¹ On Latin inscriptions, Mrozek, S., ‘À propos de la répartition chronologique des inscriptions latines dans le Haut-Empire’, Epigraphica 35 (1973), 113–18Google Scholar, cf. idem, ‘A propos de la répartition chronologique des inscriptions latines dans le Haut-Empire’, Epigraphica 50 (1988), 61–4, supported by numerous local studies, e.g. Wightman, E. M., Gallia Belgica (1985), 163Google Scholar and Duncan-Jones, R. P., The Economy of the Roman Empire: Quantitative Studies (2nd edn, 1982), 350–7Google Scholar, and discussed in an important paper by MacMullen, R., ‘The epigraphic habit in the Roman Empire’, AJP 103 (1982), 233–46Google Scholar. The evidence from the East is less well tabulated but cf. Alföldy, G., ‘Augustus und die Inschriften: Tradition und Innovation’, Gymnasium 98 (1991), 289–324Google Scholar, for the initial boom; also MacMullen, R., ‘The frequency of inscriptions in Roman Lydia’, ZPE 65 (1986), 237–8Google Scholar, and Roueché, C., Aphrodisias in Late Antiquity (1989), xix–xxiGoogle Scholar, on Asian cities. Meyer, E. A., ‘Explaining the epigraphic habit in the Roman Empire: the evidence of epitaphs’, JRS 80 (1990), 74–96Google Scholar, in the most imaginative and rigorous study of the phenomenon to date, has produced sample curves for a series of western and eastern cities. I am greatly indebted to her work in all that follows.

² On early Roman literacy, cf. most recently Stoddart, S. and Whitley, J., ‘The social context of literacy in Archaic Greece and Etruria’, Antiquity 62 (1988), 761–72CrossRefGoogle Scholar; Cornell, T., ‘The tyranny of the evidence: a discussion of the possible uses of literacy in Etruria and Latium in the archaic age’, in Humphrey, J. H. (ed.), Literacy in the Roman World (1991), 7–33Google Scholar.

³ Sailer, R. P. and Shaw, B. D., ‘Tombstones and Roman family relations in the Principate: civilians, soldiers, and slaves’, JRS 74 (1984), 124–56Google Scholar, at 124, estimate that between 170,000 and 190,000 of around 250,000 extant inscriptions are epitaphs. Such global figures conceal considerable regional contrasts. Eck, W., ‘Römische Grabinschriften. Aussagabsicht und Aussagefähigkeit im funerären Kontext’, in von Hesberg, H. and Zanker, P. (eds), Römische Gräberstraben. Selbstdarstellung — Status — Standard (1987), 25–41Google Scholar, at 25, estimates more than 35,000 of the 39,000 inscriptions in CIL VI (from the city of Rome) were funerary, while Biro, M., ‘The inscriptions of Roman Britain’, Acta Archaeologica Academiae Scientiarum Hungaricae 27 (1975), 13–58Google Scholar, at 42, estimates that votive altars outnumber tombstones in RIB (from Roman Britain) by around 3 to 1. The contrast probably reflects differences between military and civilian rather than between peripheral and central epigraphies.

⁴ cf. Corbier, M., ‘L'écriture dans l'éspace public romain’, in L'Urbs. Espace urbain et histoire I^er siècle av. J. C. — III siècle ap. J. C. Actes du collogue international organisé par le CNRS et l'Ecole française à Rome 8–12 mai 1985, Collection de l'École française à Rome 98 (1987), 27–60.Google Scholar

⁵ MacMullen, op. cit. (n. 1, 1982).

⁶ For posited connections with urbanism and demography, Jongman, W., The Economy and Society of Pompeii (1988), at 68–9Google Scholar; with the military, Biró, op. cit. (n. 3); with Romanization, Mócsy, A., Gesellschaft und Romanisation in der römischen Provinz Moesia Superior (1970), 199–212Google Scholar, Nicols, J., ‘Indigenous culture and the process of Romanisation in Iberian Galicia’, AJP 108 (1987), 129–51Google Scholar, Wightman, op. cit. (n. 1), 168–90, although for some doubts about this approach see Mann, J. C., ‘Epigraphic consciousness’, JRS 75 (1985), 204–6Google Scholar, and Cepas, A., The North of Britannia and the North-West of Hispania. An Epigraphic Comparison, BAR Int. Ser. 470 (1989), 54–8Google Scholar. For cautious links between epigraphy and literacy cf. Harris, W. V., Ancient Literacy (1989), 265–9Google Scholar, criticized by Thomas, R., Literacy and Orality in Ancient Greece (1992), 162–5CrossRefGoogle Scholar.

⁷ Meyer, op. cit. (n. 1). For a related argument about Attic epigraphy, cf. Meyer, E., ‘Epitaphs and citizenship in classical Athens’, JHS 113 (1993), 99–121CrossRefGoogle Scholar.

⁸ I am very grateful to Richard Sailer for these points.

⁹ e.g. Morris, I., Death-Ritual and Social Structure in Classical Antiquity (1992), 167–73CrossRefGoogle Scholar, drawing on Cannon, A., ‘The historical dimension in mortuary expressions of status and sentiment’, Current Anthropology 30.4 (1989), 437–58CrossRefGoogle Scholar.

¹⁰ Recent treatments showing the potential of all these media include J. L. Franklin, ‘Literacy and the parietal inscriptions of Pompeii’, in Humphrey, op. cit. (n. 1), 77–98; Veyne, P., ‘“Titulus Praelatus”: offrande, solemnisation et publicité dans les ex-voto greco-romains’, RA (1983), 281–300Google Scholar; Corbier, op. cit. (n. 4); Harris, W. V. (ed.), The Inscribed Economy: Production and Distribution in the Roman Empire in the Light of instrumentum domesticum, JRA supp. ser. 6 (1993)Google Scholar.

¹¹ Desbordes, F., Idées romaines sur l'écriture (1990)Google Scholar. For one aspect of this larger question Bowman, A. K. and Woolf, G. D. (eds), Literacy and Power in the Ancient World (1994)Google Scholar.

¹² Eck, op. cit. (n. 3), Morris, op. cit. (n. 9), 164–8, for this point in relation to tombstones.

¹³ Petronius, Satyricon 58. On writing in Petronius, cf. Horsfall, N., ‘“The uses of literacy” and the Cena Trimalchionis’, Greece and Rome 36 (1989), 74–89, 194–209CrossRefGoogle Scholar. Cf. also Pliny, Epistles VIII.6.14 discussed below.

¹⁴ Susini, G. C., The Roman Stonecutter (1973), 16–20.Google Scholar

¹⁵ ‘Exegi monumentum aere perennius ∣ regalique situ pyramidum altius, ∣ quod non imber edax, non Aquilo impotens ∣ possit diruere aut innumerabilis ∣ annorum series et fuga temporum. ∣ non omnis moriar, multaque pars mei ∣ vitabit Libitinam: usque ego postera ∣ crescam laude recens, dum Capitolium ∣ scandet cum tacita virgine pontifex’. Odes III. 30.1–9 adapting Pindar, Pythian VI.6ff. to a Roman context.

¹⁶ cf. Foxhall, L., ‘Monumental ambitions: the significance of posterity in Greece’, in Spencer, N. (ed.), Time, Tradition and Society in Greek Archaeology. Bridging the ‘Great Divide’ (1995), 132–49Google Scholar. I am very grateful to Dr Foxhall for showing me this paper in advance of publication.

¹⁷ On these letters cf. also Eck, op. cit. (n. 3), at 76–7.

¹⁸ For this sort of reading of Pliny's letters cf. W. M. Beard, ‘Ancient Literacy and the function of the written word in Roman religion’, in Humphrey, op. cit. (n. 2), 35–58, at 39–42.

¹⁹ ‘Subit indignatio cum miseratione, post decimum mortis annum reliquias neglectumque cinerem sine titulo sine nomine iacere, cuius memoria orbem terrarum gloria pervagetur. At ille mandaverat caveratque, ut divinum et immortale factum versibus inscriberetur: Hie situs est Rufus, pulso qui Vindice quondam imperium adseruit non sibi sed patriae. Tam rara in amicitiis fides, tarn parata oblivio mortuorum, ut ipsi nobis debeamus etiam conditoria exstruere omniaque heredum officia praesumere’. Pliny, Epistles VI. 10.3–5.

²⁰ ‘Huic senatus ob fidem pietatemque erga patronos ornamenta praetoria decrevit et sestertium centies quinquagies, cuius honore contentus fuit’. Pliny, Epistles VII.29.2.

²¹ ‘Delectus est celeberrimus locus, in quo legenda praesentibus, legenda futuris proderentur. Placuit aere signari omnes honores fastidiosissimi mancipi, quosque repudiasset quosque quantum ad decernentes pertinet gessit. Incisa et insculpta sunt publicis aeternisque monumentis praetoria ornamenta Pallantis, sic quasi foedera antiqua, sic quasi sacrae leges’. Pliny, Epistles VIII.6.14.

²² On aspects of the origin of this system, cf. Eck, W., ‘Senatorial self-representation: developments in the Augustan period’, in Millar, F. G. B. and Segal, E. (eds), Caesar Augustus: Seven Aspects (1984), 129–67Google Scholar; Wallace-Hadrill, A., ‘Roman arches and Greek honours: the language of power at Rome’, PCPS 216 (n.s. 36) (1990), 143–81Google Scholar.

²³ For a similar aristocratic assertion that some lives could not be monumentalized cf. Petronius, Satyricon 71 on Trimalchio's aspirations to preserve his name.

²⁴ Digest XI.7.2.6 (Ulpian), ‘Monumentum est quod memoriae servandae gratia existat’. For an example from the Greek world, cf. Rogers, G. M., The Sacred Identity of Ephesos. Foundation Myths of a Roman City (1991), 19–22.Google Scholar

²⁵ Eisner, J., ‘From the pyramids to Pausanias and Piglet: monuments, travel and writing’, in Goldhill, S. and Osborne, R. (eds), Art and Text in Ancient Greek Culture (1994), 224–54Google Scholar, at 225 and 244–52.

²⁶ Eck, op. cit. (n. 22), at 132–3, for an insistence on this point. Cf. Hölscher, T., ‘Die Geschichtsauffassung der römischer Repräsentationskunst’, JDAI 95 (1980), 265–321Google Scholar, at 279–81.

²⁷ e.g. Derks, T., ‘The perception of the Roman pantheon by a native élite: the example of votive inscriptions from Lower Germany’, in Roymans, N. and Theuws, F. (eds), Images of the Past. Studies on Ancient Societies in North-western Europe (1991), 235–65Google Scholar, at 240–1, discussing a sanctuary at Osterburken (Baden-Württemburg).

²⁸ cf. on the significance of the choice and ornamentation of scripts, McKitterick, R., ‘Text and image in the Carolingian world’, McKitterick, R. (ed.), The Uses of Literacy in Early Mediaeval Europe (1990), 297–318CrossRefGoogle Scholar, especially at 310–14.

²⁹ Williamson, C., ‘Monuments of bronze; Roman legal documents on bronze tablets’, Classical Antiquity 6 (1987), 160–83CrossRefGoogle Scholar, cf. Crawford, M. H., ‘The laws of the Romans: knowledge and diffusion’, in Estudios sobre la Tabula Siarensis, Anejos del Archivo Español de Arqueologia 9 (1988), 127–40Google Scholar.

³⁰ For investigations into the structure of these environments, cf. G. Zimmer, Locus datus decreto decurionum. Zur Statuenaufstellung zweier Forumslagen im römischen Africa (1989) and Derks, op. cit. (n. 27).

³¹ On the symbolic aspects of lists of names, Nicolet, C., L'inventaire du monde: géographie et politique aux origines de l'Empire romain9 (1988)Google Scholar.

³² Sailer and Shaw, op. cit. (n. 3).

³³ Beard, op. cit. (n. 18), at 47 and cf. in general 44–8.

³⁴ Another way of viewing these inscriptions is as gifts designed to create relationships; cf. Price, S. R. F., Rituals and Power. The Roman Imperial Cult in Asia Minor (1984), 65–77Google Scholar, on worship as a component of a system of gift exchange that constituted the symbolic unity of the Empire. Equally, funerary inscriptions set up by heirs might be seen as reciprocating testamentary bequests; cf. Meyer, op. cit. (n. 1) and Purcell, N., ‘Tomb and suburb’, in von Hesberg, H. and Zanker, P. (eds), Römische Gräberstraßen. Selbstdarstellung — Status — Standard (1987), 25–41Google Scholar. Some inscriptions were formally presented to those named in them, for example the certificates (diplomata) that marked the grant of citizenship to auxiliary veterans or tabulae patronatus like CIL VI.1492, a bronze copy of the municipal decree whereby Ferentinum co-opted Titus Pomponius Bassus as patron, which was intended for display in his house. On this genre of inscription, cf. Nicols, J., ‘Tabulae Patronatus: a study of the agreement between patron and client community’, ANRW II.13 (1980), 535–61Google Scholar.

³⁵ Gordon, R., Beard, M., Reynolds, J., and Roueché, C., ‘Roman Inscriptions 1986–90’, JRS 83 (1993), 131–58Google Scholar, at 154–5, for this formulation.

³⁶ Beard, W. M., ‘Writing and ritual. A study of diversity and expansion in the Arval Acta’, PBSR 40 (1985), 114–62Google Scholar, at 146, states clearly an analytical difficulty in accounting for long-term trends in the Arval Acta, which may be applied to the explanation of epigraphic trends in general. ‘Unless we suppose that the writing of our texts took place in complete cultural isolation, we cannot deny the influence on their character of external factors. We cannot expect that an explanation could be generated from the texts alone. Our difficulty lies in deciding which external factors should be seen as influential on the precise character of the texts’.

³⁷ Bradley, R. J., Altering the Earth, Society of Antiquaries of Scotland Monograph 8 (1993)Google Scholar, is the most readable and learned introduction to this literature, building on Bradley, R. J., ‘Studying Monuments’, in Bradley, R. J. and Gardiner, J. (eds), Neolithic Studies, BAR Int. ser. 133 (1984), 61–6Google Scholar, and Bradley, R. J., ‘The Archaeology of Monuments’, in Bradley, R. J., Consumption, Change and the Archaeological Record, University of Edinburgh Department of Archaeology Occasional Paper 13 (1985), 1–20Google Scholar. For a recent selection of approaches cf. World Archaeology 22.2 (1990) on the theme ‘Monuments and the Monumental’. For another attempt to apply some of these ideas to Roman epigraphy see Barrett, J. C., ‘Chronologies of remembrance: the interpretation of Roman inscriptions’, World Archaeology 25.2 (1993), 236–47CrossRefGoogle Scholar.

³⁸ Several late prehistoric societies did construct personal funerary monuments of comparable scale to ancient ones, but perhaps only for the burial of chiefs, which might be considered a transitional category between collective and personal monumentalization. The prominence of sumptuary legislation concerned with funerals in Archaic Greece and Rome suggests that this ‘privatization’ of monumentality may have been among the more important ways through which aristocracies established themselves in the earliest ancient states, cf. Morris, I., Burial and Society: The Rise of the Greek City-State (1987) and op. cit. (n. 9), 149–55.Google Scholar

³⁹ Bradley, op. cit. (n. 37, 1993) on these processes.

⁴⁰ Cherry, J. F., ‘Generalisation and the archaeology of the state’, in Green, D., Haselgrove, C. C. and Spriggs, M. (eds), Social Organisation and Settlement, BAR Int. ser. 47 (1978), 411–37Google Scholar; Trigger, B. G., ‘Monumental architecture: a thermodynamic explanation of symbolic behaviour’, World Archaeology 22.2 (1990), 119–32CrossRefGoogle Scholar. By ‘formative’ I mean those periods in which societies and cultures were malleable enough to be reconstructed and redirected in fundamental ways, which, once institutionalized and routinized, acted as durable and enduring social structures and practices, the formation, as it were, of habitus from practice. For example, the formative period of the Archaic Greek polis was characterized by various cultic activity around Bronze Age tombs and by the construction of monumental sanctuaries. On tomb cult, cf. Snodgrass, A. M., Archaic Greece, the Age of Experiment (1980), 38–40Google Scholar; Morris, I., ‘Tomb cult and the “Greek Renaissance”: the past in the present in the eighth century BC’, Antiquity 62 (1988), 750–61CrossRefGoogle Scholar; Whitley, J., ‘Early states and hero cults: a reappraisal’, JHS 108 (1988), 173–82CrossRefGoogle Scholar; idem, ‘The monuments that stood before Marathon: tomb cult and hero cult in archaic Attica’, AJA 98 (1994), 213–30; Antonaccio, C., ‘Contesting the past: hero cult, tomb cult and epic in early Greece’, AJA 98 (1994), 389–410CrossRefGoogle Scholar. For sanctuaries, de Polignac, F., La naissance de la cité grecque (1984)Google Scholar; Morgan, C. A., Athletes and Oracles: the Transformation of Olympia and Delphi in the Eighth Century B.C. (1990)Google Scholar; Alcock, S. E. and Osborne, R. G. (eds), Placing the Gods (1994)Google Scholar.

⁴¹ Bradley, op. cit. (n. 37), with Hodder, I. R., ‘Social and economic stress and material culture patterning’, American Anthropology 44 (1979), 446–54Google Scholar; Thomas, J., Rethinking the Neolithic (1991), 29–55.Google Scholar

⁴² Hobsbawm, E. and Ranger, T. (eds), The Invention of Tradition (1983)Google Scholar, Fentress, J. and Wickham, C., Social Memory (1992)Google Scholar. Snodgrass, op. cit. (n. 40), stresses the importance of traditions of an heroic age in connection with tom b cult in Archaic Greece.

⁴³ The rather different anxieties of the third century, for example, on which cf. Alföldy, G., ‘The crisis of the third century as seen by contemporaries’, GRBS 15 (1974), 89–111Google Scholar, found expression in very different ways to the personal anxieties that prompted most early imperial epigraphy.

⁴⁴ Recently, Price, S. R. F., ‘The future of dreams: from Freud to Artemidorus’, Past and Present 113 (1986), 3–37CrossRefGoogle Scholar; T. S. Barton, Power and Knowledge: Astrology, Physiognomies and Medicine in the Roman Empire (1994); Purcell, N., ‘Literate games: Roman urban society and the game of Alea’, Past and Present 147 (1995), 3–37CrossRefGoogle Scholar.

⁴⁵ For philosophical concerns of this type, cf. the passages of Epictetus discussed in Millar, F. G. B., ‘Epictetus and the imperial court’, JRS 55 (1965), 141–8Google Scholar. For a different philosophical response to these same anxieties, cf. Habinek, T. N., ‘An aristocracy of virtue: Seneca on the beginnings of wisdom’, YCS 19 (1992), 187–203Google Scholar. On law, Champlin, E., Final Judgements: Duty and Emotion in Roman Wills 200 B.C.–A.D. 250 (1991)Google Scholar, on the development of testamentary law, Johnston, D., Roman Law of Trusts (1988)CrossRefGoogle Scholar, on fideicommissa.

⁴⁶ Barton, op. cit. (n. 44), for these developments. The number of treatises produced on all these subjects in the early Empire is very striking.

⁴⁷ Hopkins, K., Death and Renewal. Sociological Studies in Roman History 2 (1983), at 247–55CrossRefGoogle Scholar; Johnston, D., ‘Munificence and municipia: bequests to towns in classical Roman law’, JRS 75 (1985), 105–25Google Scholar; Champlin, op. cit. (n. 45), at 25–8.

⁴⁸ cf. Barton, C. A., The Sorrows of the Ancient Romans. The Gladiator and the Monster (1993)Google Scholar, for an impressionistic but insightful exploration of these themes.

⁴⁹ cf. Edwards, C., The Politics of Immorality in Ancient Rome (1993), at 150–60CrossRefGoogle Scholar and 200–4, for parallel examples of dilemmas generated by aemulatio.

⁵⁰ cf. for example, the material gathered in Krier, J., Die Treverer ausserhalb ihrer Civitas (1981)Google Scholar and in Wierschowski, L., Die regionale Mobilität in Gallien nach den Inschriften des I. bis 3Google Scholar. Jahrhunderts n.Chr. Quantitative Studien zur Social- und Wirtschaftsgeschichte der westlichen Provinzen des römischen Reiches,Historia Einzelschriften 91 (1995).

⁵¹ For social mobility as an index of the likelihood that children will not succeed their parents in equivalent social roles, cf. Runciman, W. G., ‘Accelerating social mobility: the case of Anglo-Saxon England’, Past and Present 104 (1984), 3–30CrossRefGoogle Scholar, reprinted in idem, Confessions of a Reluctant Theorist. Selected Essays (1089), 121–47.

⁵² Although conversely the fact that the strength of familial and civic ties could no longer be taken for granted may have increased the valency of expressions of sentiment or loyalty in these directions.

⁵³ On the weakening of the civic model for religion, cf. Gordon, R., ‘Religion in the Roman Empire: the civic compromise and its limits’, in Beard, M. and North, J. (eds), Pagan Priests (1990), 235–55Google Scholar, further developed by Rives, J. B., Religion and Authority in Roman Carthage from Augustus to Constantine (1995).Google Scholar

⁵⁴ For increasing individualism in one sector of imperial Roman society, cf. Hopkins, op. cit. (n. 47), 79–81. Full discussion of what ‘individualism’ might have comprised in this period or of ancient ideas of ‘the self’ cannot be attempted here. While there are evident connections with other moral developments of the early Empire, the stress envisaged here is on the collapse, multiplication, and complexification of norms rather than on the emergence of new coercive structures, and the field in which this process operated is conceived of as encompassing both ‘public’ and ‘private’ spheres, difficult in any case to distinguish or delineate in Roman society.

⁵⁵ Social mobility may, in fact, be thought of as one form of social risk: the higher the rates of social mobility, the greater the uncertainty about any individual's future identity, prosperity, and status. For social caging, cf. Mann, M., The Sources of Social Power I (1986), 38–40Google Scholar.

⁵⁶ For discussion of how routinized behaviour generates and sustains social structure, cf. Giddens, A., The Constitution of Society. Outline of the Theory of Structuration (1984).Google Scholar

⁵⁷ e.g. by Syme, R., The Roman Revolution (1939)Google Scholar; Wiseman, T. P., New Men in the Roman Senate 139 B.C.-A.D. 14 (1971)Google Scholar; Hammond, M., ‘Composition of the Roman senate A.D. 68–235’, JRS 47 (1957), 73–81Google Scholar; Hopkins, K., ‘Elite mobility in the Roman Empire’, Past and Present 32 (1965), 12–36CrossRefGoogle Scholar, reprinted in Finley, M. I. (ed.), Studies in Ancient Society (1974), 103–20Google Scholar; idem, op. cit. (n. 47), 31–200, but cf. now Hahn, J. and Leunissen, P. M. M., ‘Statistical method and inheritance of the consulate under the early Roman Empire’, Phoenix 44 (1990), 60–81CrossRefGoogle Scholar.

⁵⁸ Purcell, N., ‘The Apparitores: a study of social mobility’, PBSR 51 (1983), 125–73Google Scholar; Duthoy, R., ‘La fonction sociale de l'augustalité’, Epigraphica 36 (1974), 134–54Google Scholar; Abramenko, A., Die municipale Mittelschicht im kaiserzeitlichen It alien. Zu einem neuen Verständnis von Sevirat und Augustalität (1993)Google Scholar; Garnsey, P. D. A., ‘Descendants of freedmen in local politics: some criteria’, in Levick, B. (ed.), The Ancient Historian and his Materials. Essays in Honour of C. E. Stevens on his Seventieth Birthday (1975), 167–80.Google Scholar

⁵⁹ ‘Most recently cf. Jongman, op. cit. (n. 6), 207–329; Mouritsen, H., ‘A note on Pompeian epigraphy and social structure’, Classica et Mediaevalia 41 (1990), 131–49Google Scholar.

⁶⁰ On luxury and status in the Vesuvian cities cf. Wallace-Hadrill, A. F., Houses and Society in Pompeii and Herculaneum (1994)Google Scholar. For the Satyricon as a source for social history cf. Veyne, P., ‘La vie de Trimalchion’, Annales ESC 16.1 (1971), 213–47Google Scholar, with Garnsey, , op. cit. (n. 58), taken into account in the revised edition published in Veyne, P., La société romaine (1990), 13–57.Google Scholar

⁶¹ The process might be compared to a common method by which belief systems have expanded, by disturbing prior systems and then suggesting a more complete vision cf. Gellner, E., The Psychoanalytic Movement. The Cunning of Unreason (2nd edn, 1993), at 40–73.Google Scholar

⁶² Taylor, L. R., ‘Freedmen and freeborn in the epitaphs of imperial Rome’, AJPh 82 (1962), 113–32Google Scholar; Eck, , op. cit. (n. 3) and especially in ‘Aussagefähigkeit epigraphischer Statistik und die Bestattung von Sklaven im kaiserzeitlichen Rom’, in Kneißl, P. und Losemann, V. (eds), Alte Geschichte und Wissenschaftsgeschichte, Festschrift für K. Christ zum 65. Geburtstag (1988), 130–9Google Scholar. The issue is discussed at length in Hope, V., Reflections of Status: A Contextual Study of the Roman Tombstones of Aquileia, Mainz and Nimes, PhD thesis, University of Reading, 1994Google Scholar. I am grateful to Dr Hope for permission to cite this work.

⁶³ Mann, op. cit. (n. 6), 204–6, on the impact of these factors on the surviving epigraphy of Roman Britain. Duncan-Jones, op. cit. (n. 1), appendix 13 discusses the problem generally.

⁶⁴ Duncan-Jones, op. cit. (n. 1), 339, and Harris, op. cit. (n. 6), 265–7, produce figures, with minor differences, for the Italian regiones. Jongman, op. cit. (n. 6), 68–9, maps Duncan-Jones' figures showing the close fit with levels of urbanization in peninsular Italy. Harris, on p. 268, extends the analysis to the western provinces.

⁶⁵ On Romano-British epigraphy, Biró, op. cit. (n. 3), and Mann, op. cit. (n. 6), on the basis of RIB I. For further analysis, cf. Blagg, T., ‘Architectural munificence in Britain: the evidence of inscriptions’, Britannia 21 (1990), 13–32CrossRefGoogle Scholar, comparing Romano-British building inscriptions with those from non-Mediterranean Gaul and Germany analysed by Frézouls, E., ‘Evergétisme et construction urbaine dans les Trois Gaules et les Germanies’, Revue du Nord 66 (1984), 27–54CrossRefGoogle Scholar.

⁶⁶ Mann, op. cit. (n. 6), 205.

⁶⁷ Goudineau, C. in Duby, G. (ed.), Histoire de la France Urbaine I. La ville antique (1980), 49Google Scholar, gives an idea of the general pattern. I hope to publish a more detailed analysis elsewhere. On the North, cf. Wightman, op. cit. (n. 1), 162–77.

⁶⁸ For the significance of these groups, cf. Purcell, op. cit. (n. 58), Crook, J. in JRS 82 (1992), 233–4Google Scholar, Wallace-Hadrill, op. cit. (n. 60).

⁶⁹ cf. MacMullen, R., ‘The legion as society’, Historia 33 (1984), 440–56Google Scholar, reprinted in idem, Changes in the Roman Empire. Essays in the Ordinary (1990), 225–35, for this perspective on military society. On the extent to which socialization obliterated previous social identities, cf. Shaw, B. D., ‘Soldiers and society: the army in Numidia’, Opus 2 (1984), 133–59Google Scholar. The prominence of some legionary veterans in their home towns is well-known, and is demonstrated by their legal classification alongside decuriones and other honestiores from the second century on; cf. Garnsey, P. D. A., Social Status and Legal Privilege in the Roman Empire (1970), at 245–51.Google Scholar

⁷⁰ Naturally not all rural societies were alike and in some, perhaps the village societies of Egypt, Syria, and Anatolia, gradations of social and economic status, economic activity and hence some forms of social mobility may have been close to conditions in smaller towns. Nor can the existence of moderately well-off or even wealthy peasantries be excluded a priori. Cities and the military camps that resembled them were, however, clearly different in kind and scale.

⁷¹ cf. Freeman, P. W. M., ‘“Romanisation” and Roman material culture’, JRA 6 (1993), 438–45Google Scholar for an insightful critique. I hope to return to this issue elsewhere.

⁷² cf. n. 1 above.

⁷³ pace Meyer, op. cit. (n. 1).

⁷⁴ pace Morris, op. cit. (n. 9), 168–70.

⁷⁵ But on a much smaller scale and with significant formal differences. For an illustration, cf. Roueché, op. cit. (n. 1), on Aphrodisias. Some 1,500 inscriptions survive from the first two hundred and fifty years A.D., but only 230 between then and the mid-sixth century. Public inscriptions declined markedly but private epitaphs are also very rare. The late imperial epigraphy of Aphrodisias is nevertheless very prolific compared to that of most Asian cities. The style and orthography of late inscriptions is also very different to those of the early Empire.

⁷⁶ e.g. (and recently) on Italy, Ward-Perkins, B., From Classical Antiquity to the Middle Ages. Urban Public Building in Northern and Central Italy (1984)Google Scholar; on Anatolia, Mitchell, S., Anatolia. Land, Men and Gods in Asia Minor (1993), I, 211–17Google Scholar; on Britain, Perring, D., ‘Spatial organisation and social change in Roman towns’, in Rich, J. and Wallace-Hadrill, A. (eds), City and Country in the Ancient World (1991), 273–93CrossRefGoogle Scholar; and more widely cf. the papers collected in Rich, J. W. (ed.), The City in Late Antiquity (1992).CrossRefGoogle Scholar

⁷⁷ On urban change, cf. most recently Barnish, S. J. B., ‘The transformation of classical cities and the Pirenne debate’, JRA 2 (1989), 385–400Google Scholar; W. Liebeschuetz, ‘The end of the ancient city’, in Rich, op. cit. (n. 76), 1–49.

⁷⁸ Alföldy, op. cit. (n. 1); Hölscher, T., Staatsdenkmal und Publikum. Vom Untergang der Republik bis zur Festigung des Kaisertums in Rom (1984)Google Scholar; Zanker, P., The Power of Images in the Age of Augustus (1988).Google Scholar