¹ See, e.g., the honorific family inscriptions from Athens, , IG ii. 2. 3822–4255Google Scholar; Delos, , IG xi. 4. 1166–1200Google Scholar; Laconia, , IG v. 455–613Google Scholar; Priene, IPriene 259–85. Most epigraphical corpora contain a section of private dedications of this nature.

² See IG xii. 1. 103–20; 813 ff.; the inscriptions in ILind (see n. 3 below) are arranged chronologically, not by type, but see the various dedications set up to priests of Athana Lindia by members of their families, ILind 96 ff.

³ See Fraser, P. M., Rhodian Funerary Monuments (Oxford 1977) 58 and nn. 323–5.Google Scholar (The Rhodian epigraphical publications mentioned throughout this article are explained ibid. 83.) Complex ‘family monuments’ other than the ones given there are IG xii. 1.72b, 107 (on these two see below); 108; 110; 111; AD 18 (1963) Mel. p. 12 no. 17; AD 21 (1966) Chron. p. 447 no. α; Clara Rhodos 2, p. 190 no. 19; Supp. rod. 18–19; 26; TCam 84; 87; 91. The Lindians seem to have preferred large communal exedrae in honour of several members of one family; cf. ILind 129; 131; 142; 197–8; 203; 244; 293; 299–300; 382; 384; 455; 465.

⁴ Charinos appears in Blinkenberg's list of sculptors known to have worked on Rhodes, ILind col. 56 no. 95. For epidamia, see Fraser (n. 3 above) 48; it was the right of permanent residence at Rhodes and seems to have functioned as a form of naturalization. The rationale of the system and its legal intricacies are not completely understood, but the right of epidamia seems to have been granted largely to sculptors, who always note the fact in their signatures. The children of the recipients of epidamia were entitled to use the ethnic Ῥόδιος. A glance at the list of sculptors in ILind shows this progression clearly; cf. col. 54 no. 74, Epicharmos of Soli. This sculptor first signs as Ἐπίχαρμος Σολεὺς ἐποίησε (ILind 234); a later base was signed after Epicharmos had been given the right of epidamia: Ἐπίχαρμος Σολεὺς ᾧ ἁ ἐπιδαμία δέδοται ἐποίησε (ILind 235); finally, a base was signed jointly by him and his son: Ἐπίχαρμος Σολεὺς ᾧ ἁ ἐπιδαμία δέδοται καὶ Ἐπίχαρμος Ἐπιχάρμου Ῥόδιος ἐποίησαν.

⁵ This conclusion is also reached by Bresson, A., DHA 7 (1981) 345 ff.CrossRefGoogle Scholar

⁶ I have assumed that Damosthenes is a brother, not a brother-in-law, of Antilochos because of his name. Since the nephew of their mother Artemisia is named Damosthenes, it seems likely that he and her (in this case) son were both named Damosthenes after an older relative in her family. Neither Artemisia nor her brother Phaullos have their patronymic preserved (lines 3–4), but if both named their respective sons Damosthenes there is a good chance that their mutual father also bore this name.

⁷ Although πάππος is restored in line 7 it is surely right; since the honorand's wife, daughter, and brothers-in-law have already been mentioned, grandchildren are the obvious category of relative remaining to honour an older man who has held the priesthood of Athana Lindia.

⁸ Cf. Fraser, op. cit. (n. 3 above) n. 325.

⁹ Peek, W., ‘Epigramme und andere Inschriften von Nisyros’, Wissenschaftl. Zeitschrift der Univ. Halle 16 (1967) (6) 374 ff.Google Scholar

¹⁰ The name Phileia is uncertain but can probably be accepted; it will be shown below that the same woman appears in IG xii. 3. 103, 2 where slightly more of the name is preserved: [Φί]/[λ]εια.

¹¹ The historical problems surrounding this inscription need only be mentioned here since my interest lies in the dedicatory part of it. For a full discussion of the status of Nisyros at the end of the third century, see Fraser, P. M. and Bean, G. E., The Rhodian Peraea and Islands (Oxford 1954) 147–52Google Scholar, with whose conclusions I agree; for a different view see the full discussion in Holleaux, M., REG 30 (1917) 95 ff.Google Scholar ( = Études 4. 169 ff.).

¹² In which case Nisyros must have been incorporated into Rhodian territory by that period; Holleaux, op. cit. (see n. 11 above) argued that the man's career structure (lines 7–13) is given in chronological order, and that he was a Nisyrian strategos of Rhodes in the First Cretan War (205–201 BC (line 7), and later was στρατευσαμένος (line 11) on a campaign under nauarchs known to have served during the years 201–190 BC (cf. Fraser and Bean, op. cit. (n. 11 above) 148 and n. 6).

¹³ In which case the man's career does not chronologically follow the sequence of the inscription, since his position as strategos (in this case in 156–153 BC) is mentioned before he ‘served as a rating’ in the years 201–190 BC. The reasons for the inversion of roles in the inscription are convincingly argued by Fraser and Bean, op. cit. (see n. II above) 149 ff. Their conclusion is that there is no evidence that Nisyros was Rhodian as early as the First Cretan War, although it had become Rhodian by the end of the century (after which time the honorand served under the nauarchs of 201–190 BC); we may note that Philip V wrote to Nisyros as an independent community shortly before this time; cf. IG xii. 3. 91 ( = Syll. ³ 573).

¹⁴ Fraser and Bean, op. cit. (see n. 11 above) 150; cf. my own discussion in JHS 104 (1984) 185–6.

¹⁵ Cf. ILind col. 54 no. 74.

¹⁶ Peek, op. cit. (see n. 9 above) 376, attempts a reconstruction of the family's stemma and the relative chronology of the two inscriptions, but I find his discussion baffling. He accepts a date of 205–201 BC (i.e. the First Cretan War) for IG xii. 3. 103, and says, p. 375, that Phileia the daughter of Gnomagoras in 103, 2 was evidently named after her aunt Phileia, the mother of the dead Sosagoras in Inschr. Nis. 3. He devises his stemma (p. 376) from this assumption. I cannot understand how Phileia (the granddaughter of 103, 2) can be the niece of a woman who commemorates the death of a young son some seventy-five years later. His reasoning is surely backwards.

¹⁷ Peek, op. cit. (see n. 9 above) 376, also arrives at this conclusion, since Sosagoras the elder is the paternal grandfather of the boy Sosagoras, but his stemma is not correct elsewhere (see below).

¹⁸ Ibid. 375.

¹⁹ Peek does not in any event explain the genitive patronymic Λαϊσθένευς in line 5. This can only be the patronymic of the Onasiphon in line 4, which shows that he is not the same man as Onasiphon the son of Sosagoras.

²⁰ We may compare the complete family dedication NS (see n. 3 above) 18, where a man with a long and distinguished career is honoured only by his grandchildren.

²¹ Peek, op. cit. (see n. 9 above) 375, supposes that it is the same Autareta in both inscriptions, but how can the girl who has lost a young brother in Inschr. Nis. 3 (c.125–100 BC) honour her grandfather (by Peek's chronology) some seventy-five years earlier?

²² Peek, op. cit. (see n. 9 above) 376.

²³ The inscription has been previously published by Loewy, E., AEMÖ 7 (1883) 108 n. 1Google Scholar; Inschriften Griechischer Bildhauer (Leipzig 1885) 189. I am grateful to Dr J. Ch. Papachristodoulou, The Ephor of Antiquities of the Dodecanese, for permission to photograph and take a squeeze of this and many other stones in the Rhodes Museum in the spring of 1983.

²⁴ One may compare in this context another family monument (ILind 56D–E, c.313 BC), where there are included two wives of the priest of Athana (Polykles the son of Polykles), Antigone and (evidently) her sister, Gorgia, the daughters of Archonidas.

²⁵ The presence of Menekrates I and II in IG xii. 1. 46–7 was pointed out by Holleaux, , RPh 1893, 177Google Scholar, but he wrongly identified the pair at the beginning of IG xii. 1. 107, 17, where the correct reading is in fact [Παρμ]ενί[ω]ν Μενεκράτευς. Menekrates the son of Menekrates does not participate in 107, but his existence can be deduced from the reconstruction of the stemma of this family. It is possible that this man is the father of a third Menekrates who is priest of Athana Lindia in 30 BC (ILind 1, sub ann.; 378b, 21), Menekrates the son of Menekrates, adopted son of Klearchos. However, it is odd that Menekrates the priest (if the son of our Menekrates II) does not appear in this dedication along with (in this case) his brother Charinos the son of Menekrates.

²⁶ Kontorini, Vassa, Inscriptions inédites relatives à l'histoire et aux cultes de Rhodes au 2e et au Ier s.av. J–C: Rhodiaka i (Louvain 1983).Google Scholar

²⁷ It is an interesting point of relative chronology that this sculpture is signed by the Tyrian sculptors Charmolas and Menodotos the sons of Artemidoros (cf. ILind col. 55 no. 82; cols. 35–6 stemma 6). Charmolas is the first of his family to receive the right of epidamia, and is the father of another Menodotos who is entitled to call himself Ῥόδιος (ILind col. 55 no. 85). This sculptor later signs (as senior partner) a base with Charinos of Laodicea, the sculptor of the family monument IG xii. 1. 107. So, when Menekrates is a hieropoios in his youth, his name appears on a base of the sculptor Charmolas, the father of the future senior partner of Charinos, the sculptor of the family monument when Menekrates (a grown man with three children of his own) honours the son of his cousin.

²⁸ Cf. n. 3 above.

²⁹ Kontorini, op. cit. (see n. 26 above) 70; on p. 71 she calls the honorand of IG xii. 1. 107 the ‘fils de la nièce’ of Menekrates, but he is the son of the cousin of Menekrates.

³⁰ If this identity is accepted, Damostratos the priest of Apollo Erithimios in 42/41 BC would be a contemporary of Aineas the son of Charmeios, priest of Athana Lindia in 50 BC.

³¹ Hiller, (IG xii. 1. 107 adnot. ad loc.)Google Scholar was confused by the term θ[ῖ]ον at the end of line 18, and he allowed the possibility that there were two people named Aineas, one the cousin who is the son of Charmeios, the other the father of the nephew Charmeios, but he did not understand how the son of a cousin of the honorand's mother could also be the nephew of the honorand. The existence in one family of two homonymous pairs in the same generation (Charmeios the son of the priest Aineas who was a mastros of the Kamyndioi, and Charmeios the son of Aineas, the nephew of the honorand)—even in such a complex family as this—is surely improbable, and it is much easier to assume that they are the same man.

³² In later family monuments, a marriage and a blood relationship are occasionally both expressed; cf. ILind. 455 (c.AD 80–100), where a man honours a woman who is both his niece and his wife, τὰν αδελφιδέαν καὶ γυναῖκα; NS (n. 3 above) 26, where a man honours his cousin and wife, τὰν ἀνεψιὰν καὶ γυναῖκα. In both cases, the blood relationship is mentioned first.

³³ See Fraser and Bean, op. cit. (n. 11 above) 147 and n. 1.

³⁴ Cf. note ad loc; van Gelder, H., Geschichte der alten Rhodier (The Hague 1900) 183.Google Scholar

³⁵ For the family see ILind cols. 35–6 stemma 6.

³⁶ The Scylla group has now been beautifully restored and displayed in the Sperlonga Museum. I am particularly grateful to Soprintendente Baldassare Conticello for a personal tour of the museum in October 1984, and for giving me the benefit of his views on the sculpture. I know that he will disagree with the conclusions I draw about the date of the sculptors of the Sperlonga groups, but I hope that he will find my arguments thought-provoking.

³⁷ JRS 67 (1977) 76 ff.; hereafter called ‘Stewart’.

³⁸ Ibid. 77 n. 5.

³⁹ ILind cols. 29–30 stemma 1.

⁴⁰ Arch. Class. 10 (1958) 160 ff.; L'Antro di Tiberio a Sperlonga (I Monumenti Romani IV) (Rome 1963) 39 ff. Jacopi discusses the right pieces of evidence but does not reach the correct conclusions: he picks out the artist but is confused over the relative chronology of the relevant Rhodian inscriptions and devises an inaccurate family tree. His errors stem partly from the fact that he believes that the sculptures date to the very end of the second century or to the beginning of the first century BC.

⁴¹ The Polyphemus and Scylla Groups at Sperlonga (Stockholm 1972) 73 ff.

⁴² Dialoghi di Archeologia 7 (1973) 112 ff. Coarelli sorted out the various pieces of chronology more satisfactorily than Jacopi but still tried to make two of the sculptors into a family and ends up with a confused stemma. He thinks that some of the Sperlonga groups were made in Rhodes in c.80–40 BC, and then moved to Italy much later.

⁴³ Stewart, 86.

⁴⁴ Cf. von Blanckenhagen, P. H., AA 1969, 257–63Google Scholar; Andreae, B., Antike Plastik 14 (1974) 93, 102, 104Google Scholar; von Blanckenhagen, , AJA 80 (1976) 102–3.CrossRefGoogle Scholar I am not convinced by Andreae's argument that Ovid's Metamorphoses xiii was the source for the picture of Odysseus presented in the so-called Pasquino and Palladium groups. See Stewart, 78, for much earlier sources for this picture of Odysseus, and p. 82 for the difficulty of proving that the sculptures are dependent upon the treatment of their themes in Augustan poetry. In addition, Stewart is of the opinion that the Laocoon does not tally with Vergil's account.

⁴⁵ Stewart, 77, argues that the installation of the groups cannot be earlier than c.30 BC, is most likely in the late Augustan or Tiberian era, and is possibly as late as Nero or Vespasian; cf. also pp. 88–90.

⁴⁶ See esp. von Blanckenhagen, , AJA 80 (1976) 99 ff.CrossRefGoogle Scholar

⁴⁷ According to von Blanckenhagen, ibid. 103, the so-called Pasquino and Palladium groups were slightly varied replicas of bronze groups of the first half of the second century BC; the Polyphemos group was an adaptation of a famous Greek composition which probably stood in a public place, and appears to have been a much freer, more colossal version of the earlier, more normal-sized piece; the Scylla group was a recreation of a theme variously depicted earlier, both in a famous Hellenistic group, and (probably) in a Hellenistic painting. All this being so, one point must be made clear about the signature. As Robertson, M. (A History of Greek Art (Cambridge 1975) 541)Google Scholar has correctly pointed out, ‘a signature in this form is that of the contemporary sculptors who carried out the work, not of artists of an earlier age whose creations are here reproduced’. The three sculptors were responsible for the Sperlonga groups in their existing form, not for the earlier Hellenistic prototypes which have been conjectured as the originals from which the designs from the grotto were re-created and combined.

⁴⁸ This follows from the presence of the various struts on the statues (e.g. those between the toes of Polyphemos), which can only be explained as safeguards against damage in transport.

⁴⁹ If, as seems likely, the Vatican Laocoon is the one mentioned by Pliny; cf. von Blanckenhagen, , AJA 80 (1976) 103.CrossRefGoogle Scholar Although the identification of the Laocoon and Sperlonga sculptors has presented some difficulties, since there are marked differences in style and quality between the Sperlonga groups and the Laocoon, it has been accepted by von Blanckenhagen, ibid.; AJA 77 (1973) 458 ff., and is assumed with greater conviction by Stewart, 76–7, who points to the unevenness of style within and among the Sperlonga groups and within the Laocoon as an indication that the same hands were at work on all five groups.

⁵⁰ von Blanckenhagen, , AJA 80 (1976) 103.CrossRefGoogle Scholar

⁵¹ Ibid.; cf. Robertson, op. cit. (n. 47 above).

⁵² I am grateful to The Lexicon of Greek Personal Names for providing this and the other information in this paper on the numerical frequency of names on Rhodes.

⁵³ Blinkenberg, , RM 42 (1927) 177 ff.Google Scholar

⁵⁴ Fraser, , Eranos 51 (1953) 42 n. 10.Google Scholar There can in fact be little doubt that their successive appearance should be connected with the tribal cycle of Lindian demes which governed the appointments to the priesthood of Athana Lindia; see ibid. 23 ff. In the operation of this system, particular provision was made for priests who entered their office by way of adoption into another deme and who thus actually broke the natural cycle while technically maintaining it. In this connection it is explicitly stated in ILind 419, 86–8 (AD 22) that priests by adoption were subject in toto to the same body of regulations as the priests who entered the cycle by direct relationship. Either Athanodoros was priest in 22 BC as a representative of a deme of tribe G in the cycle (see Fraser, 29 sub ann. 22 BC), whether or not by adoption, or his brother (if it is indeed his brother) entered by adoption a deme other than his own which belonged to a different tribe (A in the cycle). Since both the supposed brothers were adopted, and the deme of their birth is unknown, it is not possible to determine in which case the irregularity occurred, although it may conceivably have occurred in both. Blinkenberg, ILind cols. 29–30 stemma 1, suggests that the whole family may have been Lindopolitai, but there is no evidence for this, and it will be shown below that his whole stemma must be revised.

⁵⁵ Richter, G. M. A., Three Critical Periods in Greek Sculpture (Oxford 1951) 66–70Google Scholar, considered these questions in connection with the date of the Laocoon, before the discovery of the grotto of Sperlonga. Because of her stylistic analysis of the Laocoon as an original of the early second century BC, belonging to the school of Pergamene ‘baroque’, Richter was eager to disprove any connection of the Rhodian priests of 22 and 21 BC either to the first-century BC sculptor Athanodoros known from a series of signed bases (see below, p. 239, listed as her nos. 1–7, p. 67), or to her, as she thought, early second-century BC sculptor Athanodoros, one of the sculptors of the Laocoon. Her discussion of the Rhodian epigraphical evidence was very confused, and was corrected by Fraser, op. cit. (see n. 54 above) 45–7. (However, on his p. 46, second paragraph, her quoted reference to IG xii no. 847, the honorific decree for Athanodoros which Richter misinterpreted as being the priest-list, should read IG xii. 1. 847, not IG xii. 1. 857.) After the discovery of the Sperlonga sculpture, Richter stuck to her guns about an early second-century date for both it and the Laocoon, , The Sculpture and Sculptors of the Greeks (4th edn., New Haven 1970) 236 ff.Google Scholar She still, however, does not deal with the Rhodian evidence satisfactorily. On p. 237 n. 50, Richter acknowledges the corrections of Fraser, but again quotes IG xii. 1857 (this time read IG xii. 1. 847) as the inscription ‘citing the 2 brothers Athanodoros and Hagesander as priests of Athena Lindia in 22 and 21 BC’. The brothers are cited as priests only in the list of priests, ILind 1 sub annis 22 and 21 BC; IG xii. 1. 847 is the honorific decree passed in honour of Athanodoros by the Lindians, and makes no mention of his tenure of the priesthood. Furthermore, on the evidence of the Sperlonga inscription she states, p. 237, that ‘these 3 sculptors, therefore, had different fathers from those conjectured for them, and so the whole hypothesis that connected them with the 1st century inscriptions falls to the ground’. This is indeed the case with Hagesandros, as stated above, and it removes any connection of him to the priest Hagesandros, but it does not in fact affect Athanodoros, who has the same patronymic in all the inscriptions as well as in the priest-list. Of particular importance is ILind 347, which records Athanodoros son of Hagesandros as a sculptor in 42 BC, his only securely dated signature (on the importance of this inscription, see below). Richter mentions this inscription, but does not take it into account in her discussion.

⁵⁶ It is possible that the name of the father of this naval officer, Hagesandros son of Athanodoros, appeared in the fragmentary dedication ILind 271, dated by Blinkenberg to c.100 BC, but the restoration cannot be certain because of the frequency of the name Athanodoros at this period.

⁵⁷ On the relative social standing of sculptors in Greece and Rome, see Smith, R. R. R., JRS 71 (1981) 24 ff.Google Scholar Although the Greeks always looked down on manual labour as such, it seems that sculptors escaped this stigma because of the art they produced. In Greece, unlike Rome, sculptors had a relatively high social standing, and there is no reason to assume that the sculptor Athanodoros came from a family of manual labourers. It has been shown above that Hagesandros the son of Paionios belonged to a reasonably wealthy, respectable Rhodian family with some social standing.

⁵⁸ Cf. ILind. col. 56 no. 102 (Blinkenberg's stemma 1, referred to ad loc., can no longer be accepted). In Loewy, op. cit. (see n. 23 above), IGB, the only other sculptor with the name Athanodoros signed an archaic base from Olympia (no. 30a). Conticello, , Antike Plastik 14 (1974) 44 ff.Google Scholar, would agree with Richter, op. cit. (see n. 55 above) that a second-century BC date for the Sperlonga sculptures proves that there is a second-century BC sculptor Athanodoros, but this is a circular argument since the sculptures can only be dated by subjective stylistic analysis. I would be the first to agree to the existence of an earlier artist if there were independent evidence that the sculptures were second century, but since there is not we must begin from the attested existence of a sculptor in the second half of the first century BC.

⁵⁹ Overbeck, J., Die antike Schriftquellen zur Geschichte der bildenden Künste bei den Griechen (Leipzig 1868).Google Scholar Another passage in Pliny mentions an ‘Athenodorus’ but he is given no ethnic (HN xxxiv. 86 = SQ 2032: ‘nunc percensebo eos qui eiusdem generis opera fecerunt, ut … Athenodorus feminas nobilis’).

⁶⁰ Torr, C., Rhodes in Ancient Times (Cambridge 1885) 100Google Scholar refers to the signature of Athanodoros ‘on a small vase of white marble found at Olympia’. He apparently confuses no. 5 above with the Olympia base (IOlympia 239 = Kaibel, , Epigrammata Graeca (Berlin 1878) 934)Google Scholar, which is in fact in honour of the Rhodian Olympic victor Theopropos of the second to third centuries AD; cf. Moretti, L., Olympionikai, i vincitori negli Antichi Agoni Olimpici (Atti della Accademia Nazionale dei Lincei) ser. 8 vol. 8 fasc. 2 (Rome 1957) 895.Google Scholar

⁶¹ Robert, L. (ed.), Collection Froehner i: Inscriptions Grecques (Paris 1936).Google Scholar

⁶² For a description of this piece, see now Zevi, F., ‘Monumenti e Aspetti culturali di Ostia repubblicana’, in Zanker, P. (ed.), Hellenismus in Mittelitalien, Gött. Abh. 97 (1) 1976, p. 61 with figs. 26 7.Google Scholar

⁶³ See Fraser and Bean, op. cit. (n. 11 above) 52 ff. for the definition of the term peraea and a description of the Rhodian Peraea in antiquity, and pp. 59 ff. for the site of Loryma. Loryma was not a Rhodian deme centre (no known demotic is derived from its name), but appears to have been a settlement around the large Rhodian fort situated on the headland. I believe that the inhabitants there may have been assigned to the large Peraean deme of the Kasareis, situated nearby at modern Asardibi.

⁶⁴ See the lemma ad loc. for a description of the find-spot and of the arrangement of the various blocks of the monument.

⁶⁵ For the struggle between Rhodes and Cassius, see Hiller von Gaertringen, RE Supplbd. 5, cols. 806–7; van Gelder, op. cit. (n. 34 above) 169 ff.

⁶⁶ Blinkenberg, ILind col. 681, rightly pointed out that ‘L'inscription est d'une importance extraordinaire pour la chronologie et l'histoire de l'art’.

⁶⁷ See Torr, op. cit. (n. 60 above) 106 ff., for Rhodian artworks which were found at Rome. Valerius Maximus (i. 5) and Dio Cass. (xlvii. 33) said that Cassius carried off all the statues from Rhodes except the quadriga of Helios, but it is not likely that he encumbered himself to such an extent during a campaign. A century later there were still 3,000 statues at Rhodes (Pliny, , HN xxxiv. 17Google Scholar), and the island suffered less than the rest of Greece from the Roman passion for collecting (Dio Chrys. xxxi. 147).

⁶⁸ ILind col. 90.

⁶⁹ See ibid. n. 2 for examples of the phrase elsewhere. For analogous acknowledgements in Rhodian inscriptions (one from Ialysos and one from the Rhodian Peraea) about the renewed peace, which Blinkenberg ties into the aftermath of the same troubled period, see Clara Rhodos 2. 184 n. 10; Lindiaka viii. 8 n. 22. Blinkenberg notes, ILind col. 90 n. 2, that the Rhodian situation suddenly changed after the Battle of Philippi in the autumn of 42 BC. At this time there appear a series of χαριστήριον inscriptions from ‘the Lindioi’ to Athana; cf. ILind 364 ff. plus comm. ad loc.

⁷⁰ Stewart, 89 n. 120.

⁷¹ Cf. ibid. n. 121; Robertson, op. cit. (n. 47 above), 541.

⁷² Stewart, 89 n. 122.

⁷³ Ibid. n. 123; Conticello, op. cit. (see n. 58 above) 32 ff. n. 36. This is clearly seen in Fig. 2.

⁷⁴ This is apparently the view taken by Conticello, who discussed it with me in a private conversation. He is faced, however, with the need to reconcile the sculpture which he believes dates to the end of the second century BC with an inscription which he believes dates to the first century AD. He compromises by saying that the inscription was carved at a much later date when the sculptures had been moved from Rhodes to Sperlonga, some 100 years after their creation. The original signatures (contemporary with the sculpture) would have been carved on the base of the ship in the Scylla group at Rhodes. If the base was left behind when the statues were removed, a replacement ship and a new inscription would have had to be made when the group was reassembled in Italy. This reasoning could account for an Imperial-looking inscription on Hellenistic-looking sculpture. I do not believe that the sculptures could have been moved from any other setting since the evidence is overwhelming that they were designed for Sperlonga (see below). If this is the case, there is no reason why the ship should originally have been left uncarved when the statues were made, and an inscription added much later. Conticello's other argument is that the present inscription is a replacement for one that had been damaged. That may be so, but it still tells us nothing about the date of the sculptures.

⁷⁵ Cf. Robertson, op. cit., in n. 47 above.

⁷⁶ Jacopi, , Arch. Class. 10 (1958) 160Google Scholar, says that the ship is definitely ‘Lithos Lartios’ Hampe, R., Sperlonga und Vergil (Mainz 1972) 42Google Scholar, says that the inscription and the ship's stern are ‘Lithos Lartios’; von Blanckenhagen, , AJA 78 (1973) 459Google Scholar, suggests that the Polyphemus and Scylla groups are ‘Lithos Lartios’ (‘being peculiar fine-grained bluish-streaked marble, generally called Rhodian’); it is unequivocally identified as ‘Lithos Lartios’ by Lauter, H., RM 76 (1969) 165.Google Scholar

⁷⁷ That it is Rhodian stone is denied by Stewart, 76 n. 3, and Merker, G., The Hellenistic Sculpture of Rhodes (Göteborg 1973) 21 n. 13.Google Scholar

⁷⁸ Ibid. 6 with n. 21; Merker points out that ‘Lithos Lartios’ is unsuitable for large-scale sculpture.

⁷⁹ Cf. von Blanckenhagen, , AJA 80 (1976) 103CrossRefGoogle Scholar; Stewart, 88–9.

⁸⁰ Lauter, , Antike Kunst 15 (1972) 49–59.Google Scholar See esp. pp. 57 ff.

⁸¹ AJA 77 (1973) 460.

⁸² Compare Constantinopolis' discussion of Rhodian nymphaea with statue niches, Archaeology 21 (1968) 118 ff.

⁸³ This view has been expressed by others, with whom I heartily concur; Merker, op. cit. (see n. 77 above), 24 n. 90, maintains that Rhodian grottoes had little influence on Sperlonga.

⁸⁴ Aside from the Sperlonga inscription, I know of only three instances of the name at Rhodes, : IG xii. 1. 7Google Scholar; 46, 425; 107, 15. Of these, only the first is unconnected to the Paionios in question here.

⁸⁵ For a sculptor appearing in a ‘non-artistic’ context, we may compare Mnasitimos son of Aristonidas, a member of a large family of Kamiran sculptors (for their family, see ILind cols. 43–4 stemma 15). Mnasitimos appears as a sculptor in ILind 56 (= ILind col. 51 no. 8); he appears elsewhere as a hieropoios in Kamiros (TCam (see n. 3 above) 13, 12).

⁸⁶ It may be significant that we have the strikingly rare name Paionios in the family of Hagesandros. If sculpting was a family profession (regardless of whether or not Hagesandros was related to his Sperlonga colleagues), might this name not recall the famous Classical sculptor Paionios of Mende, best known for his alighting Nike at Olympia? The suggestion that artists' names may occasionally recall famous artists of the past can be supported by a second-century sculptor Polykleitos of Argos who worked at Rhodes (ILind col. 52 no. 23 = TCam 29). His name may recall the famous fifth-century sculptor Polykleitos. In addition, we may note that the father of the third-century sculptor Phyles of Halicarnassus (ILind col. 52 no. 24), Polygnotos, may recall the famous painter Polygnotos of Thasos. I am not aware of the name elsewhere at Halicarnassus.

⁸⁷ See, e.g., ILind cols. 51 ff., nos. 10, 16, 34, 36, 39, 40, 63, 83–4.

⁸⁸ Arch. Class. 10 (1958) 162.

⁸⁹ Stewart, 87 n. 101, says that the letter-forms of all the other Athanodoros signatures are apparently early Imperial; cf. Säflund, op. cit. (n. 41 above) 98, who agrees. Some of them look considerably later than that to me.

⁹⁰ See Stewart, 89 n. 119; Cf. Zevi, op. cit. (n. 62 above), 60 figs. 18 20, who records new bases found in Ostia in the precinct of Hercules for three plundered Greek bronzes, the works of Phradmon, Phyromachos, and Lysikles.

⁹¹ Stewart, 89 n. 119; Riemann, H., in Kunzinger, F. (ed.), Forschungen und Funde: Festschrift Neutsch (Innsbruck 1980) 378.Google Scholar

⁹² Coarelli, op. cit. (see n. 42 above), 114; Riemann, op. cit. (seen. 91 above), 377; the latter does not, however, consider the chronology of Athanodoros properly, and settles for a Tiberian date for the sculptures on stylistic grounds.

⁹³ Sculpture did, however, continue to be produced in quantity on the island. We may also note that all of the top artists in Italy were Greek, either free Greeks or Greek freedmen; cf. Smith, op. cit. (n. 57 above), 29.

⁹⁴ Stewart, 89 n. 119.

⁹⁵ Suet. Tib. 11. 1; Levick, B., Latomus 31 (1972) 779 ff.Google Scholar His only visit to the East at this point had been one of a few weeks or months in 20 BC, when he had put in to Rhodes on his return voyage from Armenia.

⁹⁶ Mél. d'Arch. et d'Hist. 98 (1) (1981) 297 ff.

⁹⁷ D'Arms, John H., Romans on the Bay of Naples: A Social and Cultural Study of the Villas and their Owners from 150 BC–400 AD (Cambridge, Mass. 1970)Google Scholar; McKay, A. G., Houses, Villas, and Palaces in the Roman World (London 1975) 100–36Google Scholar; cf. Riemann, op. cit. (n. 91 above) 372 ff.

⁹⁸ Conticello, op. cit. (see n. 58 above) 40 ff. (esp. p. 49).

⁹⁹ von Blanckenhagen, , AJA 80 (1976) 104–5CrossRefGoogle Scholar, is disinclined to believe this.