Riace bronzes

History

Bronze statuary in the 5th century BC

The Riace bronzes are generally regarded as Greek originals of large-scale 5th-century BC bronze statuary rather than Roman-period copies. By the 5th century BC bronze had become the preferred material for full-length Greek statuary. Between the end of the Archaic period and the Classical period, hollow lost-wax casting gradually replaced the construction of large figures from assembled sheets of hammered metal and made it possible to cast life-size statues in separate parts that were subsequently joined. Very little of this production survives: with few exceptions, ancient bronze statues were lost or melted down for their metal, and the appearance of many famous Greek originals is known mainly through Roman marble copies.

A tradition of bronze sculpture also existed in Southern Italy and Sicily, from which few originals survive; schools of bronze workers were active at Rhegion and Crotone from the late 6th century BC and especially during the first half of the 5th century BC. Klearchos and Pythagoras of Rhegion, to whom the tradition assigns a role in the formal development of the Severe style, worked at Rhegion, while the sculptor Patrokles and the bronze worker Dameas worked at Croton; Dameas was active in the second half of the 6th century BC and made the statue of the athlete Milo of Croton described by Pausanias at Olympia. No surviving original can be securely attributed to Pythagoras. The presence of these bronze-working schools is not sufficient to establish where the Riace statues were made, and their provenance remains uncertain. The two statues also differ stylistically: A has features of the Severe style, while B shows elements associated with the tradition of Polykleitos. A 5th-century BC date is also compatible with radiocarbon dating of the organic components of the casting cores, although the tests do not allow a narrower chronology.

Discovery

On 16 August 1972 Stefano Mariottini, while freediving in the waters off Riace Marina, saw part of a dark-coloured shoulder projecting from the sand; after touching it he recognized the material as bronze and began clearing the sand from a human figure lying on its right side. During his dives he noticed, about one metre from the first statue, a big toe and a knee barely emerging from the sand, and thus located a second statue lying on its back beneath a thin layer of sediment. The two statues lay a short distance apart near the submerged reef of Porto Forticchio, in the Agranci district, about 220 metres from the shoreline. In his account published in 2009, Mariottini recalled a seabed about 6 metres deep, whereas the report filed on 17 August 1972 recorded a depth of about 10 metres. The distances and topographical references in the report of the underwater unit of the Messina Carabinieri Legion are, however, approximate. No photographs or films are known of the statues in their position on the seabed before recovery.

Mariottini left the line of a marker buoy tied to the arm of the first statue and, with the help of relatives and friends, managed to locate superintendent Giuseppe Foti at his home in Reggio Calabria. Returning by boat to Riace on the afternoon of 16 August, he took bearings on land and completely covered the two statues with sand to protect them. On 17 August he went to the museum in Reggio Calabria, where he delivered a written report to Foti; Foti arranged the recovery and notified the Carabinieri divers group in Messina. The group was then in Syracuse assisting an attempt by Enzo Maiorca to set a freediving record and could not intervene at Riace until 21 August. Foti also instructed Enrico Natoli to carry out an underwater survey of the site and entrusted a letter to be delivered to the Guardia di Finanza so that the area could be watched. Recovery operations began on 21 August. Statue B was recovered on 21 August and Statue A on 22 August. During the recovery, Statue A fell back to the seabed before being brought ashore. Photographs and film footage of the operations document the use of underwater lifting bags to raise the statues. On the beach the statues were laid on improvised wooden and foam-rubber beds and then transported on a Carabinieri van.

A film reconstruction of the recovery was made at Briatico using two fiberglass copies. It was produced for the RAI series Nel mare degli antichi in 1986 and directed by Pippo Cappellano.

The official report filed on 17 August 1972 as protocol no. 2232 with the Superintendency of Antiquities of Calabria in Reggio states:

Description

The statues depict two nude, bearded men standing slightly over life size. Excluding the tenons—the projections beneath the feet used to secure them to their bases—Statue A is about 1.98 m high and Statue B about 1.97 m. Photogrammetric surveys show nearly identical poses: in Statue B the shoulders and torso tilt slightly to the right, while Statue A has a more pronounced turn of the head and wider separation of the feet. In both statues both feet touch the ground: the right foot, corresponding to the weight-bearing leg, is slightly set back, while the left leg is slightly bent at the knee. A appears tenser and more dynamic, whereas B has a more relaxed pose and a stronger bend at the right hip. The musculature of Statue A is more strongly marked; in B it is less pronounced and the torso forms a more evident S-curve.

Statue A

Statue A, inventory no. 12801, has a head cast in a single piece, with the hair modelled even in areas that were covered by the helmet. The mouth is half open; in addition to the copper lips found on both statues, Statue A retains teeth made from silver sheet worked in relief without heating, by cold repoussé, and attached to a structure inserted into the clay core. The outer locks of the hair and beard are three-dimensional, cast separately and attached with dovetail joints.

Statue B

Statue B, inventory no. 12802, has a head cast in two parts: the cranial cap was cast separately and joined by poured metal; it has no hair and an irregular surface, with gaps that in places pass through the bronze. The beard and hair were modelled directly in the wax layer used for casting, with hair limited to the frontal locks and a few tufts on the neck; the mouth is closed and no teeth are visible.

Casting technique and materials

The right arm and the left forearm of Statue B, as far as the porpax—the loop that supported the shield on the forearm—are ancient replacements rather than original parts. Chemical analysis found a lead content of 9–14 per cent in these parts, compared with about 1 per cent in the rest of Statue B and in Statue A; the structure of their casting cores also differs from that of the rest of Statue B. The replacement arms were made later and in a different place from the rest of the statue, but it cannot be established whether the substitution repaired damaged parts or deliberately altered the figure's appearance.

Between 1993 and 1996 the Istituto Centrale per il Restauro (ICR) directly examined the interiors of the statues and reconstructed several stages of their manufacture. The statues were cast by the direct lost-wax method, in which wax is modelled over the inner core; exceptions are the front portions of both feet and the replacement arms of Statue B, which were made by the indirect method, in which the wax layer is produced from a mould taken from a model. In both statues the torso and legs, rear parts of the feet, head, arms, hands, testicles, penis, forefeet and third toe of each foot were cast separately; the cranial cap of Statue B was also separately cast.

The separately cast parts were then joined by cast-on welding: the edges were brought together with a gap between them, closed below and at the sides with refractory clay capable of withstanding high temperatures, into which liquid bronze was poured; the excess metal was then removed and the exterior surface smoothed. By the 5th century BC the technique could produce strong joins even at highly stressed points such as the middle of the foot; to increase the contact surface, a series of oval cavities could be prepared and filled one by one in successive pours. In large bronze nudes, joints followed recurring patterns at the neck, the attachment of the arms to the torso, the genital region, the middle of the foot and the attachment of the central toes.

The casting cores of both statues, except for the replacement arms of Statue B, consist of concentric layers of clay arranged around iron supports. In Statue A the clay was made less fluid by adding animal hair, chiefly sheep's wool; in Statue B vegetable material was used instead, mainly leaves, straw and small flakes of wood. The cores were modelled differently: in Statue A the core does not reproduce the anatomical volumes, which were shaped primarily in wax; in Statue B, especially on the front of the legs, the outer layers of the core were trimmed into a rough but recognizable anatomical form. The same difference appears in the torso: in Statue A the inner core is relatively flat and little articulated, and the muscles and rib cage were formed mainly in the wax layer, whereas in Statue B the core more clearly reproduces the groin, abdomen and ribs. Measurements published in 1999 give an average bronze wall thickness of about 12 mm for Statue A and about 9 mm for Statue B; the difference is linked to the different methods of modelling the core and wax.

Original attributes

Surviving traces indicate that both figures carried a long shaft in the right hand and a shield on the left arm; helmets or other headgear, reconstructed in different forms, were fitted to their heads. The hair of Statue A is modelled even in concealed areas and, together with the surviving attachment devices, indicates that a helmet was placed on the head. Rebaudo considers the deformation of the occipital area of Statue B compatible either with the profile of a raised helmet or with bulky headgear; on the forehead are two copper plates with a hammered finish, which Claudio Sabbione proposed in 1984 to interpret as a leather or textile cap worn beneath the helmet.

In 2012 the right hand of Statue A was recorded with a structured-light 3D scanner to reconstruct in three dimensions the cavity left by the object originally held in the hand. The three-dimensional model allowed the cavity to be measured in fourteen cross-sections; a cylinder about 30 mm in diameter can pass through all of them without leaving the cavity. The cavity might also have contained a curved or non-cylindrical object; the reconstruction judged most likely is nevertheless a spear, with a shaft about 30 mm in diameter and about 2.20 m long, while the hypothesis of a horse rein is less compatible with the shape of the hand.

Restoration and conservation

1972–1980 restoration

After recovery, the statues underwent an initial restoration at the Museo Nazionale della Magna Grecia in Reggio Calabria, which continued until 1975. In January 1975 they were transferred to the restoration centre of the Archaeological Superintendency of Tuscany in Florence and entrusted to the restorers Renzo Giachetti and Edilberto Formigli.

During the Florence restoration, residual ancient casting core material was only partly removed by mechanical and chemical methods. The interiors were emptied through openings in the heads and soles of the feet, using steel rods and tubes, hydrogen peroxide baths and jets of compressed water; sections of the internal iron rods and much of the core material in the torsos were removed, but residues remained in the legs and partly in the torsos.

1992–1995 restoration

In 1984 the ICR began chemical and physical analyses, together with monitoring of the environmental conditions around the statues to assess their state of preservation. Radiography and an endoscopic examination in 1985 showed severe corrosion where the remaining casting core material was in contact with the bronze walls. On 6 February 1986 a complete examination of the interior with a video endoscope established that casting core material was still present in the legs and, in part, in the torsos. By the end of the subsequent treatment, 56 kg of casting core material had been removed from Statue B and 72 kg from Statue A.

During the 1992–1995 restoration, material remaining inside the statues was removed layer by layer with articulated instruments and microcameras; this "micro-excavation" revealed the construction of the inner cores and allowed much of their form and working sequence to be reconstructed in three dimensions. From the 1970s through the 1990s several studies held that the indirect method, in which wax is obtained from a mould made from a model, was already widely used in the 5th century BC, but physical evidence from workshops producing large statues was lacking. During the first emptying operation, the belief that the casting core material had been poured in liquid or semi-liquid form led to its layered arrangement not being recorded.

The later micro-excavation revealed cores built up in concentric or parallel layers; Micheli and Vidale proposed that both statues had been modelled by the direct method, with the core freely built around iron pins, rather than by filling a hollow wax mould as in the indirect method. In 1999 Micheli and Vidale stressed that the direct-method interpretation remained a hypothesis to be tested on other statues of the same period. The operations, carried out with Giuseppe Sgrò and Irene Spuri, lasted about two years and produced about 80 cross-sections at average intervals of 5 cm and about 80 hours of video recording. To counter corrosion, the internal surfaces were treated with slender tools inserted into the cavities and controlled by microcameras. The second restoration ended in July 1995. At its conclusion, the previous system attaching the statues to their bases was replaced by a new support and anti-seismic base system that had undergone simulation tests.

2009–2013 intervention

The 2009–2013 conservation intervention was carried out at Palazzo Campanella, seat of the Regional Council of Calabria, with the participation of restorers from the ICR and the Archaeological Superintendency of Calabria. In August 2010 the laboratory installed in the Salone Monteleone of Palazzo Campanella was directed by Pasquale Dapoto of the Superintendency and placed under the responsibility of Roberto Ciabattoni and Paola Donati of the Istituto Superiore per la Conservazione e il Restauro. The schedule announced in August 2010 envisaged the statues' return to the museum by March 2011. Work instead continued until 2013; the statues returned to the museum on the night of 21 December.

The analyses included chemical examinations and measurements of surface colour; gamma radiographs, images of the interior obtained with gamma rays, were also made and later repeated in selected areas. In 2010 an electronic system was installed to monitor remotely temperature, humidity and light inside and outside the statues and to trigger an automatic alarm if anomalous values were detected. Shaped supports, or counterforms, made of carbon fiber and Kevlar were produced for supporting and moving the statues; their shapes were derived from full-size resin prototypes produced by three-dimensional laser scanning.

In the 2013 reinstallation the statues were placed on new anti-seismic bases of Carrara marble designed by ENEA. Anti-seismic bases of this type were used for the first time with the Riace bronzes. The base was designed to permit large horizontal displacements with little energy dissipation and to reduce the effects of both horizontal and vertical seismic motion. The design called for materials compatible with the display and simple maintenance. The device consists of two superimposed load-bearing blocks: balls between the blocks allow the base to slide horizontally; in the upper block pairs of cylinders and pistons reduce vertical oscillations, while damping systems dissipate oscillation energy. Before installation, ENEA subjected the bases to experimental tests.

Conservation and documentation

The room housing the Riace bronzes was equipped with an air-control and filtration system to protect the statues from environmental fluctuations. When the room reopened in December 2013, access was organized in groups of twenty people, with about twenty minutes in the pre-filter room, about three minutes in the filter room and about twenty minutes to view the statues. In March 2023 the museum and the ICR agreed on a periodic monitoring programme for the Riace bronzes and the Porticello bronzes, begun in September 2023. In 2024 checks took place on 15–16 April, 27–30 May, 17–18 June, 1–2 July and 16–18 September; they included direct examination of the statues, instrumental tests, and graphic, photographic and very-high-resolution three-dimensional recording. On 7 January 2026 the National Archaeological Museum, Florence and the Museo Nazionale della Magna Grecia signed an agreement to catalogue and digitize the documentation held in Florence on the restoration and exhibition of the Riace bronzes and to make it available to both museums.

Exhibitions and public initiatives

On 15 December 1980, after the Florence restoration, the two statues were exhibited to the public for the first time at the National Archaeological Museum, Florence. The display, initially planned as temporary, was extended until 24 June 1981; it was followed by an exhibition at the Quirinal Palace, which ended on 12 July 1981. The Florence exhibition received 250,000 visitors in its first month, while the Quirinal exhibition attracted 300,000 visitors in twelve days. During the intervention at Palazzo Campanella, the restoration laboratory was open to the public and received almost 80,000 visitors between 1 January and 31 July 2010. From 4 May to 4 September 2016 the Liebieghaus Skulpturensammlung in Frankfurt hosted the exhibition Athens: The Triumph of Imagery, in which the reconstruction of Statue A made in 2015 was presented for the first time together with a reconstructive cast of Statue B.

From 19 June 2019 to 27 September 2020 the Altes Museum in Berlin hosted the exhibition Starke Typen. Griechische Porträts der Antike; the reconstructive casts of Statues A and B from the Liebieghaus were placed in the museum's rotunda and remained on display until 12 January 2020. From 5 July 2022 to 26 March 2023 the Metropolitan Museum of Art in New York hosted Chroma: Ancient Sculpture in Color, organized in collaboration with the Liebieghaus Skulpturensammlung; both reconstructions of the Riace bronzes, made by Vinzenz Brinkmann and Ulrike Koch-Brinkmann in 2015–2016, were exhibited.

On 12 July 2022, for the fiftieth anniversary of the discovery, the Directorate-General of Museums, the Colosseum Archaeological Park and the Museo Nazionale della Magna Grecia organized a meeting in the Curia Julia in collaboration with the Metropolitan Museum of Art. A bronze reproduction of the head of Statue A, made by Vinzenz Brinkmann to illustrate a reconstruction of its original colour appearance, was shown during the meeting; a video produced by the Metropolitan Museum for an exhibition on the polychromy of ancient sculpture that had opened in New York on 5 July 2022 was shown with it. The reconstructed head and video remained on display in the Curia Julia until 18 July 2022. From 10 to 12 November 2022 the Museo Nazionale della Magna Grecia hosted an international conference for the fiftieth anniversary, with more than thirty scholars and researchers. The programme had four sections devoted to underwater archaeology, ancient art history, restoration, and presentation of the bronzes to the public; the first section also presented results from campaigns conducted in the discovery area through 29 October 2022.

Identification, dating and attribution

Hypotheses on dating, provenance and authorship

Since the late 1970s, different proposals have been advanced concerning the dating and provenance of the bronzes, their sculptors, the identity of the figures, the original monument, and their history before the discovery. A 5th-century BC date is predominant, but there is no agreement on the identity of the figures, their sculptors, or the monument for which they were made. Style is one of the criteria used for dating them: Statue A has the characteristics of the Severe style, while Statue B shows elements connected with the tradition of Polykleitos; A is generally placed around 460 BC and B around 430 BC. The resemblance of Statue B to works in the Polykleitan tradition does not imply an attribution to Polykleitos: Phidias, Myron, Alcamenes, Onatas, and Pythagoras of Rhegion, among others, have been proposed for one or both statues; the attributions include sculptors and schools from Attica, Aegina, the Peloponnese, and Magna Graecia.

Because most of the original attributes are lost, few elements remain for identifying the figures: the helmet and shield characterize at least Statue A as a warrior, but proposed identifications include mythological heroes, Athenian eponymous heroes, generals, victors in athletic contests, athletes, and deities. Delphi, Olympia, Athens, centres of Magna Graecia, and, in some hypotheses, Sicily have been proposed as their original location. The form and number of the surviving tenons beneath the feet, two on Statue A and four on Statue B, have been used to propose different dates and restorations, but the interpretations do not agree. Werner Fuchs, Antonio Giuliano and Maurizio Harari connected the bronzes with the Athenian dedication, the votive monument erected along the Sacred Way at Delphi after the victory at Marathon and attributed by the ancient sources to Phidias; Claude Rolley, however, observed that the statues reconstructed from bases associated with that dedication would have been smaller than the Riace bronzes. Analyses of the casting earth and other materials left inside the statues have been used to investigate their provenance, but by themselves cannot identify the sculptors or the original monument. Underwater investigations in the discovery area have produced few finds and have not securely identified a ship's cargo; explanations proposed for the statues' presence on the seabed include loss in a shipwreck and jettisoning to lighten a vessel during a storm, while the route, date of the event, and the bronzes' earlier history remain unknown.

Hypotheses from 1979 to 1992

In 1979, and again in 1981 and 1982, Werner Fuchs connected the two statues with the dedication erected by the Athenians at Delphi after the victory at the Battle of Marathon and attributed them to Phidias. Alberto Busignani adopted the same interpretation; Antonio Giuliano developed the attribution to Phidias and his workshop, while Maurizio Harari revived the connection with the Delphic dedication in 1988. In 1982 Carlo Odo Pavese proposed that the attribute held in the right hand should be a laurel or olive branch. Antonino Di Vita developed this suggestion together with evidence concerning the helmets and comparisons with Attic pottery, interpreting both statues in 1984 as victors in the armed race at Panhellenic or Panathenaic contests and at the same time as politically prominent Athenians. In 1983 Hans Peter Isler and in 1984 Paolo Enrico Arias identified the figures as Athenian heroes, connected Statue A with Phidias or his circle and Statue B with the Phidian school, and dated the works respectively to about 460–450 BC and roughly thirty years later. Arias also regarded both statues as works of an Attic milieu.

In 1983 Claude Rolley proposed identifying Statue A as an Attic eponymous hero, one of the heroes who gave their names to the Athenian tribes, sculpted by an Athenian artist around 460 BC; Statue B would be a second Attic eponymous hero, attributable to the circle of Phidias and datable to about 430 BC. He repeated the proposal in 1984. Rolley reconstructed A's head without a helmet: the cut-down and peened pin might have supported protection against bird droppings, while the band at temple level would have been a diadem, understood as a sign of the hero's royalty. Castrizio disputes both interpretations: he considers it more likely that the pin intended to support a Corinthian helmet broke and was replaced by the stronger one preserved at the centre of the head; regarding the band, he notes that surface marks indicate almost complete coverage by the helmet and that the two hanging ends typical of diadems are absent, and therefore interprets it as padding placed between the head and the helmet.

Enrico Paribeni excluded an Attic origin for Statue A and turned his attention to the sculpture of Magna Graecia. He gave particular weight to the discovery site, in the Locrian area of southern Calabria, and rejected the hypothesis that the statues had been abandoned by a passing ship. He identified A as a hero, perhaps Ajax the Lesser, made by a Peloponnesian artist between 460 and 450 BC, and B as a strategos, made by an artist influenced by Attic sculpture between 410 and 400 BC. For Paribeni, A still belonged to the Severe style, whereas B belonged to mature Classicism. Castrizio objects that the identification of A with Ajax the Lesser rested mainly on the hero's role among the Ozolian Locrians of Greece, whom he described as the founders of Italian Locri; he also disputes the large chronological gap between the statues, arguing that it depended on comparison with the development of Athenian sculpture and was less suitable for explaining Peloponnesian, South Italian Greek and Siceliot works. Paribeni was more cautious about Statue B and proposed no precise identification, although the comparisons he selected could suggest a sword in the right hand; Castrizio, by contrast, regards the presence of lead in the cavity of the hand as evidence that the object held was a spear.

Hypotheses since 1998

In 1998 Paolo Moreno proposed identifying Statue A as Tydeus and Statue B as Amphiaraus, two protagonists of the Seven against Thebes, and connected both to a statuary group from Argos; in a later study he reaffirmed the identifications, attributing Tydeus to Ageladas II and Amphiaraus to Alcamenes I. Moreno excluded an athlete identification for Statue A because of the figure's hostile attitude; in Statue B he considered the facial expression too intense for a historical portrait, interpreting both as mythological figures belonging to the same statuary group. Moreno used technical data from the restoration to argue that the two statues belonged to the same group; from the casting earth he derived a hypothesis of production at Argos for A and, more cautiously, at Argos or in Attica for B.

Statue A would represent Tydeus, a hero of the Theban Cycle, associated with Calydon, father of Diomedes and protected by Athena. An iconographic comparison used for this identification is the high relief from Temple A at Pyrgi, dated to 470–460 BC, in which Tydeus bites Melanippus's skull with his teeth clearly visible, as in Statue A.

Statue B would represent Amphiaraus, king of Argos and a seer, who foresaw the disastrous outcome of the expedition of the Seven against Thebes and knew that he would die there.

Tydeus and Amphiaraus took part in the expedition of the Seven against Thebes, which ended in defeat for the Argive army. In Moreno's proposal the two statues are dated respectively to about 450 BC and 440 BC.

Identification of the artists

Moreno attributed Statue A to Ageladas the Younger, an Argive sculptor, and Statue B to Alcamenes the Elder, originally from Lemnos but professionally trained in Athens. The attribution of Statue B to Alcamenes revived a name already proposed by Georgios Dontas in 1988. Since some ancient sources call Alcamenes Athenian and others Lemnian, it has also been conjectured that he was born on Lemnos and later received Athenian citizenship as a reward for his artistic merit.

Ageladas is remembered as an Argive sculptor and as the maker, at Delphi, of a group of Messapian women and horses dedicated by the Tarentines; because the works attributed to him by the sources span the period from 520 to 450 BC, some scholars have hypothesized two sculptors of the same name. Moreno connected the two sculptors with the Temple of Zeus at Olympia, attributing to them the entire sculptural programme of the two pediments and the metopes, completed in 457 BC. In 2014 the casting earths had been interpreted as certainly Argive for Statue A and very probably Athenian for Statue B.

Summary tables

Proposals from 1979 to 1988

Proposals from 1998 to 2025