II. NUMINA AUGUSTORUM*

Shortly before the death of , Tiberius dedicated the celebrated Ara Numinis Augusti, thus formally enshrining the of Augustus within the

Imperial Cult. I The step was a radical one, fundamental to the whole develop­ ment of the emperor's "divinity". Whereas the cult of the emperor's genius had continued a traditional Republican practice,2 to ascribe numen to the princeps was to establish Augustus as a 9!;to~ cX\lTjp, through whom divinity could function as an intermediary.3 For to pay cult to the Numen Augusti was to ascribe to the human emperor the quintessential property of a god. 4 The cult of the numen seems, however, to have been slow in establishing itself as a popular form of worship and to have been restricted in the areas where it took hold. S In the north-east and the east it occurs rarely, apart from the late

formula devotus numini maiestatique eiusl eorum, 6 and is attested only to a very limited extent in Africa, where numini Augustorum gives way in the third century to D.N.M.Q.E.7 Spain has produced scores of examples of the devotus formula, yet extremely few direct dedications. 8 But in Narbonensis and more especially in the Celtic provinces of the north-west evidence for the worship of the Imperial numen is very great indeed. In no other quarter of the Roman world, in fact, does the Imperial numen appear so frequently in dedications of the widest variety. The major concern here is with the form the cult took and what exactly this implies. In the vast majority of cases dedications are made to the Numina

• Classical Quarterly 20 (1970), 191-197.

I Degrassi,lnscrlt 13, 2, pp. 115,401: Praenestini ad 17th January. For arguments in support of a date ca. A.D. 6 see A. A1fbldi, Die zwei Lorbeerbaume des Augustus (Antiquitas 14), Bonn, 1973, 42-44. 2 "Augustus and the West", above, Vol. I, 1, pp. 84f. On the cult of the genius see in general Weinstock, DJ 205-217. 3 For the significance of the step from genius to numen see "Genius and Numen", above, p. 386); further "Divinity and Worship", above, Vol. I, 1, pp. 4lf. with note 71. 4 On the nature of numen see W. Pbtscher, "'Numen' und 'numen Augusti''', ANRW 2, 16, 1 (1978) 355-392. , Early examples occur at Narbo, CIL 12,4333 (= lLS 112): A.D. 11; Forum Clodii, ClL II, 3303 (= ILS 154): A.D. 18; Lepcis Magna, lRT 324a: A.D. 11-12; Tarraco, RlT 48: ?reign of Augustus. • D.N.M.Q.E. and variants first appear in the early third century; cf. H. G. Gundel, "Devotus numini maiestatique eius. Zur Devotionsformel in Weihinschriften der rbmischen Kaiserzeit", Epigraphica 15 (1953),128-150; A.D. Nock, CP 57 (1962), 115. 7 The commonest form of dedication in Africa is numini (never numinibus) Augustorum: else­ where this form is comparatively rare; cf. below, note 34. See further D. Fishwick, "Le Numen imperial en Afrique romaine," 115e Congres national des societes savantes, Avignon, 1990, Ve Colloque sur I'histoire et I'archifologie d'Afrique du Nord, forthcoming. 8 Etienne, Culte imperial 309-13. NUMINA AUGUSTORUM 389

Augustorum, either written in full or in a variety of abbreviations. 9 When the plural Augustorum is abbreviated to Augg.!Auggg., there is usually no doubt that the numina of two or more reigning Augusti are intended. 10 Examples of this type occur frequently in Britain, which is unique in attesting an early­ third-century cult of the numenlnumina of the living emperor or emperors. But what exactly is implied by numinibus AugustorumlAugustor.! Aug.; num(inibus?) Aug(ustorum)?11 The fact that the double or treble G is expressly used to denote reigning emperors certainly seems to imply that some other sense is intended when an abbreviation has only the single G. Must one then assume that if the reduplicated G denotes living emperors to the exclu­ sion of the dead, therefore the single G denotes dead emperors to the exclu­ sion of the living? The question is prompted by the opinion of RIB on the sub­ ject, stated in the editorial note ad RIB 152 (p. 49): " ... In an abbreviated form, however, the sole means of distinguishing between the numina of two or more reigning Augusti and those of deceased emperors treated collectively [my italics] would be to use for the former the normal AUGG. or AUGGG., and for the latter AUG. with the plural NUMINA ... " The question is an important one since it is basic to the whole cult of the Imperial numen. To whom is the cult actually addressed in its most frequent manifestation, that is, in dedications to the Numina Augustorum? First, a look at the problem in the context of the ruler cult as a whole within the general area of the Celtic World. It is important to observe that in the Gauls, the Germanies and Britain emperor worship was something different from what it was in the Romanized provinces of the West. In Narbonensis, Baetica and Proconsularis, for example, a great deal was originally left to local initiative and the establishment of an official provincial cult deferred until the time of Vespasian. 12 But in newly pacified territories the cult per­ formed the important role of focusing loyalty on the person of the emperor and helped to further the process of Romanization; hence its installation at

• On the expansion of the various abbreviations and the general nature of the cult in Britain see "The Imperial Numen in Roman Britain", below, pp. 397-422. On abbreviated forms in the Germanies and the Gauls see J. Deininger, "Numinibus Augustorum. Anmerkungen zur Datierung der Trierer Bronzeprora", Germania 44 (1966), 138-142. I' But see "The Imperial Numen in Roman Britain", below, p. 404. In elL 13, 1710 it is possi­ ble that Augg. may signal the inclusion of two living emperors (Severus and Caracalla) with deified dead emperors. See "The Severi and the Provincial Cult of the Three Gauls", above, Vol. I, 2, 321f.

11 For the expansion of num. in the plural see "The Imperial Numen in Roman Britain", pp. 407-412.

12 For Narbonensis see "The Provincial Cult of Gallia Narbonensis: Three Temples at Narbo", above, Vol. I, 2, pp. 240-256; for Baetica: "The Origins of the Provincial Cult of Baetica", ibid. pp. 219-239; for Proconsularis: "The Foundation of the Provincial Cult in Africa Procon­ sularis", ibid. pp. 257-268.