BMCR 2006.03.19

The Roman Nude: Heroic Portrait Statuary 200 BC-AD 300. Oxford Studies in Ancient Culture and Representation

, The Roman nude : heroic portrait statuary 200 B.C.-A.D. 300. Oxford studies in ancient culture and representation. Oxford: Oxford University Press, 2005. xxi, 391 pages : illustrations ; 26 cm.. ISBN 0199240493. $150.00.

I am not a Romanist and therefore should not have undertaken to review a book focusing on Roman statuary. My reasons for doing so, however, are twofold: my appreciation of the author’s published work;1 and my interest in the Greek tradition that forms the basis for most of the Roman renderings. Moreover, Hallett’s text (a considerably expanded version of a 1993 Berkeley dissertation) was virtually completed in 1999 (p. v), but delays in obtaining illustrations required revisions and a much later date of publication, although only two new items could be referenced in the discussion. This review may therefore usefully cite some recent pertinent works that H. could not take into account.

The above-mentioned problems with photographs are occasionally reflected in a wrong cross-reference or a caption not in full agreement with the text, but this elegant book has otherwise remarkably few typographical errors.2 It also uses different lettering sizes and styles in chapter titles and subtitles that articulate the structure of the discussion, which in turn is based on the typological divisions of the catalogue (Appendix B). The latter lists 342 sculptures in stone and bronze, both in the round and in relief; the main text, however, includes also the evidence of coins and gems which, in some cases, provide the earliest examples of certain renderings. Because the book is intended not only for students of the ancient world but also for a general audience (pp. 3-4), all ancient and modern sources are cited in English, but their original versions are usefully given in footnotes. The seven basic chapters are followed by 13 Appendices (A-M) which, besides the Catalogue, cover peripheral aspects of the topic. The extensive bibliography is supplemented by occasional references in the footnotes. Further help is provided by an Index Locorum for the ancient authors; a Museum index for the portraits of the Catalogue; and a General Index. Acknowledgments, Table of Contents, a List of Figures, one of Plates, and one of Abbreviations complete the apparatus criticus.3

H. starts from the premise that our negative aesthetic reaction to Roman portrait statuary joining a youthful, nude (“Greek,” ideal) body to a mature, realistic head is a modern phenomenon that could not correspond to ancient Roman perceptions. In order to establish the proper historical context of such renderings, H. begins by exploring the possible Greek reasons for nudity in life and art (Chs. 1-2). He then extensively uses ancient sources to demonstrate that Romans viewed exposed bodies as improper and largely shameful, as for slaves and prisoners (Ch. 3), but eventually as acceptable in art through absorption of Greek practices and adoption of a specific symbolism (Chs. 4-5) which could be expanded by the use of divine attributes (Ch. 6), whose meaning is usually misunderstood by modern critics.4 The final, excellent chapter (7) returns to ancient quotations to show the Romans’ emphasis on heads and facial expressions—not on physiognomic grounds but as illustrating a person’s life and mores. A clever analysis of current approaches to eclectic (or rather, juxtaposed) styles—defined as our “Period Consciousness” and “Culture Consciousness”—is counterbalanced by a defense of “hybrid art” as practiced in antiquity. Typical examples are the Egyptian pharaonic sphinxes, or the Greek erotes, satyrs, and centaurs, which required a “suspension of disbelief” but which we too tend to accept, although we still react with discomfort when confronted with what we perceive as belonging to different periods of art—like a Fayyum Roman portrait on a typically Egyptian mummy case (Pl. 159). As H. concludes (p. 307), “Perhaps when we disparage such works with talk of ‘a confusion of cultures,’ and designate such pieces as ‘hybrid art,’ we say less about the statues than we think; and rather more about ourselves.”

This complex material is made accessible by a roughly chronological approach, for both Greek and Roman times, that tries to determine the initial appearance of various clothing formulas and their eventual disappearance or momentary omission. In addition, certain sections within chapters pull the various strands of the argument together: Ch. 4 contains two “Conclusions” (pp. 137 and 158 respectively); Ch. 5 ends with a lengthy “Conclusion: The Function and Meaning of the Nude Heroic Costume in Roman Portraiture” (pp. 217-222); and Ch. 6 summarizes “The Function and Meaning of the Nude Portrait with Divine Attributes” (pp. 259-264) before turning to “The Last Nude Portraits of the Ancient World.”

H. contends that nude bodies in Greek art were used to depict deities and heroes, or—in humans—athletes, who routinely disrobed during gymnastic practices. On Classical multi-figured gravestones, various forms of dress/undress signified generational differences. In the Hellenistic period, rulers were given “heroic” costume (nude with weapons) basically to symbolize “agonal” qualities and youthful vigor (p. 57). This type, especially in the Eastern Mediterranean areas, was then appropriated by Greeks to honor Romans, both literally, with pre-existing monuments conveniently relabelled, and with newly erected statuary; Magna Graecians may have used the formula for Roman portraits dedicated in their own cities or even in Rome. Gradually, Late Republican officials and private citizens adopted portrait types in various stages of dress or undress (more on this point below). Augustus at first followed suit, then preferred to stress his civic and religious role, although under his rule the nude image was not completely eschewed and became increasingly popular under later emperors.5 Yet even in cases of total nudity and with the addition of divine attributes, the portrait was not meant to depict an emperor deified after death, as generally assumed. H. convincingly argues that acknowledged images of divi show them wearing the toga instead (pp. 225-229).

The use of divine attributes requires a more complex interpretation. H. (p. 238) makes a novel distinction between images of mortals carrying objects that refer to specific qualities shared by both the deity and the human, and statues of divinities that have been given the idealized features of an individual but are truly representing the gods. The lines were blurred even in antiquity (pp. 239-240) but inscriptions and context would have clarified the issue, whereas the modern viewer remains confused. This section (pp. 240-242) includes a few comments on women’s images “in the guise of” goddesses, and it is noted that such renderings are not limited to members of the Imperial family; but the female nude portrait is not treated in equal depth and only 16 items appear within the Catalogue. It is, however, acknowledged that the formula is a Roman invention that could not rely on Greek antecedents (p. 219).6

The term “nudity” as used by H.—and seemingly also by the Romans—does not always imply total absence of clothing. Figures in which the entire body is exposed may carry a chlamys wrapped around one arm and bunched over the shoulder, in a popular formula of clear Greek derivation.7 Other forms of “naked costumes” are the hip-mantle that leaves the full torso uncovered; the large cloak or “Jupiter Costume” reserved for emperors (p. 259) and worn similarly or draped over the legs of an enthroned figure—the latter a Roman innovation—to express that the ruler was “like Jupiter” and his ruling viceroy on earth (p. 258); and the mantle fastened with a fibula on the right shoulder. This costume classification is extended to busts and numismatic images where traces of clothing can be distinguished.

H. correctly points out that the hip-mantle fashion is not a Roman novelty, as occasionally implied, but has Greek antecedents, which he exemplifies through fourth-century vase painting (p. 123 and n. 16) and treats as a single item. I would make a distinction, however, between the (shorter?) hip-mantle that crosses the body around the hips to cascade over the left arm, and the (longer?) one that covers the back and rests on one shoulder. Both these renderings occur in original sculptures at least as early as the late fifth century: the first on a relief figure from the Nemesis Base at Rhamnous, the second on a statue in the round dedicated by Lysikleides.8 I suspect that only the first version signifies a young person, whereas the second, as H. notes, is often used for mature males, like the Poseidon of Melos (his Pl. 73). Perhaps the same age distinction obtained in Roman times.

The fashion mentioned above as popular recurs on the earliest proven Roman Republican statue (ca. 110 B.C.), although its head is missing: an over-lifesized marble from Delos inscribed to C. Ofellius Ferus (H.’s B1 and Pl. 51, pp. 103-104). The man thus honored was a negotiator, a businessman; yet his statue carried a sword in the parade position and his mantle is of the fringed military type. H. acknowledges this point (pp. 132, 158) but considers the situation “in many respects exceptional” (p. 153) without further elaboration. A suggestion, that Ofellius may have been an early governor of the province of Asia, has been rejected because the title is not included in the dedication, but more recently Filippo Coarelli has revisited his theory that the Agora of the Italians, where the monument was set up, served as a slave market, arguing that Ofellius may have been instrumental in restoring justice and nomos after the slave revolt around 130 B.C., hence deserving the military overtones.9

Other statues probably as early are truly naked, without garments of any sort, but may not be portraying Romans. The most famous among them is the bronze “Hellenistic Ruler” in the Terme (H.’s Pl. 30) whose identity has been endlessly debated. H concedes that he may be a Roman (pp. 46-47) but he leaves him off his Catalogue and seems inclined toward the more traditional label (p. 297). I can add a recent, extensive discussion that favors the Roman Consul Manlius Vulso in connection with his 187 B.C. triumph.10 At any rate, other examples can be adduced, and this “fully naked” type is the most persistent, since it recurs as late as the well-dated “Rulers from the Yemen” (between A.D. 290 and 319; pls. 144-145) and was perhaps used even for Constantine.11

I basically agree with H.’s chronological outline of interactions between Greeks and Romans, but I would nuance it somewhat. H. correctly points out (p. 30) that East Greek gravestones rarely depict naked athletes; but even in the Archaic period the eastern regions showed a preference for draped male figures and “kouroi.”12 This tendency toward modesty continues with the above-mentioned stelai, where men usually wear a tunic under the himation. Honorary marble statues of Romans, both Republican and early Imperial, from Kos and Rhodes use the same draped formulas, occasionally with the single himation and the breast partially exposed. So does a recently found bronze, if correctly identified as L. Cornelius Lentulus Spinther, ca. 56-53 B.C.13 Even those sources that mention statues as “nude” often mean only partial nudity, and in some cases a pun might be intended—as for the images of Verres’ son, whose condition is metaphorically paralleled by that of the Sicilian province stripped bare by the father, obviously neither to be taken literally. And Latin writers could occasionally misrepresent the evidence, for their own moralistic purposes.14

The athletic physique of Hellenistic rulers may, I suspect, have been partly influenced by Egyptian beliefs that connected the fertility and welfare of the country to the pharaoh’s virility, as tested by his periodic running of the Heb-sed festival. The other strand of influence briefly acknowledged by H. (p. 94) comes to the Romans from Etruscan tomb sculptures; yet I would stress that their partial nudity is usually expressed by a decidedly plump body that has given rise to the sobriquet “fat Etruscan.” Far from being a realistic trait, this rendering was meant to express wealth and leisure, and was thus as “unreal” a body as that of the Greek “hero,” yet was obviously deemed an unsuitable formula to convey Roman frugal ideals. Some Etruscan “portraits,” on the other hand, have now been shown to recur on statues of unrelated individuals. The result would therefore be comparable to those Roman images that H. would define as the combination of an idealized (youthful) body and an idealized (mature) head. The former was derived from the Greeks as “a wholly artificial one, far removed from the circumstances of real life,” (p. 222); the latter, although usually a recognizable likeness, was a comparable manipulation of reality “according to the Roman ideals of age and character” (p. 295). The apparent contradiction for the modern viewer was therefore nonexistent for Roman eyes.

That Greek sculptors of the Late Hellenistic period could adopt similar mixtures of age renderings in their mythological subjects had not even registered on me until it was pointed out that the body of the famous Laokoon has a virile appearance at odds with his elderly face.15 Even Laokoon’s heavily veined hands and feet look considerably older than his “heroic” torso; yet the total effect goes unchallenged because of its dramatic content. Other comparable examples could be cited among the purely narrative sculptures, and their existence convinces me that H. must be right in his interpretation. His is an intelligent book that must be taken seriously and can be read with profit beyond its specific subject.

Finally, a brief review of the Appendices. A summarizes previous studies on the subject of nudity in art and portraiture. B is the Catalogue, in two chronological groupings: Republican or Early Augustan (nos. 1-26), and Imperial (nos. 27-342). The latter has a section on Antinoos (nos. 180-196), lists of non-Imperial and uncertain subjects (nos. 197-326) and a coda on women, none certainly imperial (nos. 327-342). Appendix C discusses the cuirassed portraits from the Temple of Mithradates Eupator on Delos. D analyzes the terminology for the cloak worn by portraits in heroic costume, and includes comments on colors already touched upon on p. 136 in connection with Greek evidence.16 Appendix E reviews scale armor in Roman art and literature. F amplifies entries B142 and B310: two portraits in heroic costume identified as Pompeius Magnus. G treats the larger cloak worn by some Imperial hip-mantle portraits. H explores the Imperial use of the aegis in cameos and major portrait statuary. I, “Flavian (and later) Cuirassed Portraits added to Pre-existing Julio-Claudian Portrait Groups,” could include the evidence from the Julian Basilica at Corinth (now perhaps dated early Tiberian).17 J deals with “The Return of Military Imagery to Imperial Iconography after its Renunciation by Augustus.” K mentions attested or suspected galleries of multiple portraits depicting a single individual.18 L returns to color in discussing the toga picta of Roman magistrates. The last Appendix, M, focusses on the portrait venatorio habitu from the Tomb of Claudia Semne. All in all, a very rich fare that deserves a wide readership.


1. See, e.g., his recent review essay, “Emulation versus replication: redefining Roman copying,” JRA 18 (2005) 419-435.

2. Some examples: p. 104 n. 3, for Pl. 61 read Pl. 72; p. 262, for Pl. 121 read 119; p. 268, for Pl. 46 read 47; p. 307, for Pls. 94, 98 read 93, 97, and for Pls.132-3 read 130-1. The caption to Pl. 11 mentions that Kleoboulos, on his gravestone, holds a rabbit, but p. 27 is chronologically correct in calling the animal a hare (this ambivalence recurs elsewhere: rabbit on p. 29 n. 7, p. 41; hare on pp. 31 and 54 n. 39). The caption to Pl. 44 (the small bronze from Herculaneum often considered Demetrios Poliorketes) describes the Hellenistic ruler as having bull’s horns, but the text (p. 44) more neutrally calls him horned; his are in fact goat’s (or Pan’s) horns, as extensively argued by P. Laubscher ( AM 100, 1985, 333-353) and by A. Rumpf before him ( AM 78, 1964, 176-199). Typos are largely irrelevant; note, however, that British, rather than American, spelling and conventions are employed throughout.

3. After all this considerable help, it seems invidious to demand more, but, since figures and plates are scattered throughout the book, it would have been useful to have page references added to their lists and to the individual catalogue entries, since not all items in it could possibly have been illustrated. Perhaps a future (paperback?) edition could add this information, as well as, in the catalogue, page references for text mentions, especially since the text proleptically gives each work its B-number (cf. p. 103 n. 1) and a computer search could easily retrieve them. One last quibble: since some appendices occupy a single page, they carry no page number, making them hard to cite. This is a design layout that works well for the initial page of chapters but not for addenda, and could be remedied in the future.

4. On the nakedness of slaves see now G. Pucci, ” Detrahis vestimenta venalibus. Iconografia della vendita di schiavi nell’antichità e oltre,” JRA 18 (2005) 235-240, esp. 235-237.

5. Augustus’ renunciation of fully nude portraits is mentioned on p. 160, although somewhat nuanced in the following pages. That this renunciation has probably been overstated is suggested by T. Stevenson in dealing partly with similar issues: “The ‘Problem’ with Nude Honorific Statuary and Portraits in Late Republican and Augustan Rome,” Greece and Rome 45 (1998) 45-69, esp. p. 61.

6. This shortcoming is acknowledged in the Introduction (p. 4) and partly justified in light of the many studies of the subject in recent publications. The topic is also touched upon in Ch. 5 as part of the chronological review: p. 199 with n. 52 and Pls. 121-122; pp. 219-222. The same brevity characterizes the discussion of sarcophagi (pp. 216-217) and of cameos, which are occasionally used as evidence but admittedly “consistently employ a richer, more exalted iconography than do the extant portraits in the round” (p. 170).

7. H. himself clarifies this point (p. 216): “…while Roman portraits are often fully naked ( in the sense of having the genitalia exposed [my italics]), their nakedness is almost always qualified by the addition of items of clothing or other attributes.” Cf. also pp. 172-176. One Greek example of the shoulder chlamys from Delphi is erroneously drawn in H., fig. 2 on p. 25, (extreme left on the Daochos Monument), although correctly described on p. 48. Note, moreover, that the figure can no longer be identified as Sisyphos II (H., p. 26), since it is in Parian rather than Pentelic marble, like the adjacent statues; technical and iconographic problems also exist: O. Palagia and N. Herz, “Investigation of Marbles at Delphi,” ASMOSIA 5 (1998) [2002] 240-249, esp. 245-247. Cf. also A. Jacquemin and D. Laroche, “Le Monument de Daochos ou le Trésor des Thessaliens,” BCH 125 (2001) 305-332, esp. 325-327.

8. Rhamnous base: see, e.g., H. Kyrieleis, ed., Archaische und Klassische griechische Plastik vol. 2 (Mainz 1986) pls. 114.1 (extreme left) and 115.1; also fig. 6 on p. 90 (B. Petrakos); I believe the same fashion recurs on one of the fragmentary bronzes from the Porticello wreck. Lysikleides statue, also from Rhamnous: Athens, NM 199, N. Kaltsas, Sculpture in the National Archaeological Museum, Athens (Los Angeles, 2002) no. 223.

9. Ofellius as governor of Asia: theory expressed orally by R. Étienne, cited (and refuted) by F. Queyrel in MonPiot 82 (2003) 67 n. 238. F. Coarelli, “L’Agora des Italiens: lo statarion di Delo?” JRA 18 (2005) 196-212, esp. 210.

10. For H.’s open-ended position on the Terme Ruler see pp. 145-148, esp. 146, and p. 297 for a Hellenistic label. Manlius Vulso: M. Papini, “Il Principe delle Terme: ‘Grieche oder Römer?’,” Bullettino della Commissione archeologica comunale di Roma 103 (2002) 9-42. I tend toward a Roman identification: Hellenistic Sculpture II (Madison 2000) 305-309, with ref. to possible Etruscan influence on the technique of the beard; cf. also n. 25 on p. 329 for additional bibliography on the “Levy Ruler” discussed by H., Pl. 84, pp. 145-146.

11. Nude statue of Constantine, on porphyry column, in Constantinople: H., pp. 265-268 and Pl. 143. See now S. Bassett, The Urban Image of Late Antique Constantinople (Cambridge 2004), under “Forum of Constantine,” Cat. 109 άβ, pp. 192-199 (literary sources), 199-200 (graphic evidence), and 200-204 (column and bronze statue). Note also her Cat. 48, p. 166, for a statue in the Baths of Zeuxippos believed to be of Julius Caesar with aegis and thunderbolt, and thus probably a nude bronze—whatever its true date, another example of a naked image in the Christian city. Also in the third cent. A.D. the LIMC (vol. 4, s.v. Herakles, no. 372, Pl. 470) correctly places the bronze Hercules in the Conservatori that H. (p. 91, Pl. 42) tentatively accepts as the Hercules Aemilianus dedicated in 142 B.C. On the colossal Constantine in Rome (B102, p. 265) see now P. Pensabene, L. Lazzarini, B. Turi, “New archaeometric investigations on the fragments of the colossal statue of Constantine in the Palazzo dei Conservatori,” ASMOSIA 5 (1998) [2002] 250-255, documenting its reworking from a single older work in Parian marble, with replacement pieces in Carrara marble.

12. They do show, however, naked “eroes,” as H. demonstrates. But R. R. R. Smith, Hellenistic Sculpture: a handbook (London 1991) 190, states that, by the Hellenistic period, “hero means little more than ‘late beloved.'” On the new approach to the cult of the dead, cf. Ridgway 2000 (supra, n. 10) 192-193 and n. 13, with refs., and 202-204 on round altars on graves.

13. Additional bibliography: A. Oliver, “Honors to Romans: Bronze Portraits,” in C. C. Mattusch, ed., The Fire of Hephaistos (Cambridge, Mass. 1996) 138-160. C. B. Rose, “The Imperial Image in the Eastern Mediterranean,” in S. E. Alcock, ed., The Early Roman Empire in the East (Oxford 1997) 108-120. H. cites the possibility of Republican palliati but considers it “unlikely” (p. 139). See, however, R. Kabus-Preisshofen, Die hellenistische Plastik der Insel Kos ( AM Beiheft 14, Berlin 1989), and, for the bronze now in Adana, C. Brüns-Özgan and R. Özgan, “Eine bronzene Bildnisstatue aus Kilikien,” Antike Plastik 23 (1994) 81-89, Pls. 37-43. Cf. also Ridgway, Hellenistic Sculpture III (Madison 2002) 121-124. H. also discounts the possibility of emperors portrayed wearing the himation (p. 209 and n. 65), but there is a strong probability that a headless bronze from Bubon belonged to Marcus Aurelius: BClevMus 74.3 (March 1987); the entire issue is devoted to the statue, but see esp. A. Kozloff, 85-99 with a forceful statement on p. 97.

14. Cicero, Verr. 2.2.154, cited by H., p. 97 and n. 62. To be sure, nakedness in this case could have signified young age, but that Romans could pun on the words is also shown by Ovid as noted by H. 290-291. For the range of meaning of the Latin nudus, see H. 61-62. Ancient sources contradicted by archaeological evidence (e.g., Suetonius): H. 250-251. I also wonder whether Augustus, by throwing his toga off his shoulders, could truly have bared his breast when begging the Romans not to make him a dictator: Suet. Aug. 52, cited by H., 100 and n. 66. H. mentions (pp. 292-293, with sources) that candidates canvassing for votes used to wear the toga without the tunic beneath it, but this could hardly have been Augustus’s attire when he was actually rejecting office.

15. G. Gebauer, “Le corps du Laocoon,” in E. Décultot et al., Le Laocoon, histoire et réception (Paris 2003) 237-250, esp. 243-246.

16. I am, however, skeptical about the fidelity of renderings in Campanian wall paintings and would have wished for a more extensive discussion of the symbolism of the star-spangled mantle on the marble Trajan from Samos, which H. can acknowledge only in n. 38, because of recent publication.

17. B. S. Ridgway, “Sculpture from Corinth,” Hesperia 50 (1981) 435 and Pl.93.d for a cuirassed torso of probable Late Hadrianic/Early Antonine date; see also pp. 432-433 for comments on the likely different dates of the Julio-Claudian portraits. H.’s Pl. 98 (p. 173) is misleading because the statue of Augustus was found near the southwest corner, while those of Lucius and Gaius came from the east aisle and cryptoporticus together with a Nero and a Hadrian. I owe this and the chronological information to the kindness of N. Bookidis. To be sure, later destruction might have been responsible for the scatter.

18. This is in addition to the main text, pp. 208-212. I would accept that the three statues in the Villa Doria Pamphili (Pls. 125-127; B231, B239) portray the same man, but why is the military image (with sword) supported by a tree trunk ending in a boar’s head? The original publication (by H. von Heintze, Antike Plastik 1, 1962, 13-15) does not mention it among the restorations.