Introduction
Much speculation surrounds the origins and cultural identity of the large bronze ‘Lion’ that tops one of the two large columns in the Piazzetta by St Mark’s Basilica in Venice (Figure 1). Hypotheses on its origins include a twelfth-century AD Venetian foundry (Pilutti Namer Reference Pilutti Namer2013) or an unspecified location in Anatolia or northern Syria in the Hellenistic period (323–30 BC) (Ward Perkins Reference Ward Perkins1947; Wills Reference Wills1999), with possible Romanesque, Gothic, Assyrian, Etruscan, Sassanian and Chinese influences (Elam Reference Elam1990). B.M. Scarfì (Reference Scarfì1990) viewed the ‘Lion’ as a Hellenistic interpretation of Mesopotamian or Persian lion-headed griffins, made in the fourth or third centuries BC as the vehicle of Sandon, deity of the city of Tarsos in Türkiye, who was depicted on coins standing on a big horned cat. In this article, in contrast, we argue that the St Mark’s Lion was originally a monumental Chinese bronze, heavily modified by later adjustments. Similarities in the morphology of the snout and stylistic features of Tang Dynasty (AD 618–907) renderings of lions have already been noted, as have the scars left by the removal of horns (Scarfi Reference Scarfì1990: 72, fig. 58). Here, we reappraise this stylistic comparison, drawing on the results of lead isotope analyses to identify a Lower Yangzi origin for the metal alloys that compose the original bronzework.
Figure 1. The great bronze ‘Lion’ of St Mark’s Square, Venice (Scarfì Reference Scarfì1990). The statue is about 4m long from muzzle to tail and 2.2m high at the head.
Evolution of a super-symbol
According to Tigler (Reference Tigler2000: 3) the large columns of the Piazzetta arrived in Venice not much earlier than AD 1261, and were erected a few years later, with the bases and capitals sculpted on site. The column supporting the lion is made of a violet granitoid rock from north-western Anatolia (marmor troadensium) (Lazzarini Reference Lazzarini2010). Yet, the only historical documents mentioning St Mark’s Lion on its column are dated 14 May 1293, when the sculpture was already damaged and needed restoration (Cessi Reference Cessi1931: 339). The next precise historical information we have deals with its removal to Paris in 1797 by Napoleon, as a symbol of the defeat and humiliation of Venice, and its return in pieces from there in 1815 (see below). The final restoration of the sculpture (Catra Reference Catra2014), with new wings, then added to a complex micro-history of discontinuous, sequential castings, explained in detail in Scarfì Reference Scarfì1990 (with analytical data on each metallurgical step).
The icon of the ‘Winged Lion’ had developed from the apocalyptic tetramorphs (lion, ox, eagle and winged man) and the lion protomes, winged and nimbate (surrounded by a halo or circle of light), commonly depicted in wall mosaics between the fifth and thirteenth centuries AD. The meanings and uses of the Venetian Winged Lion, and its transformation between 1261 and 1268 from religious icon to state emblem, are a focus of recent scholarship (Aldrighetti Reference Aldrighetti2002; Pedani Reference Pedani2004: 3; Griffith Reference Griffith2005; Rizzi Reference Rizzi2012: vol. 1, 17–26, 33, 65).
In the thirteenth century, Mongolian heraldry adopted the Winged Lion (e.g. von Fircks Reference von Fircks2018: 182, fig. 13) and similar lions appear in Mamluk dinars minted by sultans Baybars (1260–1277) and Baraka (1277–1280), and in coeval military insignia (Pedani Reference Pedani2004). A late-eleventh-century panel depicting two rampant winged lions appears in the doge’s chapel in St Mark’s Basilica (Zuliani Reference Zuliani1969: 113–42).
The earliest representations of Winged Lions in Venice are in moeca—with radiating wings turned towards the head, like the legs of a crab. The oldest are three sculpted reliefs, two of which are now in the Museo Correr, that have been subject to reuse and damage since their creation, sometime between the eleventh and fourteenth centuries AD (Rizzi Reference Rizzi2012: vol. 2, cat. 190 & 191, 32; cat. 523, 55). One of the reliefs at the Museo Correr, originally from the church of St Aponal (twelfth–thirteenth centuries), may be a ‘portrait’ of the St Mark’s Lion before its installation on top of the column (after 1172, but before 1293: Piazza Reference Piazza, Pala and Usai2023; Artioli et al. Reference Artioli, Angelini, Vidale, Henderson, Morgan and Salonia2024a). The third relief, whose copy is still in situ in Rio de San Marcuola, is also reminiscent of the bronze Lion, with its curls and odd human-like lateral ears.
The Winged Lion came to the fore as an official symbol of Venice in the early 1260s (Rizzi Reference Rizzi2012: vol. 1, 18). Holding a book—a symbol of identity—in its paws, it supplanted the traditional veneration of the Byzantine St Theodoros, condensing into a single image the celestial authority and the earthly power of the Republic (Rizzi Reference Rizzi2012: vol. 1, 17–26, 33, 65; Crouzet-Pavan Reference Crouzet-Pavan2017: 107–17). A seal from 1261 depicts Doge Ranieri Zeno receiving a banner with the lion in moeca from St Mark (Rosada Reference Rosada and Ricci1985). Two bronze capacity measures for wheat with winged lions cast in relief on surface, held in the Venice State Archives, date from 1262 and 1263 (Rizzi Reference Rizzi2012: vol. 2, cat. 663, 65; vol. 1, fig. 2, 18), and a poem by Martino da Canal, written in 1274, describes the Winged Lion on the city’s banner (Tigler Reference Tigler2000: 20).
The adoption of this Marcian banner, from 1261 onwards, coincided with an abrupt redefinition of Venice’s state boundaries, when its primacy in the eastern Mediterranean was faltering. On 25 July 1261, the Byzantine emperor Michael VIII Palaeologus reconquered Constantinople, inflicting a deadly blow to the Latin Empire. Patriarch Pantaleone Giustinian and Venetian podestà (chief magistrate) Marco Gradenigo took refuge in Negroponte, where a “political lion” soon appeared (Pedani Reference Pedani2006: footnote 8, 187). Nicolò Querini, bailo (commander) in Acre had a Winged Lion seal coined in 1263–1264 (Dal Gian Reference Dal Gian1958: 22). Both Negroponte and Acre became border strongholds. Around the same time, c. 1262–1264, Niccolò and Maffeo Polo, the father and uncle of Marco Polo, were trading in Bukhara, before joining a diplomatic mission by khan Hulagu to Khanbaliq (Beijing) and the court of Kublai Khan (Wright Reference Wright1948; Jackson Reference Jackson1998). While there, it is possible that they encountered the original bronze sculpture of the St Mark’s Lion.
Stylistic considerations
The St Mark’s Lion (hereafter, the ‘Lion’) shows little or no iconographic relationship with medieval Romanesque lions (early and middle phases, c. AD 1000–1150) nor with Gothic lions of the early and ‘classical’ phases (twelfth to mid-thirteenth centuries AD). While the style also diverges from Moorish sculptures, around AD 1377–1380 (e.g. The Court of the Lions in the Alhambra, Granada, Spain), the potential Chinese provenance of the bronzework has never been discussed in detail.
Ancient Chinese art is replete with lions and lion-like creatures (Sirén Reference Sirén1925, Reference Sirén1930, Reference Sirén1960; Munsterberg Reference Munsterberg1948; Segalen Reference Segalen1995; Falco Howard Reference Falco Howard2006). Particular comparisons may be drawn with Tang Dynasty artefacts, including an earthenware zhènmùshòu (镇墓兽 ‘tomb guardian’) from the early eighth-century AD Tomb M2 at Fujiagou Village, Lingtai County, Gansu (Figure 2), a series of tricolor-glazed ceramic zhènmùshòu on display in the Luoyang Museum (Henan) and a specimen probably originally from Shaanxi Province now on display at the Cleveland Museum of Art (Figure 3). Other tomb guardians have been lost to the antiquities market (Figure 4). These hybrid creatures share leonine muzzles, flaming manes, horns and raised wings attached to the shoulders, pointed upraised ears and, sometimes, partially humanised facial features.
Figure 2. A Tang-dynasty painted and gilt earthenware zhènmùshòu (镇墓兽 ‘tomb guardian’) excavated from Tomb M2 at Fujiagou Village in Lingtai County (Gansu, China) (image from https://www.timesunion.com/entertainment/article/Clark-Art-Institute-explores-Unearthed-3738598.php).
Figure 3. A Tang-dynasty zhènmùshòu in glazed sancai (three-colour) earthenware, probably from Shaanxi province (courtesy of the Cleveland Museum of Art, 2000.118.1).
Figure 4. A Tang-dynasty white terracotta zhènmùshòu that shows similarities, particularly in the details of the muzzle, to the ‘Lion’ (after Artioli et al. Reference Artioli, Angelini, Vidale, Henderson, Morgan and Salonia2024: fig. 8.7; zhènmùshòu images from https://www.tjitra.nl/T9210-qp-97/view). The sculpture is 620mm high.
Although rendered in different materials, the muzzles of the Tang hybrids resemble the ‘Lion’ in their bulbous noses, the lateral position of the ears, the swellings under the chin, the gnashing of the mouth exposing four powerful canines, and the twisted wrinkle of the forehead at the root of the disproportionately large nose. Zhènmùshòu also often display erect, pointed ears, covered with hair flocks, and the unusual human-like lateral ears of the ‘Lion’ could be the remnants of such bat-like ears. Rather than tapering, the upper edge of the auricles on the ‘Lion’ is thick, blunt and rounded, suggesting that an upper portion may have been sawn off.
Additional modifications are apparent. A ‘wig’ is added to the top of the head, which covers the scars left by the removal of one or two horns (Figure 5; Scarfì Reference Scarfì1990: 76, fig. 66). Traces of the plumage below the attachment of the later wings just behind the shoulders indicates that the original bronze was also winged (Scarfì Reference Scarfì1990: 54–56, fig. 27). These features recall apotropaic funerary figurines of fantastic hybrids from ancient China, such as the píxiū (貔貅), a winged leonine creature, usually with a horn (if male) or two (if female), protruding eyes and sharp fangs, or the bìxié (辟邪), a hornless winged leonine hybrid (Till Reference Till1980).
Figure 5. View of the top of the head of St Mark’s ‘Lion’ once the ‘wig’ was removed. Note the linear-angular contour of the ears and the signs that something—probably horns—has been cut off. The abrupt end of the fur flocks suggest that ears were also sawn off (Scarfì Reference Scarfì1990: 76, fig. 66).
The streaming waves that compose the mane of the ‘Lion’ do, however, lack parallels in ancient Chinese art. After the second century AD, Chinese leonine sculptures, hybridised and winged, had either a solid mane without tufts or, since the sixth–seventh centuries AD, manes of dense curls. Examples of these may be found in the sitting lions of the sacred, monumental ‘Spirit Road’ (神道 shéndào) leading to the tombs of the aristocracy, or in the low-relief sculptures in Buddhist rock temples and cast copper alloys devotional or votive artefacts (Sirén Reference Sirén1925; Paludan Reference Paludan1991).
Large, pre-Ming (before 1368) bronze statues have not survived. Between the fifth and the twelfth centuries AD, Mahayana Buddhism enjoyed the favour of the imperial house. Wealthy monasteries under imperial protection commissioned large devotional, talismanic statues, such as the ‘Buddhist lion’ (佛狮子Fú shīzi) (also popularly called ‘Dog of Buddha’ or ‘Fu Dog’), to guard the entrances to Buddhist temples and monasteries. Yet, between AD 574 and 577 Emperor Wu (r. 543–578, Northern Zhou dynasty) confiscated Taoist and Buddhist estates, and bronze statues were smashed and remelted. In 845, Tang Emperor Wuzong (r. 840–846) enacted the ‘Persecution of Buddhism in the Huichang Era’; 4600 temples and monasteries were destroyed, their properties confiscated, and all bronze statues reused for minting coins. In a third state persecution by Emperor Shizong (r. 954–959, later Zhou dynasty), 3336 temples and monasteries were erased, and all statues weighing more than 2.5kg were melted down, again for coins (Ch’en Reference Ch′En1954; Reischauer Reference Reischauer1955; Palumbo Reference Palumbo2017). As a result, almost no evidence of large bronze Buddhist statues from the seventh to the tenth centuries remains.
From the tenth century AD onwards (Song, Liao, Jin and Yuan dynasties), Buddhist sculptors worked with wood, stucco or stone. When the Mongols conquered China in 1206 and the Yuan Dynasty was founded (1271–1272), Tibetan Vajrayana Buddhism became a de facto state religion. It embedded prominent symbols of Tibetan Buddhism, among them the celestial wingless ‘snow lion’ guardian (Tibetan 雪獅 xuěshī, gangs seng ge). Examples of this later style can be seen in the sculpture of a lion with cubs found in the infill of the Beijing city walls (rebuilt from 1402), now at the Beijing Stone Carving Art Museum, and in the lions of the ‘Black Ditch Bridge’ (卢沟桥 Lúgōu Qiáo), also known as the ‘Marco Polo Bridge’, constructed in the late twelfth century (Ciarla, in press). The bulbous manes of these lions again differ from the waves portrayed on the St Mark’s Lion.
Casting process and isotopic evidence
Although suggestive, stylistic comparisons remain subjective. Examination of the construction and chemical composition of the ‘Lion’ provides a more objective assessment of its provenance. Such assessment builds on previous compositional and stylistic mapping of the different phases of casting and subsequent modification/restoration (Figure 6; Scarfi Reference Scarfì1990), and is made possible through the availability of three samples of the original parts, the development of lead isotope analysis protocols for metal provenancing (Gale & Stos-Gale Reference Gale and Stos-Gale1982; Albarède et al. Reference Albarède, Desaulty and Blichert-Toft2012; Pernicka Reference Pernicka, Roberts and Thornton2014; Baron et al. Reference Baron, Tǎmaş and Le Carlier2014; Villa Reference Villa, Grupe and McGlynn2016; Artioli et al. Reference Artioli, Canovaro, Nimis and Angelini2020), and the recent publication of extensive reference databases of lead isotope data for ores and deposits (Killick et al. Reference Killick, Stephens and Fenn2020; Tomczyk Reference Tomczyk2022).
Figure 6. Six different phases of casting and casting-on identified by Scarfì (Reference Scarfì1990: tab. XV). Phase 1 (from which two of our samples came) and Phase 2 are ascribed to the original Tang sculpture (figure by authors modified from Scarfì Reference Scarfì1990).
Bronze casting
The traditional Chinese bronze casting technique—epitomised by the Shang-Zhou ritual vases of the fifteenth–fourth centuries BC—was piece- (or section) mould casting (Barnard Reference Barnard1961; Gettens Reference Gettens1969; Tan & Lian Reference Tan and Lian2011), enhanced since the fifth–fourth centuries BC by the direct and indirect lost-wax casting techniques (Peng Reference Peng2020, Reference Peng2023). Traditional piece-mould casting was still used in the fifth century AD to cast Buddhist icons, such as the gilt-bronze Standing Buddha Maitreya displayed in the Metropolitan Museum in Washington D.C. (n. 26.123), dated AD 486 (Strahan Reference Strahan, Patry and Strahan2010: 58). The outer surface of the ‘Lion’ does not exhibit any joints suggestive of piece-moulding, but copper core pins and spacers left by lost-wax castings on clay cores are visible both on the earliest-cast parts (Phase 1) and on those from the remodelling phases. Phase 1 core pins have a smaller diameter than those from the later phases, but, as the ‘Lion’ remains atop its column, it is not possible at present to quantify or evaluate the difference. The casting sequence is unusually complex (Figure 6; Scarfì Reference Scarfì1990), perhaps reflecting the difficult fit of the top of a zhènmùshòu-like seated form onto a horizontal leonine body. If this is the case, a reconsideration of the casting sequence is needed.
Lead isotope analysis
Characterisation of the chemical composition of metal artefacts, particularly the isotopic signature of lead traces, provides a reliable means of linking metals to their original ore deposits (Gale & Stos-Gale Reference Gale and Stos-Gale1982; Pernicka Reference Pernicka, Roberts and Thornton2014). Three samples of metal alloy from the ‘Lion’ (Figure 6) were examined using inductively-coupled plasma multi-collector mass spectrometry (ICP-MC-MS) and compared with samples tested previously but never fully interpreted (Sentimenti et al. Reference Sentimenti, Guerriero, Trincherini and Scarfì1990). Sample treatment and the generation of results followed established protocols (Artioli et al. Reference Artioli, Angelini, Kaufmann, Canovaro, Dal Sasso and Villa2017, Reference Artioli, Angelini, Canovaro, Kaufmann and Villa2024b). The three new samples were taken from the frontal part of the curly mane (sample 97, casting phase 1), a frontal repair dowel (sample 98, casting phase 1) and the right wing (sample 99, casting phase 5). These are added to the three samples previously analysed in duplicate from phases 1, 2 and 3 (samples 12, 15 & 18; Sentimenti et al. Reference Sentimenti, Guerriero, Trincherini and Scarfì1990) to provide a total of nine lead isotope measurements now available for the ‘Lion’ (Table 1).
Table 1. Lead isotope analysis data from the St Mark’s ‘Lion’.
The lead isotope ratios of the three new samples are consistent with the results of previous analyses (Table 1; Sentimenti et al. Reference Sentimenti, Guerriero, Trincherini and Scarfì1990: 183, tab. 11). When compared with reference databases for Eurasian ores, including all available data for identified deposits from Iberia to Anatolia, the Caucasus, Iran and China (Iberlid: www.ehu.eus/ibercron/iberlid; OXALID: oxalid.arch.ox.ac.uk; Leuven repository: https://doi.org/10.48804/D4DPLJ; Hsu & Sabatini Reference Hsu and Sabatini2019; Tomczyk Reference Tomczyk2022), the lead isotope ratios from the ‘Lion’ are compatible with a limited range of copper (Cu) ore deposits. Plotting against the worldwide database (Killick et al. Reference Killick, Stephens and Fenn2020) indicates a broad fit with ore deposits related to the Alpine/Tethyan orogenic event that formed Pb-Zn-Cu deposits from North Africa to China (Figure S1). A closer comparison with the available Eurasian deposits (Figure 7), indicates a partial fit with some copper deposits in the Balkans, particularly Bulgaria, though stylistic and historical considerations rule these out as possible origins. A much weaker fit includes peripheral portions of the data clouds of the Swiss, German and Anatolian (Pontic Area) deposits. Available data on other Western Asian deposits (Armenian, Iranian and Omani ones) are also plotted for comparison in Figure 7, but can be excluded as possible origins due to differences in data trends, especially on the geologically significant 208Pb/204Pb versus 206Pb/204Pb plot.
Figure 7. Bi-dimensional plots of lead isotope analysis data for the St Mark’s ‘Lion’ samples (yellow stars) compared with the best-fitting values from Eurasian lead deposits (figure by authors).
Extending the search to a database of Chinese ores (Hsu & Sabatini Reference Hsu and Sabatini2019), copper and polymetallic mines along the Lower Yangzi River show a good fit with the lead isotope ratios of the ‘Lion’ (Figure 8). Some of the Upper Yangzi (Yunnan province) deposits also show a partial overlap with the Lower Yangzi deposits in the 207Pb/204Pb versus 206Pb/204Pb diagram, though the 208Pb/204Pb versus 206Pb/204Pb plot indicates a clear geological discrimination between the two tectonic units. The lead isotope ratios from the ‘Lion’ are particularly compatible with the ores of the Guishan and Yaojialing deposits in Anhui province and the Anji deposits in Zhejiang province—all within the Lower Yangzi tectonic unit (Hsu & Sabatini Reference Hsu and Sabatini2019; a summary table of the statistically most affine deposits and their geographical location appears in Figure S2).
Figure 8. Bi-dimensional plots of lead isotope analysis data for the St Mark’s ‘Lion’ samples (yellow stars) compared with the best-fitting values from Chinese lead deposits (figure by authors).
Chemical composition
Metal samples 97, 98 and 99 were also embedded in resin and semi-quantitively analysed by scanning electron microscopy using energy dispersive x-rays (SEM-EDS) to determine their major elements content (i.e. elements > 1wt%) (Table 2). These results were also compared with previous analyses (Sentimenti et al. Reference Sentimenti, Guerriero, Trincherini and Scarfì1990). Sample 97, from the mane, is a tin (Sn) bronze (Sn 15.7wt%) with a measurable amount of lead (Pb) (Figure 9a), similar to the mean composition of previous samples from phase 1 (Sentimenti et al. Reference Sentimenti, Guerriero, Trincherini and Scarfì1990: 163, tab. 2). Sample 99, from the right wing, has about half the Sn content of sample 97, but shows higher Pb and 2.5wt% of zinc (Zn) (Figure 9b). This composition is comparable to previous samples from phase 5 (Sentimenti et al. Reference Sentimenti, Guerriero, Trincherini and Scarfì1990: 163, tab. 2), which were quite variable in Sn, Zn and Pb content, and is consistent with the known addition of Zn or brass during the recasting of the wings in 1815. Sample 98, however, does not fit well with the compositional class from phase 1, to which it is presumed to belong, due to its low Sn, high Pb and trace amounts of Zn. This composition may result from a mixing of the original metal with that of phase 5, but the complex composition does not identify a clear boundary between the parts that were made and recast at different times.
Table 2. Chemical composition of the samples measured by SEM-EDS analyses (oxygen, iron, nickel, copper, zinc, tin and lead) expressed as the mean and standard deviation (SD) of three or four areas. Inclusions and segregations identified in the samples are listed in the last column.
Figure 9. A) Backscattered electron image of sample 97 from the mane; the high content of the delta phase (pale grey) and the small, rare segregations of lead (white) are visible; B) backscattered electron image of sample 99 from the wing; the white areas testify to the high lead content (figure by authors).
Discussion
In alloys containing very small amounts of lead (i.e. casting phase 1 in Scarfì Reference Scarfì1990), the lead isotope signal reflects the source of copper, as the lead is an incidental addition from the original copper ore. When the lead quantity is appreciable (about a few wt% or higher, such as in the analysed samples, Table 2), the lead isotope signal most likely reflects the origin of the lead added to lower the melting point of the alloy for brazing and amending casting defects (i.e. phases 2–3).
The apparent paradox of the relative homogeneity of lead isotope ratios in different reconstruction and restoration steps can be reconciled if phases 1, 2 and 3 were actually related to the adjustment and amendment of casting defects carried out shortly after casting, well before the statue appeared in Venice (Scarfì Reference Scarfì1990: 55–64). Restoration of the ‘Lion’ was undertaken in 1815 by Bartolomeo Ferrari after the broken parts were returned from France (phase 5; Scarfì Reference Scarfì1990: 49–50). The replaced parts, including the wings (sample 99), were obtained by remelting fragments of the original pieces with added brass, detectable by the high Zn content. Yet the lead signal of sample 99 is comparable with that of the other samples and thus still reflects the original formulation. This is possible only if pure Zn or brass with a very low lead content was added.
All analysed samples therefore show a remarkable homogeneity in the lead isotope composition, despite the complex casting and remaking history of the ‘Lion’. In turn, all samples are compatible with a source in the ores outcrops along the Lower Yangzi River, among the most important deposits of mainland China. Comparison with available data from Chinese artefacts reveals that a bronze fragment from the late Shang period found in Tomb-493 at the Yinxu Stadium cemetery (sample AY-3; Li et al. Reference Li, Wei, Hayan, Tang and Zhang2024) has the same lead isotope signal as the ‘Lion’ (see Figure S3). Although an isolated find, this suggests that the Lower Yangzi deposits were known and exploited at least from Shang times (c. 1600–1050 BC).
In general, information on contact between Europe and China in the medieval period is scarce. While a celadon jar in St Mark’s Treasury is a Qingbai porcelain, made in south-eastern China during the Song or Yuan dynasties and coeval with the travels of the Polo family (Meicun & Zhang Reference Meicun and Zhang2018), a systematic study of Chinese medieval ware imported to Venice has not, so far, been attempted. The violet granite of the column on which the ‘Lion’ stands was possibly brought to Venice with horses and loot from the sack of Constantinople (Venturi Reference Venturi1902; Tigler Reference Tigler2000: 17, 20–22), but this is an unlikely origin for the still mysterious history of the ‘Lion’ itself. If it had been a victim of one of the above-mentioned systematic destructions, it might have belonged in a hoard of sculptural pieces that had escaped melting and reuse for half a century. In turn, that would suggest an early medieval or late antique trade connection between Constantinople and eastern China, but there is little historical or material evidence for that.
Another possibility is that Niccolò and Maffeo Polo, in AD 1264–1268 (when the ‘political’ Winged Lion was actively established), encountered—perhaps in a foundry or imperial storeroom at Khanbaliq—a dismantled Tang bronze funerary sculpture, whose impressive head could be taken as leonine. In the general effort to spread the Republic’s new powerful symbol, the Polos may have had the somewhat brazen idea of readapting the sculpture into a plausible (when viewed from afar) Winged Lion. Although we cannot exclude that the statue was modified by Chinese hands, changes were more likely performed while transforming a hybrid monster into the desired leonine image. Whether and to what extent the Pope and coeval ‘Vatican diplomacy’ were informed by the Venetian emissaries of such a visionary and hazardous project, remains unknown. The two Polos, belonging to a diplomatic web linking the Mongol court, the Pope and other powers, may have sent home large pieces of the hybrid statue, which were then discreetly and laboriously refitted into the reborn holy emblem of St Mark. Was this the birth of the big ‘Lion’ that Marco, Niccolò and Maffeo eventually saw in place, when disembarking at Venice in 1295? Of course, this is only one possible scenario based on the intersection of historical and archaeometallurgical data. The word now goes back to the historians.
Conclusion
Against traditional narratives that hypothesised local, Anatolian or Syrian production, we argue that the muzzle and mane of the hybrid bronze creature have similarities with Tang Dynasty zhènmùshòu (even though comparisons with earlier and later Chinese sculptures, in principle, could not be excluded). It is possible that Marco Polo’s father and uncle, during the four years they spent at the court of Kublai Khan during their first journey, were responsible for the acquisition of the sculpture. Lead isotope analysis of the bronze supports a Chinese origin, identifying likely copper sources in the Lower Yangzi River region. In a puzzling absence of written information, the intention and logistics behind its journey to Venice remain elusive and open to interpretation. If the installation of the ‘Lion’ was meant to send a strong, defensive political message, we can now also read it as a symbol of the impressive connectedness of the medieval world.
Acknowledgements
Many thanks to the anonymous referees for their help improving the article. The authors are deeply indebted to Gisella Capponi, formerly the Director of the Istituto Superiore per la Conservazione ed il Restauro, Rome, and the late Maurizio Marabelli, an unforgettable friend, for their kind permission to analyse the three samples of the bronze ‘Lion’. Thanks also to Eugenio Burgio for his constant collaboration.
Funding statement
This study was funded by support from the University of Padova.
Online supplementary material (OSM)
To view supplementary material for this article, please visit https://doi.org/10.15184/aqy.2025.10159 and select the supplementary materials tab.
Open access
