suet., Cal. 37Google Scholar for the fastigium of a basilica.

⁵¹ On the extraordinary honour of the fastigium and its divine connotations, see Weinstock, S., Divus Julius (1971) 280–1Google Scholar. Suet., Jul. 81Google Scholar for the dream. (In Caesar's case, the fastigium was external not internal to the house.) The psychological impact of the fastigiate facade is underlined by comparison with the villas of the slave-plantation owners of the American South, illustrated in Carandini, , Settefinestre vol. i. pl. 177, 179, 181Google Scholar. As Carandini comments, the use of the language of classical forms to legitimate the hierarchies of a slave-owning society is conscious (187 f).

⁵² Of course Greek houses too may occasionally have had pediments, see e.g. the purely hypothetical reconstruction in Wiegand, and Schrader, , Priene 286Google Scholar; that has no bearing on the significance the feature had for the Romans.

⁵³ Cf. Vitruvius v. 6. 9 (above n. 50), linking columns with pediments and statues as the apparatus of a regal stage setting. Note too the force of poetic evocations of grandeur: Verg., Aen. 7. 170Google Scholar ‘tectum augustum, ingens, centum sublime columnis; Prop. 3.2.9 ‘non Taenareis domus est mihi fulta columnis’; Ovid., Met. 2. 1Google Scholar ‘regia Solis erat sublimibus alta columnis’; Stat., Silv. 1.2.Google Scholar 147/152 ‘digna deae sedes … innumeris fastigia nixa columnis’.

⁵⁴ Pliny, NH xvii. 1.6Google Scholar, cf. Val. Max. ix. 1.4. Similarly Pliny, NH xxxvi. 2. 5Google Scholar describes Scaurus dragging marble columns past the terracotta pediments of the gods; see the discussion of Gros, P., Architecture et Société (1978) 65 fGoogle Scholar. Pliny exaggerates: there was a marble temple in Rome shortly after 146 B.C. Cf. also NH xxxvi. 60Google Scholar on the onyx columns decorating the dining-room of Callistus.

⁵⁵ So Mau-Kelsey, , Pompeii: its life and art 245 ff.Google Scholar, vigorously contested by Sulze, H., RE Supp. VII (1940), 950–71Google Scholar (s.v. Peristylium). For Maiuri's traditional views, see his ‘Portico e peristilio’, Parola del Passato 1 (1946) 306–22Google Scholar. See too McKay, , Houses, Villas and Palaces 34–5Google Scholar against the traditional view.

⁵⁶ Well demonstrated by Sulze 951–4. Columns are attested in Greek houses as early as the fifth century, cf. Jones, J. E., Sackett, L. H. and Graham, A. J., ‘The Dema house in Attica’, ABSA lvii (1962) 75 ff. esp. 107 n. 70Google Scholar. For columns as a sign of splendour in a private house, cf. Aristophanes, , Clouds 815Google Scholar (reference owed to Peter Wiseman).

⁵⁷ So Gros, P., Architecture et Société 26Google Scholar, pointing out that the peristyles of the C. del Fauno, in contrast to Greek domestic courts, lack surrounding rooms.

⁵⁸ Cic., ad Att. i. 6Google Scholar; i. 10, etc; cf. Varro, RR iiGoogle Scholar pref. ‘gymnasia urbana’; for other passages, see Sulze 966.

⁵⁹ Dwyer, E.J., Pompeian Domestic Sculpture (1982), 117 and 125Google Scholar makes too little of this aspect of the use of statuary.

⁶⁰ Grimal, P., Les Jardins Romains (1943) 76Google Scholar notes gymnasia and particularly philosophical academies as part of the background for the Roman peristyle; 226 f. stresses the public nature of the peristyle garden. Ridgeway, B. S., ‘Greek antecedents of garden sculpture’ in Ancient Roman Gardens (Dumbarton Oaks Colloquium on the History of Landscape Architecture VII, ed. MacDougall, E. B., Jashemski, W. F., 1981) 9–28Google Scholar on the lack of evidence for Greek domestic gardens or garden sculpture.

⁶¹ Cic., de leg. iii. 31Google Scholar. On this see Pape, M., Griechische Kunstwerke aus Kriegesbeute und ihre öffentliche Aufstellung in Rom (Diss. Hamburg 1975)Google Scholar.

⁶² Hill, K., ‘Roman domestic garden sculpture’ in Ancient Roman Gardens 83–94Google Scholar for some of the standard pieces of the reproduction trade.

⁶³ Maiuri, A., ‘Gli oeci vitruviani in Palladio e nella casa pompeiana ed ercolanese’, Palladia n.s. 2 (1952) 1–8Google Scholar for the links between Vitruvius and the remains. For Settefinestre, Carandini, A. et al. , Settefinestre (1985) 1**, 20–3Google Scholar.

⁶⁴ Described by Callixeinos in Athenaeus v. 196–209; discussed by Studnicza, F., Das Symposion Ptolemaios II (1914) esp. 32–4Google Scholar for the Pompeian parallels; Rice, E., The grand procession of Ptolemy Philadelphus (1983)Google Scholar discusses Callixeinos, but not this passage. Fittschen, K., ‘Zur Herkunft und Entstehung des 2. Stils’, in Hellenismus in Mittelilalien vol. ii, 544–9Google Scholar brings out the relevance of these descriptions for the decorative versions of such oeci.

⁶⁵ Athenaeus v. 207 d–e on the gymnasium, promenades, temple of Aphrodite and library which formed part of the dining-complex.

⁶⁶ Apsed rooms are catalogued and discussed by Tamm, B., Auditorium und Palatium 147–88Google Scholar, bringing out the ‘sacred’ connotations of the apse. For the Auditorium Maecenatis, currently interpreted as a nymphaeum∣triclinium, see Rizzo, S. and de Vos, M. in L' archeologia in Roma capitale tra sterro e scavo (Roma Capitale 1870–1911 exhibition catalogue, 1983) 225–47Google Scholar.

⁶⁷ See the illuminating discussion of Coarelli (above, n. 46).

⁶⁸ Petr., Sat. 77. 4Google Scholar: ‘aedificavi hanc domum. ut scitis, + cusuc + erat; nunc templum est.’ The sense of the corrupt word, ‘cusuc’ is fairly clear.

⁶⁹ The importance of curtains is nicely captured by Y. Thébert (op. cit. n. 11) 388 f.

⁷⁰ For Eastern examples of fabric style decorations see Alterlümer von Pergamon xiv, 86–92Google Scholar; Rostovtzeff, M., JHS 39 (1919) 151–3Google Scholar. For Italian examples, see Barbet, A., La peinture murale romaine 203Google Scholar classified, in several cases wrongly, as ‘zone supérieure’; Barbet, A. and Allag, C., ‘Techniques de préparation des parois dans la peinture murale romaine’, MEFRA 84 (1972) 992–1006Google Scholar; Carandini, , Settefinestre 1** 231–2Google Scholar. The style is increasingly attested in the western provinces: see Drack, W., Die römische Wandmalerei der Schweiz (Monog. zur Ur- und Frühgesch. der Schwciz 8, 1950) 31–4Google Scholar; Ling, R., Romano-British Wall Painting (1985), 34–6Google Scholar; Ling, R., ‘Two Silchester wall-decorations recovered’, Antiquaries Journal lxiv (1984) 280–97CrossRefGoogle Scholar.

⁷¹ On the Greek background, see Bruno, V.J., ‘Antecedents of the Pompeian first style’, AJA 73 (1969) 305–17CrossRefGoogle Scholar; Barbet, , Peinture murale 12–25Google Scholar; Laidlaw, A., The First Style in Pompeii: Painting and Architecture (1985) 34–7Google Scholar.

⁷² Laidlaw 307 ff. and 330 for the continued use of first style in public buildings and funerary monuments; 42–6 for preservation of old decoration in private houses.

⁷³ The traditional case, propounded esp. by Beyen, H., Die Pompejanische Wanddekoration vol. I (1938), 279 ffCrossRefGoogle Scholar. was attacked by Engemann, J., Architekturdarstellungen des frühen Zweiten Stils (Röm. Mitt. Suppl. xii, 1967)Google Scholar (with further references). Some degree at least of theatrical inspiration must be conceded: see Barbet, A., La peinture murale romaine, 44 fGoogle Scholar.

⁷⁴ Dependence on hellenistic palaces is argued by Fittschen, K., ‘Zur Herkunft und Entstehung des 2. Stils’ in Hellenismus in Mittelitalien (ed. Zanker, P., 1976) vol. ii, 539–63Google Scholar; also Schefold, K., ‘Der zweite Stil als Zeugnis alexandrinische Architektur’, in Neue Forschungen in Pompeji, ed. Andreae, B. and Kyrieleis, H., 53–60Google Scholar. Dependence on Roman villas was proposed by Lehmann, P. W., Roman Wall Paintings from Boscoreale (1953) 82–131Google Scholar. Leach, E. W., ‘Patrons, painters, and patterns’ (op. cit. n. 3) 141–59Google Scholar rightly sees that both villas and stage settings may be evoked.

⁷⁵ Carettoni, G., Boll. d'Arte 46 (1961), 189–99Google Scholar and Das Haus des Augustus 23 f.; Beyen, , BABESCH 39 (1964), 140–3Google Scholar; against Engemann's attack, Allroggen-Bedel, A., Maskendarstellungen in der römischenkampanischen Wandmalerei (1974) 28–33Google Scholar.

⁷⁶ Fittschen (above n. 74) 544 ff.

⁷⁷ Lucan's observation about marble incrustation is almost but not quite correct, if Pliny, NH xxxvi. 48Google Scholar correctly attributes its introduction to Mamurra: cf. Fittschen (n. 74), 555.

⁷⁸ Notably in the fragmentary wall in the C. di Fabio Rufo, e.g. Barbet, , Peinture murale 45Google Scholar with pl. IIa.

⁷⁹ Conventionally classed as ‘megalographiae’, e.g. Barbet 52–6. On the Boscoreale paintings, Fittschen, K., Neue Forschungen in Pompeji 93–100Google Scholar.

⁸⁰ Described by Pausanias, , Attika xvGoogle Scholar, (cf. Pliny., NH xxxv. 59Google Scholar).

⁸¹ The Porticus Pompei and Octaviae were notable repositories of paintings and other works of art, cf. Platner-Ashby, , Topographical Dictionary of Ancient Rome 427–8Google Scholar; Pape, M., Griechische Kunstwerke aus Kriegesbeute 46 ffGoogle Scholar.

⁸² The use of telamones in public architecture is illustrated by Castiglione, L. in Neue Forschungen in Pompeji 211–24Google Scholar.

⁸³ The temporary theatre of Aemilius Scaurus of 58 B.C. was notorious: Pliny, NH xxxiv. 36Google Scholar; xxxvi, 50 and 113–5, though Pliny regards Curio's of 52 B.C. as more extravagant, xxxvi, 116–20. On these, see Gros, P., Architecture et société 20 f.Google Scholar; Bieber, M., A history of the Greek and Roman Theatre (1961) 167 ff.Google Scholar; Rawson, E., PBSR 53 (1985) 100Google Scholar for the continued erection of scaenae frontes at Rome; Paoletti in Carandini, , Settefinestre 1** 227–8Google Scholar has good remarks on the social context of scenographic paintings.

⁸⁴ Fittschen (n. 74) 543.

⁸⁵ This analysis follows Barbet, , Peinture murale 70 with fig. 27Google Scholar.

⁸⁶ Bastet, and de Vos, , Il terzo stile pompeiano 8–16Google Scholar for the evidence. The Casa di Augusto could be of extreme importance for dating the shift, as suggested by Carettoni, G., Das Haus des Augustus 86 ff.Google Scholar; but inadequate evidence about the structures of the house has been published to confirm Carettoni's hypothesis of building in the period 36–28 B.C. Note also room 12 at Settefinestre with scenographic decoration similar to that of the C. di Augusto, dated to the period of Caesar/Octavian: Carandini, , Settefinestre 1** 215–28Google Scholar.

⁸⁷ Leach, E. W., ‘Patrons, painters, and patterns’ (op. cit. n. 3) 166Google Scholar: ‘In the subtle tone of the early third style we may see reflected the changed temper of the Augustan world where the princeps championed the virtues of solid citizens in whose lives quiet prosperity and dutiful service had replaced the republican passions for honour and display.’

⁸⁸ Documented by Eck, W., ‘Senatorial self-representation: developments in the Augustan period’, in Caesar Augustus (ed. Millar, F. and Segal, E., 1984) 129–67Google Scholar; see too JRS 76 (1986) pp. 66–87Google Scholar.

⁸⁹ See Saller, R. P., Personal Patronage in the Early Empire (1982)CrossRefGoogle Scholar, rightly stressing survival of patronage into the empire.

⁹⁰ Pinacothecae: Varro, RR i. 2. 10Google Scholar; 59. 2; Cicero, de leg. iii. 31Google Scholar; Vitruvius vi. 5. 2 on galleries as an import from the public sector. See further van Buren, A. W., RE Supp. viii (1956) 500–2Google Scholar; Schefold, K., La peinture pompéienne (1972) 50 ff.Google Scholar; Leach (op. cit. n. 3) 162.

⁹¹ Pliny, NH xxxv. 118Google Scholar contrasts the public context of Greek art with the private context of Roman.

⁹² Pape, M., Griechische Kunstwerke aus Kriegesbeute 73–80Google Scholar on the protests from the elder Cato onwards against private possession of masterpieces; Pliny, NH xxxv. 26Google Scholar for Agrippa's proposals. Note too the protest against appropriation of Lysippus' Apoxyomenos by Tiberius, : Pliny, NH xxxiv. 62Google Scholar.

⁹³ Pliny, NH xxxv. 24–6Google Scholar dates public possession of paintings back to the triumph of Mummius, but regards Caesar's dictatorship as the turning point. He omits to mention the importance of the Porticus Pompei.

⁹⁴ Carettoni, , Das Haus des Augustus 90–2Google Scholar thinks Augustus brought an Alexandrian craftsman back in 29 B.C; but Egyptianising art was a widespread vogue, see de Vos, M., L'Egittomania in pitture e mosaici Romani-Campani della prima età imperiale (Et. Prelim. Rel. Or. Emp. Rom. vol.84, 1980), esp. 75–95Google Scholar.

⁹⁵ Strabo's description of contemporary high-life at Canopus is tantalising: vii. 1. 17 (p. 801).

⁹⁶ For a useful preliminary attempt at classification see Barbet, , Peinture murale 193–203Google Scholar.

⁹⁷ Cf. Barbet, , Peinture murale 123–6Google Scholar; Vitruvius vii. 7–14 (note the sheer space he devotes to colour): Pliny, NH xxxv. 29–50Google Scholar.

⁹⁸ Pliny, NH xxxv. 30, 44–7, 50Google Scholar.

⁹⁹ On Vestorius, Vitruvius vii. 11. The use of blue is notable at Herculaneum: e.g. C. dell' Atrio a mosaico, room 9; C. dei Cervi, room 16; C. dell' Alcova, room 8; C. del Gran Portale, room 6. Blue monochromes are apparently not used in the third style (Barbet p. 126) but note the atrium of the C. dei Quadretti Teatrali (16. 11) and the tablinum of the C. della Caccia Antica (VII 4. 48).

¹⁰⁰ At Herculaneum, esp. C. del Salone Nero, room G; C. dei Cervi, room 5; at Pompeii, esp. C. di Fabio Rufo, room D (also C); Villa dei Misteri tablinum 2Google Scholar; at Rome, Villa Farnesina, room C. For others, cf. Barbet p. 124.

¹⁰¹ E.g. the much illustrated tablinum of the C. di Lucretius Fronto (V 4a), evidently the climactic point of its surrounding decorative scheme.

¹⁰² The Casa del Gran Portale (V. 34–5) at Herculaneum is an excellent example, where the triclinium lies directly on the axis of the fauces and the ‘aedicle’ of the decoration continues the vista: see Maiuri, , Ercolano vol. i. 379, fig. 309Google Scholar. At Pompeii I have noted similar arrangements at I 7. 18 (a small shop/house, illustrated), and I 11. 17 (unpublished).

¹⁰³ On this pattern, see Barbet, , La peinture murale 130–5Google Scholar. The black triclinium of the Casa del Frutteto (I 9. 5, room 11) is a good example.

¹⁰⁴ Strocka, V. M., ‘Pompejanische Nebenzimmer’ in Neue Forschungen in Pompeji 101–14Google Scholar; also see his analysis of the C. del Principe di Napoli (below).

¹⁰⁵ E.g. C. del Principe di Napoli, triclinium k, or earlier the white rooms from the Villa Farnesina at Rome and the Villa Imperiale at Pompeii.

¹⁰⁶ Of the black rooms cited in n. 100, those in the C. del Salone Nero, and the C. dei Cervi lack panels and motifs (a style favoured in general at Herculaneum) and that in the Villa dei Misteri, of outstanding polish and elegance, has only subordinate motifs.

¹⁰⁷ E.g. the elegance of the black tablinum at the Villa dei Misteri gives it suitable grandeur, while its restraint in decorative elaboration allows the more private Sala dei Misteri to come as a climax.

¹⁰⁸ Cf. Coarelli, , Dialoghi di Archeologia n.s. 2 (1984), 152–4Google Scholar.

¹⁰⁹ Cf. the explosion of cursus honorum inscriptions for equites and freedmen in the early empire, originating in a shift in senatorial practice under Augustus: Eck, W., ‘Senatorial self-representation’ (op. cit. n.88) 149–52Google Scholar.

¹¹⁰ Cf. Carandini, , Settefinestre 1** 111–13Google Scholar for the seigniorial and slave quarters of a villa; 187–206 for valuable comparative material on American slave plantations.

¹¹¹ Petronius, Sat. 30Google Scholar. On the various types of household slave, Marquardt, J., Das Privatleben der Römer (1886) 142–7Google Scholar is helpful.

¹¹² Dig. 44. 15. 10.44 ‘multum interest qualis servus sit, bonae frugi, ordinarius, dispensator, an vero vulgaris vel mediastinus vel qualisqualis’.

¹¹³ For illustrations, Franciscis, A. de in Neue Forschungen in Pompeji pl. 38–9Google Scholar; Strocka ibid. pl. 75 illustrates an example in a passageway in the Praedia Iuliae Felicis (II 4. 10); Maiuri, , Ercolano 420Google Scholar for an example on the exterior of the modest C. dell' Ara Laterizia (III. 17) at Herculaneum, comparing (n. 216) the stairway at the Porta Marina at Pompeii. Eschebach, H., Die Stabianer Thermen in Pompeji (1979) pl. 67Google Scholar for the latrine (0) at a public baths. There are traces of similar decoration in the passages of the amphitheatre at Pompeii. It is also to be found in the entrance of the C. di Iulius Polybius (unpublished).

¹¹⁴ Elia, O., NSc 1934, 321–39Google Scholar.

¹¹⁵ Sherwin-White, A. N., The Letters of Pliny (1966) 188Google Scholar well observes the silence.

¹¹⁶ Cf. my comments in Antiquaries Journal 66. 2 (1986) 433 fCrossRefGoogle Scholar. on the C. del Principe di Napoli.

¹¹⁷ The Roman lack of interest in private as opposed to public bathing is visible in the absence of evidence for private bath tubs (as opposed to bath suites) in Vesuvian houses; contrast the frequency of bath tubs at Olynthos, , Robinson, and Graham, , Excavations at Olynthos VIII, 198 ffGoogle Scholar. On communal latrines as a standard feature of Roman life, see Drexel, F., ‘Das Latrinenwesen’, excursus in Friedländer, , Sittengeschichte¹⁰ iv, 310–11Google Scholar; and recently Scobie, A., ‘Slums, sanitation and mortality’, Klio 68 (1986) 429CrossRefGoogle Scholar.

¹¹⁸ Carettoni, , Das Haus des Augustus 92Google Scholar identifies this ‘Syracuse’ with one upper-floor room discovered in the C. di Augusto. But the language of Suet., Aug. 72Google Scholar, ‘locus in edito singularis … huc transibat’ suggests something far more remote than the ‘retired place at the top of the house’ of Rolfe's Loeb translation. It was surely a separate building.

¹¹⁹ Cf. Drerup, H., ‘Bildraum u. Realraum in der römischen Architektur’, Röm. Mitt. 66 (1959) 147–74, esp. 158–9Google Scholar for the social basis of the phenomenon.

¹²⁰ This important observation of ‘optical symmetry’ was made by Bek, Lise, Towards Paradise on Earth (Analecta Romana IX, 1980) esp. 17 fGoogle Scholar. and 181–9, and documented in more detail by Jung, F., ‘Gebaute Bilder’, Antike Kunst 27 (1984) 71–122Google Scholar. Symmetrical vistas could also be appreciated from the interior looking outwards: cf. Pliny, ep. ii. 17. 5Google Scholar.

¹²¹ Examples of painted nature: Pompeii, C. di Sallustio I 7. 19 (at the rear of the C. del Ephebo); VI 8. 22 and 23, C. della Fontana Grande and Piccola; also VI 14. 20 C. di Orfeo, where the figure of Orpheus dominates the scene of nature.

¹²² Cf. Bek, op. cit. 186, well citing Horace, Epist. 1. 10. 23Google Scholar. ‘laudatur domus longos quae prospicit agros’. The Case del Fauno, di Pansa, del Labirinto, etc. at Pompeii give views centred on Vesuvius. The Casa del Menandro combines all three types, since the vista passes through the peristyle garden to apses painted with scenes of wild nature at the end of the peristyle; while above them from the front doorstep of the entrance is visible, neatly framed, a peak of the Monti Lattari.

¹²³ Cf. Jung, op. cit. 77. For houses at Delos with some optical symmetry, sec Explorations à Délos VIII b pl. XIII (Maison du Trident) and XXVII pl. A (Maison des Tritons, dated to the late second century B.C., ibid. p. 424).

¹²⁴ E.g. Pliny, ep. iii. 5. 8 ffGoogle Scholar. on the elder Pliny, iii. 1. 4–9 on Spurinna; ix. 36 on himself; Suet., Aug. 78Google Scholar; Vesp. 21; Martial iv. 8.

¹²⁵ E.g. the triclinium (8) of the C. degli Amanti (I 10. 11) is reached directly from the atrium, but is oriented towards the peristyle; the C. di Lucretius Fronto (V 4, a) has a finely decorated triclinium in the atrium, but this supplements one in the garden. The smaller the house, the less likely a clear atrium/peristyle distinction.

¹²⁶ See above n. 55. Cf. Y. Thébert (op. cit. n. 11) 357 f. on the social significance of the atrium house type.

¹²⁷ E.g. on a very small scale the C. di Fabius Amandio (I 7. 23), or in what is surely a craftsman's house, the C. del Fabbro (I 10. 7).

¹²⁸ Strocka (op. cit. n. 24).

¹²⁹ Strocka 33 and 49 f.

¹³⁰ The question of upper bedrooms and who populated them needs further investigation. Note that the interpretation of Petronius, Sat. 77Google Scholar which locates the bedrooms of Trimalchio and his wife upstairs (McKay, , Houses, Villas and Palaces 113Google Scholar) depends on a questionable reading of the text.

¹³¹ Thus in the C. del Menandro, the recess in the peristyle containing apparently a shrine of the ancestors has its late republican decoration preserved though the rest of the peristyle has been redecorated: Maiuri, , La Casa del Menandro (1933) 98–106Google Scholar; Ling, R., Antiquaries Journal 63 (1983) 45CrossRefGoogle Scholar.

¹³² Laidlaw, 42–6 for the deliberate preservation of first style decoration. Maiuri (op. cit. n. 32) needlessly identifies the peristyle area as a ‘gynaeceum’. Jashemski, W., The Gardens of Pompeii (1979) 168–70Google Scholar sees the house as a hotel; but it shows no affinities with other more plausible hotels and inns, cf. J. Packer, ‘Inns at Pompeii’ (op. cit. n. 40).

¹³³ E.g. at Pompeii the C. dei Quadretti teatrali (I 6.11), or at Herculaneum the C. del Bicentenario (V. 15). These might have provoked the barbs of Martial (12. 50. 7 f.): … atria longa patent, sed nee cenantibus usquam∣nec somno locus est. quam bene non habitas!

¹³⁴ Meiggs, R., Roman Ostia 2 (1960) 253 fGoogle Scholar. for a clear account of the change; also Tamm, , Auditorium und Palatium 145–6Google Scholar. On the Casa di Fortuna Annonaria at Ostia see Becatti, G., Case Ostiensi del Tardo Impero (1923) 23–5Google Scholar; Boersma, J. S., Amoenissima Civitas: Block V. ii at Ostia: description and analysis of its visible remains (1985) 47 ff., 138 ffGoogle Scholar.

¹³⁵ McKay, , Houses, Villas and Palaces 66Google Scholar fails to pursue the connection.

¹³⁶ See Saller, R. P., Personal Patronage under the Early Empire (1982) 7 ffCrossRefGoogle Scholar.

¹³⁷ Cf. Jung (op. cit. n. 120) 116 f.

¹³⁸ Cf. Y. Thébert (op. cit. n. 11) 353–87 for a fine discussion of the public face of the grand houses of Roman Africa in the post-Pompeian period.

¹³⁹ Tamm, , Auditorium und Palatium 189–205Google Scholar on the varied terminology for dining rooms, rightly suggesting that many such rooms may have served ‘audience’ functions.

¹⁴⁰ See Girouard, M., Life in the English Country House 194 fGoogle Scholar. discusses the use of strings of reception rooms in the eighteenth century; on growing differentiation of function in the nineteenth century, 239, 300 ff. Ariès, P., Centuries of Childhood 378–81Google Scholar is excellent on the contrast between the seventeenth century ‘big house’, crowded with friends, clients, relatives and protégés, lacking in either privacy or functional differentiation of rooms, and the growing privacy and differentiation of the eighteenth and nineteenth century house.

¹⁴¹ Although unpublished, there is a plan and illustrations of this suite in Barbet, , Peinture murale 241–3Google Scholar.

¹⁴² Maiuri, , Ercolano 306–10Google Scholar for the suite and its adaptations.

¹⁴³ Elia, O., NSc 1934, 282 ffGoogle Scholar.

¹⁴⁴ Petronius, Sat. 77Google Scholar; but the account of his slaves (47 and 53) implies a servile household larger than 20 bedrooms would accommodate.

¹⁴⁵ Suet., Aug. 72Google Scholar.

¹⁴⁶ Pliny, ep. ii. 17. 6–13Google Scholar is a giddy succession of cubicula; 20–24, more, including a favourite for the Saturnalia; v. 6. 21, 24, 28, 31 (two suites of four and three cubicula), 37.

¹⁴⁷ Richardson, L. Jr., ‘A contribution to the study of Pompeian dining-rooms’, Pompeii Herculaneum Stabiae. Bollettino dell' associazone internazionale Amici di Pompei i (1983) 61–71Google Scholar identifies the pattern of linked suites; but his assertion (p. 69) that women ate apart from the men and seated is based on a misreading of passages (Isidorus xx. 11.9, Valerius Maximus ii. 1. 2) recalling ‘ancestral’ Roman habits in contrast to those of the historical period—they used to sit, but now they lie to eat with the men.

¹⁴⁸ Cf. Carandini, , Settefinestre 1* 120Google Scholar; also Maiuri, , Ercolano 325Google Scholar (Casa dell'Albergo). Also impressive is the cluster of cubicula around the oecus Corinthius of the C. del Labirinto.

¹⁴⁹ So too the excavators identify the bedroom (28) of the ‘oecus Corinthius’ at Settefinestre as that of the dominus (Settefinestre 1** p. 41); doubtless he used it, but to share with his wife he might also have used rooms 3 and 25.

¹⁵⁰ Maiuri (op. cit. n. 32), 456–7; cf. de Vos, A. and , M., Pompei, Ercolano, Stabia (Guida Laterza 1982)Google Scholar.

¹⁵¹ Girouard, , English Country House, 144 ffGoogle Scholar. for an illuminating discussion of etiquette and status as the organising principles of the eighteenth century ‘formal’ house.

¹⁵² This point emerges forcibly from the examination of the historical development of the Menander block (I 10) at Pompeii: Ling, R., ‘The insula of the Menander at Pompeii: interim report’, Antiquaries Journal 63 (1983) 34–57CrossRefGoogle Scholar.

¹⁵³ Evidence of this is abundant: note particularly the observations of A. Maiuri, Ercolano passim.