Spilled Wine, Spilled Blood: Spilling the Secrets of the Covered Cup from the Chungul Kurgan

The covered cup that is the object of this study has spent the majority of its existence in darkness (fig. 1).^[1] Placed in a grave in the steppelands of southern Ukraine only a few decades after its manufacture, it was brought to light again in 1981 with the excavation of the nomadic burial mound in which it had been interred. It now sits, as of this writing, in a secure and undisclosed location to prevent its destruction or looting in the ongoing Russian attacks on Kyiv.^[2] Apart from being an impressively large and well-preserved example of secular metalwork from around the year 1200, the cup conceals a couple of surprises that further underscore its importance. First, when one removes the lid, one finds within the bowl of the cup a cast silver-gilt lion poised on its hind legs against a centrally positioned post (fig. 2). Second, that same post conceals a hitherto undetected siphon mechanism, making the cup one of the first preserved automata from Europe and so far the earliest known Western medieval example of this type of trick vessel. Although the cup finds no exact stylistic parallels in Western European metalwork of the late twelfth century, its presence in the same burial with other works that seem to emerge from the same approximate date and region can suggest something of the historical circumstances that might have resulted in the cup’s burial with a nomadic leader in the steppe.

1

Covered cup from the Chungul Kurgan burial (conserved state), late 12th – beginning of the 13th century, gilded silver, 29.5 × 16.5 cm. Kyiv, Treasury of the National Museum of History of Ukraine

2

Covered cup from the Chungul Kurgan burial, view of cup interior

The covered cup was discovered in 1981 in the course of the salvage excavation of a kurgan, or burial mound, in the Zaporizhzhia Oblast of southern Ukraine (fig. 3). The dig was carried out by a team from the Institute of Archaeology of the (then) Academy of Sciences of the Ukrainian SSR, with the objective of clearing the way for new irrigation systems.^[3] With the expansion of Soviet agriculture setting the timeline, the kurgan was fully excavated in a single season using motorized earth-moving equipment.^[4] The mound, designated as the Chungul Kurgan, contained a series of burials dating from the early Bronze Age (Yamna culture, ca. 3300–2600 BCE) to the late Bronze Age (Zrubna culture, ca. 1900–1200 BCE).^[5] At its center, the excavators discovered a medieval burial, dating to the early thirteenth century CE, sunk through the preexisting Bronze Age kurgan and topped with an impressive tumulus reaching almost five meters in height. It contained the skeleton of a male, estimated to have been 50–60 years old at his death. Around the perimeter of the burial pit were found the skeletons of five sacrificial horses, at least three of which were buried with bridles, saddles, and other trappings.^[6] The burial chamber itself contained two large amphorae— likely to have been filled with wine—and the carcasses of approximately ten sheep to supply the deceased with food and drink for the afterlife, along with a variety of grave goods (fig. 4).^[7] The back of the skull of the deceased bore a sharp wound, as from a sword. Pinholes on either side of this fissure, as though from a suture, and a trepanation drilled into the skull with some signs of bone regrowth at its edges attest that he survived for a time after this blow and received medical attention.^[8] The glazed albarello found in the grave (of a type associated with Ayyubid Raqqa in the early thirteenth century) probably contained medicinal remedies or analgesics meant to mitigate the suffering of the wounded leader.^[9]

3

Map of the Black Sea region, with enlarged detail showing Dnipro and Azov watersheds

4

Chungul Kurgan, view of burial pit at time of excavation, showing skeleton and grave goods

At the time of its excavation, the lower half of the cup was partially filled with remnants of organic matter, some of it consisting of charred herbs.^[10] Together with the contents of the albarello, these botanical remains were evidently part of the pharmacopoeia placed in the burial to ease the pain of the deceased in the afterlife. The presence of these burned materials led to the initial—and erroneous—description of the cup as a censer (kuryl’nytsia or kadilo in Ukrainian), an error repeated in some subsequent publications.^[11] Presumably the cup was still used as a drinking vessel by its owner up to the time of his death and its final reuse in the burial. The confusion of the cup for an incense burner is the more understandable due to the presence of perforations, as though to facilitate the aeration of coals, in the base of the interior column. Rather than being intended for air, however, these holes are integral to the function of the hidden siphon, as will be explained below.

A burial of this type in this location, carried out within a nomadic religious tradition of providing necessities for the afterlife, must have belonged to a leader of the Qıpčaqs (known as Cumans or Polovtsy in Byzantine and Slavic sources, respectively), a nomadic confederation that dominated the Pontic steppe from the mid-eleventh century until the Mongol invasion in the mid-thirteenth century.^[12] The grave contained weapons and a gilded iron helmet of a type comparable to finds from Kyivan Rus’, armor apparently of nomadic manufacture, a splash-glazed ceramic bottle and a ring-handled silver cup with enamel decoration possibly from Constantinople, and silk textiles of Eastern Mediterranean origin.^[13] The metalwork attributable to Western Europe includes the covered cup itself, three elaborate belt buckles also of gilded silver, and two finger rings of gold, mounted with stones. The dates that can be assigned to the grave goods mean that the burial can date no earlier than the first decade of the thirteenth century. The mobilization of Qıpčaq labor necessary to erect this impressive tomb would have been well-nigh impossible after the consolidation of Mongol control over the Black Sea Steppe in 1237.^[14] There is, therefore, a window of roughly thirty years in the early thirteenth century when this burial is likely to have been deposited.^[15]

Within the Qıpčaq notable’s burial, the covered cup was found lying just above the right shoulder of the deceased. At some time in the modern era, the mound suffered a partial collapse, leading to the crushing of the internal platform that protected the burial pit and sending debris into the pit itself. Fortunately, the cup escaped destruction. Although the foot was separated from the body and lay at an angle to the cup’s upper section, the vessel emerged largely intact from the excavation. Local museum officials in the Zaporizhzhia Regional Museum, anxious to put this impressive find on display, carried out a hasty restoration process in the months following the excavation, cleansing the cup of encrusted corrosion, reattaching the footto the vessel, and restoring the broken portions of the acanthus collar. Unfortunately, apart from photographs taken of the cup at the time of excavation, the process of restoration was not documented.

The cup measures about 29.5 cm high and 16.5 cm in maximum diameter; it weighs a bit under a kilogram in total. The vessel sits on a flaring conical foot, 10.5 cm in maximum diameter, open at the bottom. A ring of acanthus leaves marks the transition from the foot to the bowl of the cup. The profile of the cup and its lid approximates a sphere compressed at the waist. This slightly “cinched” shape is remarkably consistent across both Byzantine and Western medieval covered drinking vessels of the twelfth and thirteenth centuries. The two parallel bands of ornament at the waist of the cup—rather than a single band attached to the lower section—seem to distinguish covered cups of Western manufacture from those that can be attributed to Byzantium or its eastern neighbors.^[16] The lid features a high finial consisting of an openwork sphere on a flaring conical foot; both are rendered as a network of vines and tendrils with cupped leaves, crowned by a bud with four radial petals. This is an unusually elaborate handle for a cup lid, and its closest parallels can be found among much larger objects such as reliquaries and shrines.

The surface of the cup and its lid are decorated with engraved ornament against a matte ground. The principal motif, seen in the six wide bands (two each on the lid, bowl, and foot), consists of interlaced sprays of very attenuated acanthus leaves (figs. 1, 5). The bands where the lid and cup meet are decorated with a motif of four leaves radiating diagonally from a central point. On the lower surface of the cup, there is an additional band formed as a row of tendrils, which branch alternately to the right and to the left. The corresponding position on the lid has only a narrow band of engraved lozenges, within which is secured the cast finial. The way in which this ring of ornament frames the attachment of the finial makes clear, as will be discussed below, that the cast and engraved components of the cup’s decoration were coordinated from the outset.

5

Covered cup from the Chungul Kurgan, detail of cup and attachment of foot

At the join between the cup and the foot is a ring of acanthus leaves apparently worked by cutting and hammering silver sheet. Each leaf is formed as a five-lobed structure, articulated with parallel grooves. They are separated at their bases by round notches around the collar from which all eight leaves emerge and curve upward toward the bowl of the cup. Two of the leaves were broken off prior to the cup’s excavation and were re-attached as part of the restoration.

The finial, as already mentioned, consists of a cast openwork of tendrils that form a ball and a flaring conical foot, which ends in an attachment ring connected to the lid with rivets (fig. 6). The upper and lower halves of the ball each consist of six tendrils, spiraling alternately clockwise and counterclockwise. In this respect they resemble the tendril ornament of the cup’s bowl, only here translated from two dimensions into three. The two zones of scrolls are separated around the circumference of the ball by a torus-shaped “equator.” Another torus separates the neck of the finial’s foot from the ball. The foot is formed of four larger tendrils, all right-hand spirals, with their smaller branches. The foot terminates in a solid ring with a scalloped edge, 6.2 cm in diameter, in which are placed the rivets affixing the finial to the cup’s lid.

6

Covered cup from the Chungul Kurgan, detail of finial of lid

When the lid is removed, the inside of the cup reveals a central post, which supports the cast figure of a lion reared on its hind legs. The lion measures nearly 8 cm long from its rear paws to its muzzle, from which its enormous tongue protrudes as though the creature were licking the post’s berry-like finial (fig. 7). Its eyes are inlaid with a glossy black material, probably jet, and its mane is articulated with engraved decoration. The body is lean and sinewy, and the ribs are clearly articulated on the lion’s flanks. The tail is modeled as passing between the hind legs and then curling around the belly and over the left side of the back, ending in a tuft. The lion sits atop a raised base, 3.4 cm in diameter, with a flat flange through which pass the four rivets that attach it to the cup. Six holes are pierced in the domical part of the base, out of the center of which rises the post, 1.5 cm wide and 6.7 cm tall, terminating in a stylized berry. This form, representing a berry, pinecone, or, perhaps, beehive is articulated with rows of indentations from a ring punch. This object is now partially torn away, revealing an inner silver cylinder within the encasing silver-gilt post.

7

Lion and post from interior of covered cup

The interior surface of the cup, in contrast to the exterior and to the central lion and pillar, is mostly plain silver apart from a few details picked out in gilding. Four small gilt roundels on the sides of the bowl, each about 2.5 cm in diameter, feature engraved animals: a canid (?), a griffin (fig. 8), a ram (?), and an eagle. As the question marks indicate, these are rather sketchily rendered, and they form a contrast to the far more careful anatomical delineation of the cast lion. Around the latter, near the bottom of the bowl, is a narrow, gilded band engraved with overlapping S-shaped motifs to create a wavy pattern.

8

Engraved roundel with griffin on interior of covered cup

Morphology

The cup’s form, style, and artistic techniques place its creation in northwest Europe sometime in the last quarter of the twelfth or beginning of the thirteenth century. As we shall see, there is no absolute match that can pinpoint a date and location of manufacture, but comparisons can help us to define a zone of probable origin and plausible date. The Chungul cup is by no means alone in this respect—dating and localization of treasury objects in the decades around 1200 seems to be a particularly thorny issue, both because of the number of objects found far from their place of manufacture and because of the emulation of similar motifs by artisans in a number of different locales.

We can begin with the covered cup form, the origins of which have concerned historians of metalwork for many decades.^[17] Whether sacred or secular, this vessel type, with a bulbous, domed cover approximately matching the shape of the bowl and with a cinched waist where bowl and cover meet, appears rather suddenly in works around 1150 and disappears again around the middle of the thirteenth century. Examples appear more or less simultaneously around Western Europe, from England to Scandinavia to the Limousin.^[18] These vessels include a number that have an unambiguously liturgical function as a container for the reservation and distribution of consecrated hosts. The Ciborium of Master Alpais in the Louvre (fig. 9) is very similar in its overall shape and dimensions to the Chungul cup, but it was clearly meant as a sacred vessel, as attested by its decoration of angels and apostles. There is an engraved roundel with the dextera Domini within the lid, while in the bottom of the bowl, the figure of a gesticulating angel is surrounded with the maker’s inscription: +MAGISTER G. ALPAIS. ME FECIT. LEMOVICARUM (“Master G. Alpais made me at Limoges”).^[19] The location information in the inscription is in keeping with the medium of gilded copper and the techniques of champlevé enamel and die-cast appliqués typical of Limoges.

9

Ciborium of Master Alpais, ca. 1200, gilded copper with champlevé enamel and glass cabochons, 30.1 × 16.8 cm. Paris, Musée du Louvre

Unfortunately, no other cups of this form from the period so helpfully name their place of manufacture. A case in point is the covered cup known as the “Coupe de Charlemagne” in the treasury of the abbey of St. Maurice d’Agaune in Switzerland (fig. 10). Like Master Alpais’s ciborium, it shares with the Chungul cup the same general form of a near-spherical vessel pinched at the waist and mounted on a trumpetlike foot. In contrast to the Chungul cup, however, it is constructed as a double shell, with an outer part supporting the repoussé roundels and an inner lining providing a smooth interior surface. The attribution of the St. Maurice cup has been much disputed: England, the Meuse valley, and Cologne have all been proposed.^[20] Furthermore, there has been debate over whether it was originally meant as a eucharistic ciborium or a secular drinking vessel. Despite the repoussé decoration of angels and narrative scenes from the Infancy of Christ, recent scholarship seems relatively united in the opinion that the cup is, in origin, a secular object. Furthermore, the cast group of a centaur teaching a young student (likely meant to represent Chiron and Achilles) is currently attached as the finial of the lid, but in an inventory of 1659, it was recorded as being mounted to the bottom of the bowl.^[21] Scholars are not in agreement as to which was the original position.^[22] Whatever its original placement, the cast and gilded silver figural group is significant as a parallel to the gilded silver lion from the Chungul cup.

10

Coupe de Charlemagne, ca. 1210 – 1220, gilded silver, 27.8 × 20 cm. Saint-Maurice d’Agaune, Treasury of the Abbey

Another close formal parallel to the Chungul cup is the gilded silver “Sainte Coupe” in the Cathedral Treasury of Sens. Despite lacking any obvious iconographic signifiers of its function, its finial on the lid is fitted with an integral suspension ring, which suggests that it was always intended to hold eucharistic wafers and be suspended above an altar (fig. 11).^[23] Traditionally, the cup has been linked with Thomas à Becket and his exile at Sens from 1166 to 1170. The implied manufacture in England prior to 1166 seems a bit early for the elegant style of the cup, however, and scholars seem more comfortable placing it somewhat nearer the year 1200.^[24]

Apart from the Coupe de Charlemagne at St. Maurice—assuming that it was, in fact, a drinking cup—most of the secular examples of covered cups from Western Europe survive only as fragments.^[25] The partly gilded and nielloed silver bowl from the Basilevsky collection, now in the Cloisters, for instance, lacks its original foot as well as its lid.^[26] The Dune Treasure has a bowl with its conical foot but without a lid, as well as another drinking vessel lacking both lid and foot.^[27] Numerous other examples have been found in the former territories of the Russian Empire and Soviet Union. From the trove of vessels found near Muzhi in the Ob’ valley of northwest Siberia, for instance, we have two separate lids from lost cups as well as a cup not associated with either lid.^[28] An extraordinary lid of solid gold—once again lacking its cup—was found in Chernihiv in 1957. It bears an elaborate inhabited rinceau in repoussé, and the interlace motif around its rim recalls that on the cup from Tahancha in central Ukraine, now in Warsaw.^[29] Again, it bears emphasizing that the Chungul cup is exceptional in being found in damaged but complete condition.^[30]

11

Sainte Coupe de Sens, ca. 1200, gilded silver, 31 × 19 cm. Sens, Cathedral Treasury

Representations of this covered cup form also appear in various artistic media from the second half of the twelfth century and into the early thirteenth century. Notable among these are the twin cups held by the king at the top of Fortune’s wheel in the Hortus Deliciarum of Herrad of Landsberg.^[31] On a leaf of pen sketches within a twelfth-century manuscript of Isidore of Seville’s Etymologies in Vienna, a figure in secular dress holds up a similar cup by its foot (fig. 12).^[32] In a Parisian Bible Moralisée, likewise in Vienna, a gilded lidded cup sits among the spoils of Egypt taken by the fleeing Israelites.^[33] One of the Magi presents such a cup to the infant Christ on a Limoges casket in the Walters Art Museum.^[34] The use of this type of vessel for liquids—and not exclusively for holding consecrated eucharistic hosts—is made clear by a set of cast and gilded bronze personifications of the four elements, attributed to the Meuse valley and now in Munich. The figure of Water holds a covered cup in her hands (fig. 13).^[35]

12

Leaf of drawings from a volume of Isidore of Seville, second half of 12th century, ink on parchment, 30.6 × 21.4 cm. Vienna, Österreichische Nationalbibliothek, Cod. 67, fol. 1v

13

Personification of Water, ca. 1180, gilded bronze, 10.5 × 5.7 × 4.9 cm. Munich, Bayerisches Nationalmuseum

In considering the history of the vessel’s form, we should be attentive to how fragmentary our surviving evidence is. A striking reminder of this comes in Galbert of Bruges’s account of the assassination of Count Charles the Good of Flanders in 1127—a text to which we shall return. Galbert makes explicit mention of a gold covered cup that belonged to the count (and was misappropriated by clerics after his murder) that, with its lid, weighed some seven marks of gold (about 1.6 kg).^[36] The cup does not survive, despite being recovered from the church and passed on to William Clito, Charles’s successor as Count of Flanders, so we cannot reconstruct its shape, but it raises the possibility that such lidded vessels were in circulation a good half century earlier than we have surviving evidence for them.

Ornament

The patterns chased and engraved on the exterior of the Chungul covered cup consist of attenuated, “leggy” acanthus interlaced against a finely matted ground. I have not been able to find any examples of twelfth- or thirteenth-century metalwork with precisely this pattern. A cup of Western European style found in Kyiv in1876 and now in the Hermitage features a narrow band of windblown acanthus ornament around its lip and a central rosette of long acanthus leaves in its bowl,^[37] but the angular quality of the engraving and the technique of the matted background are distinctly different from those of the Chungul cup. The engraved ornament of a pyx of the late twelfth or early thirteenth century from Saint-Omer shows a similar overall aesthetic to the Chungul cup, but again, without exact parallel to the form of leafy interlace.^[38] The intertwined forms on the Saint-Omer vessel are thicker and bolder than the leggy acanthus of the Chungul cup, and they merge without any border into the burnished surface of the pyx’s stem. This feature stands in contrast to the clean divisions between the bands of plain gilded silver and matted-ground registers of ornament on the Chungul cup. Vaguely similar, too, is the border formed of interlaced leafy motifs on an enamel plaque of ca. 1180 in Chantilly, attributed to the master of the portable altar of St. Gregory in the treasury of St. Servatius in Siegburg.^[39] The Gregory altar itself, generally placed in the 1170s, features analogous forms around the perimeter of its top, but again in champlevé enamel rather than engraved silver.^[40] The interlaced acanthus spray turns up again in several variations in the enamels of the tower reliquary in Darmstadt, attributed to Cologne ca. 1180.^[41] Finally, there are somewhat similar sprays of acanthus and other vegetation in the enameled background of some of the apostles from the shrine of St. Anno made for St. Michael’s in Siegburg, dated to about the time of the archbishop’s canonization in 1183 (fig. 14).^[42]

14

Figure of St. Bartholomew and surrounding ornament, detail of Shrine of St. Anno, ca. 1183, gilded bronze, champlevé enamel, rock crystal and other stones over a wooden core, overall dimensions 78 × 157 × 46 cm. Siegburg, St. Michael

Turning from engraved acanthus to the hammered leaves of the collar below the cup’s bowl, one finds similar leafy transitional zones on a number of works of ars sacra found in western Germany and eastern France in the late twelfth and early thirteenth centuries. The elaborate enameled finials of the shrines associated with late twelfth- and early thirteenth-century Cologne, such as the shrine of St. Anno just mentioned, are frequently set off with collars of gilt leaves. The champlevé enamel orb from a shrine at St. Ursula in Cologne, for instance, formerly in the Stoclet collection and dated ca. 1180, bears rings of gilded copper-alloy leaves above and below. The upper group centers a berry with ring-punched drupelets very similar to the form that crowns the central post of the Chungul cup.^[43] The pinnately lobed leaves, however, are distinct from the leaves of the Chungul cup, and they are also cast in copper alloy rather than hammered from a sheet of silver. Somewhat closer in form to the Chungul acanthus collar is a gilt copper and enamel finial in the Metropolitan Museum of Art, attributed to a Rhenish workshop and dated 1185–1200 (fig. 15).^[44] Here the leaves, rather than being scored by a sharp division at the central vein, are arranged in parallel lobes much like the leaves on the Chungul cup. The closest analogy of form and medium, however, comes with the leafy collars on the silver reliquary of the True Cross from Laon (fig. 16), where rings of alternately upward- and downward-facing leaves surround the knop at the foot of the cross. The socles that support the figures of the Virgin Mary and John the Evangelist are formed as cups of leaves that, like the leaves on the Chungul cup, spring from a ring and are bent 180° back towards their stem (fig. 17).^[45] The inscription, CRUS [sic] HUGONIS ABBATIS, allows the cross to be associated with Hugh, the abbot of Saint-Vincent at Laon from 1174 to 1205, thus establishing a window for the dating of the object and a probable manufacture in or around northern France.^[46]

15

Spherical knop, late 12th century, gilded copper and champlevé enamel, 13.7 × 9.4 cm. New York, The Metropolitan Museum of Art

16

True Cross reliquary of Abbot Hugh from Saint-Vincent, Laon, datable to between 1174 and 1205, silver with gilding, filigree, stones, and colored glass, 46.8 × 12.5 cm. Paris, Musée du Louvre

17

Figure of St. John on a base of leaves, detail of the True Cross reliquary of Abbot Hugh. Paris, Musée du Louvre

The other applied ornament of the cup’s exterior, the elaborate cast finial of openwork design (see fig. 8), is both highly distinctive and surprisingly difficult to place. Numerous analogies suggest themselves, but none are exactly parallel. Somewhat similar to the foliage of the Chungul finial is the curling head of a crozier in the treasury of St. David’s Cathedral in Wales, dating to the twelfth century but of uncertain attribution.^[47] The cupped leaves of the gilt bronze crook echo the forms of the finial, although their scale is somewhat larger (the whole measures 16 cm across) and the rinceau is formed into a disk rather than a ball. Some of the capitals from the colonnettes of the St. Anno shrine in Siegburg (see fig. 14) show the same round profile as the tendrils from the Chungul cup finial, together with somewhat analogous curled-over leaves. Here, again, the work is cast in bronze rather than in silver.^[48]

For analogies to the spherical form taken by the tendrils, the most obvious parallel is the finial of the domed reliquary from the Guelph Treasure, now in the Kunstgewerbemuseum in Berlin and dated on stylistic grounds to about 1200 (fig. 18).^[49] The round profile of the vines matches the scrolls of the Chungul cup finial; the compound leaf forms on the Berlin reliquary, however, are both larger and more precisely defined than the leaves on the cup lid, there is no division around the midline of the sphere, and the sphere is supported on a short, solid collar rather than the high foot of further openwork. The material of the Berlin reliquary finial, moreover, is cast and gilded bronze rather than the gilded silver of the Chungul finial. Another openwork finial appears on the silver-gilt mounted mazer of ca. 1180 known as the “skyphos of St. Nicholas,” in Brauweiler, which is crowned by a cast, openwork knob.^[50] While the technique is the same as the Chungul cup, the Brauweiler knob is divided into four vertical sections, each dominated by a pinnate leaf form. The sphere sits directly on a thin silver collar, which is connected in turn to the wooden lid of the mazer; there is no openwork neck, as on the finial of the Chungul cup, nor any tendrils.

18

Domed reliquary from the Guelph Treasure, end of the 12th century, champlevé enamel and vernis brun on copper, gilded bronze, silver and gilded silver, walrus ivory, over a wooden core, 45.3 × 41 × 41 cm. Berlin, Kunstgewerbemuseum

Interestingly, in his posthumously published revised and expanded study of “oriental” silver, Boris Marshak proposed an attribution for the Chungul cup. Working from published photographs, Marshak attributed the cup itself to Picardy or Flanders around 1200, but, based on the false assumption that the vessel functioned as an incense burner, he proposed that it was later altered into a censer by a second craftsman’s addition of the openwork finial, which he attributed to a Cologne workshop of the 1170s or 1180s.^[51] He arrived at the attribution to Picardy or Flanders by comparison with the engraved surface decoration of the pyx from Saint-Omer, mentioned above.^[52] The Cologne attribution of the knob comes from comparisons we have already made—with the finial from the Guelph reliquary now in Berlin and the finial of the mazer in Brauweiler—as well as with the knobs of filigree openwork atop the Shrine of St. Anno.^[53] These formal comparisons are reasonable—although the wire filigree knob of the St. Anno shrine is quite different from the much smaller, cast silver knob on the Chungul cup—but close examination shows the scenario Marshak proposed to be highly unlikely. The lid of the cup has no perforations for smoke to escape; furthermore, as already mentioned, the perimeter of the finial’s foot matches almost precisely the innermost ring of engraved decoration on the lid, and no decoration can be seen within the circumference covered by the finial. The base ring of the finial is affixed to the lid by four rivets that match the gilded silver finish of the remainder of the cup’s exterior. Moreover, they are consistent with the attachment of the inner group of the lion and post to the bottom of the bowl. Finally, the tendril pattern engraved near the bottom of the cup (see fig. 5) shares the same organizing principle as the cast tendrils of the finial. In short, Marshak’s hypothesis of a bricolage of Cologne and northeast French work of different dates is almost certainly wrong.^[54] The confusion of the great Russian scholar of medieval metalwork is itself, however, illustrative of the conundrum posed by the Chungul cup and its decoration.

Most puzzling of all is the cast and gilded silver lion dominating the interior of the bowl of the Chungul cup. The twelfth-century treatise on the arts by Theophilus recommends animal figures in repoussé (not cast) silver for the insides of secular drinking vessels: “Also, in the same workmanship, one fashions, in the middle of gold or silver cups or dishes, horsemen fighting against dragons, or lions or griffins fighting, the figure of Samson or David breaking the mouths of lions, also lions and griffins by themselves …,”^[55] Such a figural group of a lion and a basilisk is found (in cast silver, rather than in repoussé) in the center of one of the drinking bowls from the Dune Treasure in Sweden.^[56] Another cast and gilded silver figure of a hybrid creature (a senmurv?) sits within a footed cup found in the Kyiv region, now in a private collection.^[57] As already mentioned, if the cast group of Chiron and Achilles from the Coupe de Charlemagne at St. Maurice d’Agaune originally sat within the bowl, it would provide still another parallel from the period (see fig. 10).^[58]

The lion from the Chungul cup is distinctly more naturalistic than most twelfth-century depictions. It bears little resemblance, for instance, to the cast and gilded bronze evangelist symbol of St. Mark found on the cross base from Saint-Omer or to the strangely anthropomorphic lion-headed angel on the St. Anno shrine.^[59] In the articulation of the ribcage and the form of the mane, the lion from the Chungul cup calls to mind the griffin-form feet from the domed shrine in the Victoria and Albert Museum, attributed to a Cologne workshop around 1180 (fig. 19).^[60] The rearing pose is surprisingly rare in metalwork of this period; it finds its closest parallels in the trio of bronze lions that form the foot of a cross from the Guelph Treasure, now in the Kunstgewerbemuseum in Berlin, attributed to Saxony or Lower Saxony and dated ca. 1130–1140 (fig. 20).^[61] The lions of the cross base sharewith the Chungul lion the detail of the tail wrapped between the legs and around the body, albeit in opposite directions (wrapped right to left rather than left to right). The object that the lion licks, on the other hand, can be compared to finials on Rhenish metalwork such as the shrine of St. Anno mentioned above, where it crowns the central ball-form finial. Furthermore, one of the capitals of the St. Anno shrine features a pair of lions, each with his tail wrapped between his legs and terminating in a tuft on his flank.^[62] This same wrapping of the lion’s tail is seen in the drawing from the Österreichischen Nationalbibliothek in Vienna (see fig. 12). Again, the comparisons are not precise, but details shared with other objects seem to confirm an attribution of the Chungul cup to the region of the lower Rhine or Meuse valley and a date in the final decades of the twelfth century.

19

Griffin-form foot, detail of a domed tabernacle, probably from St. Panteleon in Cologne, ca. 1180, gilded copper, overall dimensions 55.5 × 51 × 50.8 cm. London, Victoria and Albert Museum

20

Base of an altar cross in the form of three rearing lions, gilded bronze, base 13.4 × 19.1 cm. Berlin, Kunstgewerbemuseum

The Siphon

As noted above, the lion and post stand on a shallow dome-shaped foot, about 3.4 cm across, which has a flattened flange around its outer edge. This flange is pierced by four rivets that secure the casting to the base of the cup (fig. 21). Above these, on the domical foot of the post, six holes are pierced. These holes, together with the concentric tubes making up the central post, were noted in earlier publications of the cup, although no conclusions were drawn as to their function.^[63]

21

Section drawing and details of Chungul cup

Their logic becomes apparent when compared with another extraordinary cup of much earlier date, the fourth-century Roman Tantalus cup from Vinkovci in eastern Croatia (fig. 22).^[64] This cup was discovered in 2012 in excavations in the town of Vinkovci, which stands on the site of the ancient Roman settlement of Cibalae.^[65] This cup has a central figure of Tantalus, who is seated on a rock, bending forward, with his hands outstretched in front of him. Around the base of the rock, which is raised from sheet silver, there are four openings. These small holes permit the flow of liquid into a siphon tube concealed within the figural group. Thus when the bowl was filled to a level just below Tantalus’s outstretched hands, the siphon would be triggered and the bowl’s contents would drain out through the pierced foot of the cup, potentially soaking the unwary drinker.^[66] An inscription around the rim, a quotation from Phaedrus, makes the figure of Tantalus a metaphor for the greedy who are unable to grasp the good things around them.^[67] The engraved aphorism and the repoussé figure thus work together to enhance the humorous effect of the vessel’s secret mechanism.

22

Tantalus cup from Vinkovci, Croatia, silver, 8.3 × 19.6 cm diameter. Zagreb, Archaeological Museum in Zagreb

The Chungul cup would have functioned in a similar manner to the Vinkovci bowl. When filled to the level of the top of the hidden tube within the post, the wine in the cup would trigger a siphon mechanism, causing the cup’s contents to drain out via the holes placed around the feet of the lion (fig. 7, 23). An opening within the foot of the cup (fig. 24) would direct the liquid either onto the ground or onto the person holding the cup. Although the Chungul cup has no inscription, the lion straining at the berry or beehive at the top of the column conveys something analogous to Tantalus’s futile reach for the liquid in the Vinkovci vessel.

23

Conceptual diagram of siphon mechanism within the Chungul cup

24

Chungul cup seen from below

The Tantalus cup from Vinkovci is the lone identified example from antiquity of a type of trick vessel described in the works of Hero of Alexandria.^[68] Hero’s twelfth model in the Pneumatica is entitled “A type of vessel, which if not filled up, does not run, but if filled empties itself of all the liquid it holds.”^[69] His example consists of a basin fitted with a tube in the form of either a bent siphon or an enclosed tube within a larger cylinder (a so-called concentric siphon) that pierces the bottom of the vessel.^[70] When filled to the level of the bend in the bent siphon—or the top of the inner tube of the concentric siphon— the siphon action is triggered.^[71] As Hero notes, if the opening of the siphon is placed sufficiently close to the bottom of the vessel, the liquid will drain completely.^[72] The diagram illustrating Hero’s trick cup in Marcianus graecus 516, fol. 171r (a manuscript of the fourteenth century, which is generally assumed to be modeled on the late antique original of Hero’s text) shows both the bent and the concentric siphon within the same vessel (fig. 25).^[73] The concentric siphon mechanism is the same one used in both the fourth-century Vinkovci cup and the twelfth-century cup from the Chungul Kurgan. This variant siphon mechanism appears in the work on fine mechanics (Kitab al-hiyal) by the Banu Musa in the mid-ninth century, where it is introduced in the first example vessel,^[74] but is absent from the most famous work on pneumatics in the Islamic world, the Kitab fi ma ‘rifat al-hiyal al-handasiya (The Book of Knowledge of Ingenious Mechanical Devices) of al-Jazari, completed in 1206, which appears to rely on the bent siphon tube exclusively for its elaborate automata.^[75] Both the bent and the concentric siphons can be found among the so-called “Pythagoras cups” produced as novelty items in Greece and Italy since the nineteenth century. The attachment of the philosopher’s name to the device seems to be a product of modern whimsy, although in continuity with the near-mythical status accorded Pythagoras from antiquity onward.^[76]

25

Hero of Alexandria, Pneumatica, illustration (lower) of vessel with bent and concentric siphons, ink on paper, 30.7 × 22 cm. Venice, Biblioteca Nazionale Marciana, MS gr. 516, fol. 171r

The notebook of Villard de Honnecourt (datable to ca. 1220–1240) also contains a sketched design for a cup with a central siphon mechanism (fig. 26).^[77] Villard accompanies the drawing with a short description of its function in Old French:

26

Villard de Honnecourt, sketchbook leaf showing drawing of a chantepleure, ink on parchment, 23.5 × 15.5 cm. Paris, Bibliothèque nationale de France, MS fr. 19093, fol. 9r

See here a chantepleure [lit.: sing-and-cry] that can be made in a goblet, in such manner that in the middle of the goblet there must be a little tower. And in the middle of the little tower there must be a tube that touches the bottom of the goblet, so that the tube is as long as the goblet is deep. And in the little tower there must be three crosspieces across the bottom of the goblet, so that the wine in the goblet can go into the tube. And above the little tower there must be a bird that must hold its beak so low that, when the goblet is full, it drinks. Then the wine will flow through the middle of the tube and through the middle of the foot of the goblet, which is double. And understand well that the bird must be hollow.^[78]

This elaborated hanap, or chantepleure, as Villard terms it, has at its center a crenelated tower.^[79] When the cup was filled past the trigger-point of the siphon, the bird perched atop the tower would appear to suck up all the liquid from the cup. There are some inconsistencies between Villard’s description and his drawing, and the somewhat garbled account of the cup’s functioning has led previous authors to suggest that Villard encountered such a vessel only through hearsay.^[80] On the one hand, Villard’s insistence that the bird be hollow and “drink” when the cup is full implies a bent siphon (which would not function in the cup as he has drawn it, because the bird sits too high above the rim of the vessel).^[81] On the other hand, the interior fittings of the turret that he describes match with the concentric siphon found in the Vinkovci Tantalus cup and in the covered cup from the Chungul Kurgan. The key to the interpretation lies in the enigmatic “IIJ. traveçons par sontre le fons del henap.”^[82] These seem to be tubes—explicitly mentioned as being at the bottom of the cup—that allow the wine to flow into the siphon. This would seem to rule out a bent siphon channeling wine from the beak of the bird, but would agree with a concentric siphon concealed within the tower.

Earlier reconstructions of Villard’s chantepleure have failed to make full sense of the information given in his description. The discovery of actual late antique and medieval trick vessels can help us now in clarifying how Villard’s cup must have been meant to function. In their reconstructions, both Hans Hahnloser and Roland Bechmann passed over the possibility of a siphon in favor of a mechanism relying on displacement of air by the overflowing liquid.^[83] Both scholars gave priority to the movement of the “drinking” bird—a movement not actually specified in Villard’s text—over the drainage of the liquid itself. Bechmann’s reconstruction, in fact, renders the object hardly usable as a drinking cup at all, as the reservoir of air that is meant to power the bird’s supposed movement takes up almost the entire volume of the vessel. Already in the mid-nineteenth century, Robert Willis, commenting on Villard’s notebook, correctly recognized the cup as an iteration of Hero of Alexandria’s twelfth example from the Pneumatica.^[84] The bird, in this analysis, has no mechanical function, but rather by appearing to drink aids in the illusion and amusement created by the cup, much as the figure of Tantalus does for the cup from Vinkovci.

Willis also recognized a surviving parallel to Villard’s chantepleure in the cup known as the Swan Mazer, belonging to Corpus Christi College, Cambridge (fig. 27).^[85] This vessel must be dated to ca. 1380, when it is recorded as entering the collections of the college. It consists of a silver-gilt mounted wooden drinking cup, or mazer, with a central, crenelated turret, crowned by the eponymous swan.^[86] The conceit is highly reminiscent of Villard de Honnecourt’s chantepleure. In this case, it is clear that the bird’s position is fixed and that the illusion of its drinking the cup’s contents is only notional. The tower, like the central features of the Vinkovci and Chungul cups, is fitted with a concentric siphon that, when the cup is filled to just below the swan’s beak, empties its contents into the lap of the over-eager drinker.^[87] Interestingly, this cup was originally fitted with a lid, which in the sixteenth century was converted into a second cup and given away by the college to one of its benefactors. Its present whereabouts are unknown.^[88]

27

The Swan Mazer, ca. 1380, maple wood with gilded silver, 7 × 13 cm diameter. Cambridge, Corpus Christi College

Despite Villard’s drawing of a cup with a siphon, the longstanding scholarly consensus has been that such mechanical toys were not produced in the medieval West prior to the turn of the fourteenth century. Elaborate automata were, of course, well-known features of court ceremonial in Constantinople and Baghdad, and were even emulated in Mongol Karakorum.^[89] In Western Europe, however, no automata seem to have been known or conserved from antiquity into the Middle Ages, and no evidence has hitherto come to light for the actual production of such devices prior to the end of the thirteenth century.^[90] In terms of the transmission of the Alexandrian tradition of fine technology, the current consensus of textual historians is that no complete Latin translation of Hero’s Pneumatica was in circulation in Western Europe prior to the fifteenth century.^[91] On the other hand, Western romance literature drew extensively on automata as motifs in their narratives, often as a means of enhancing the exoticism of the settings.^[92] In her study of medieval automata, Elly Truitt places the rise of Western European manufacture of— as opposed to fantasy about—automata at the very end of the thirteenth century, with the furnishing of the park at Hesdin by Robert II, Count of Artois (1250–1302) with elaborate fountains and other hydraulic tricks.^[93] The table fountain preserved in the Cleveland Museum of Art, with its miniature water-wheels and bells, represents a small-scale version (ca. 1320–1340) of this phenomenon of playful engineering, the popularity of which is attested in French and Burgundian court inventories of the fourteenth and fifteenth centuries.^[94] With the Italian Renaissance, of course, came both renewed copying and translation of Hero of Alexandria’s works and the application of their principles in gardens, theaters, and other courtly entertainments.^[95]

The discovery of the siphon mechanism hidden in the Chungul cup obviously complicates this narrative. There is also another, tantalizing mention of a hydraulic automaton in the medieval West about a century before Villard’s drawing of a self-draining cup. An analogous vessel appears in the Vita of Charles the Good, Count of Flanders, composed by Galbert of Bruges. Just days prior to his murder in March 1117, Charles paid Lombard merchants at Ypres 21 marks of silver for a wine jug (kanna in the Latin text, pot a vin in the Old French version) which, when filled, emptied itself by a hidden mechanism, to the wonder of onlookers.^[96] This is obviously not the same vessel as the cup from the Chungul Kurgan, nor can we tell at this point whether it was of local manufacture, produced in northern Italy, or created in some more distant center before being brought to Flanders. One might nevertheless suggest that the presence of this self-emptying wine jug in the treasury of the Counts of Flanders might have helped to inspire a silversmith later in the twelfth century to undertake the fabrication of a cup with the same siphon mechanism. The trick wine jug—like the gold covered cup of Count Charles the Good mentioned above—achieved uncharacteristic prominence in Galbert’s life of the unfortunate noble, as the clergy of Bruges absconded with both vessels in the wake of Charles’s assassination.^[97] Several months later, the cup and jug would be returned to their rightful owner, Charles’s successor as Count of Flanders, William Clito. Presumably, these were in turn passed down along with the count’s other properties to William’s successor, Thierry of Alsace, Thierry’s daughter, Margaret of Flanders, and Margaret’s son, Baldwin IX, who would eventually become Latin Emperor of Constantinople as Baldwin I.

The Spoils of Adrianople?

It has been suggested that the fourth-century Vinkovci cup was deposited in the panic and confusion of 378 CE, when the native son of Cibalae, the Eastern Roman Emperor Valens, was killed on the field of battle at Adrianople.^[98] The Goths, having risen up against the authority of the ruler in Constantinople, rampaged erratically around Thrace for months afterward until being brought under control by the new emperor, Theodosius I. A second violent confrontation of the steppe and the settled peoples occurred in the same location centuries later, during the 1205 siege of Adrianople by the Crusader forces led by Baldwin of Flanders. The light cavalry of Qipčaq (Cuman) mercenaries allied to the Bulgarian leader Kalojan routed the heavily armed, but slow-moving knights of the Crusaders. The French knight Robert de Clari gives a description of the disaster:

When they came to this city, they laid siege to it, and while they were encamped there, behold one day John the Vlach [i.e., Kalojan], he and the Comans, with a very great force, came into the land of Constantinople, as they had done before, and found the emperor with all his host encamped before Adrianople. When they of the host saw these Comans clothed in their sheepskins, they had no more fear or care for them than for a troop of children. And these Comans and this horde came at a great pace and they rushed upon the French and slew many of them and defeated them all in this battle. And the emperor was lost, so that it was never known what became of him, and Count Louis [of Blois] and many other high men, and so many others that we do not know the number of them, but fully three hundred knights were lost there.^[99]

Robert de Clari presents two mysteries: first, the Qipčaqs, whose ways of war seem to have taken the Crusader knights completely by surprise, and second, the fate of the emperor Baldwin. About the latter, we are better informed than Robert, as we know from an exchange of letters between Kalojan and Pope Innocent III that Baldwin had died in Kalojan’s custody by about September of 1205.^[100]

What became of the belongings of Baldwin and his vassals who were captured or killed at Adrianople? Could the Qipčaqs have carried them back to their core lands in the Black Sea steppe?^[101] While there are dozens of one-off finds of Western medieval metalwork in the territories of Ukraine and Russia, the concentration of finds in the Chungul Kurgan burial is extraordinary. Even though we have not been able to pinpoint a precise locus of manufacture for the Chungul cup, we can at the least sketch out a probable origin within a zone stretching from Picardy in the west to Lower Saxony in the east and encompassing Flanders, the Low Countries, and the middle Rhine. There is also a coherence within the finds themselves that suggests a common origin. A silver-gilt belt buckle with a siren (fig. 28), for instance, bears a resemblance to various works of German, Mosan, and northern French metalwork to which we have already compared the ornament of the cup.^[102] More closely than any of these works, however, the buckle resembles the workmanship of the cup itself, with its use of cast silver for an animal figure, the extensive employment of the ring punch to create textures and details, and the cupped leaf forms of the siren’s palmette tail. These details all suggest that the two works originated from a common center. Similarly, among the trappings of the five horses buried alongside the notable in the Chungul Kurgan there was found a partly gilded and engraved silver knob from an otherwise lost object.^[103] Its windblown acanthus design, with leaves facing alternately to the right and to the left, echoes the pattern found on the lips of the cup and lid of the Coupe de Charlemagne at St. Maurice d’Agaune.^[104] Still more intriguingly, attached to the bridle of one of the horses was a tablet of rock crystal (4.6 × 5 × 1.5 cm) drilled through crosswise (fig. 29).^[105] This object can be identified as the center piece of a rock crystal cross of a type similar to the cross from Scheldewindeke that survives intact in Brussels, adorned with filigree and champlevé enamels (fig. 30).^[106] This fragment of what must have been an impressive and valuable cross raises a question: from what sort of enemy might the Qipčaqs have looted such an object? Might it not have been an army of Crusaders? Evidence linking the 1205 Battle of Adrianople to the finds from the Chungul Kurgan is, of course, entirely circumstantial, but it would explain both the richness of the finds and the inclusion among them of works such as the siphon cup that bespeak patronage at the highest levels. Furthermore, the origins of Emperor Baldwin and many of his high nobles in the counties of Flanders, Hainault, and Naumur and in contiguous areas of northern France would explain the concentration of artifacts that can be associated with this region.^[107]

28

Belt buckle with a siren excavated from the Chungul Kurgan, ca. 1200, gilded silver, 2.1 × 6.7 cm. Kyiv, Treasury of the National Museum of History of Ukraine

29

Rock crystal tablet excavated from the Chungul Kurgan, late 12th century, rock crystal, 4.6 × 5.0 × 1.4 cm, diameter of perforations 1.1 cm. Kyiv, Treasury of the National Museum of History of Ukraine

30

Rock crystal cross from Scheldewindeke, ca. 1175–1200, rock crystal on wooden armature with gilded copper, champlevé enamel, and stones, 42 × 38.5 cm. Brussels, Musées Royaux d’Art et d’Histoire

Just as cups of precious metal were an accoutrement of rulers in medieval Western Europe, so they were also an attribute of leadership in the nomadic world. The shared vessel, passed around from the leader to the members of his entourage, was an important symbolic marker of princely status.^[108] The size of the covered cup from the Chungul burial marks it as a vessel meant for sharing, whereas the smaller cup found in the grave—the one with enameled decoration— which was probably created in a Byzantine workshop in the twelfth century, is characterized as a personal drinking vessel for a single individual by the presence of the ring handle.^[109] Archaeological evidence from the time of the Avars indicates that cups often came in two distinct forms, one for personal, the other for communal use.^[110] The nature of the trick concealed in the Chungul cup naturally implies shared drinking; the joke would be rather pointless if the cup were used exclusively by its owner. While we cannot be sure that the Qipcaq buried with the covered cup knew the secret of its hidden siphon, other evidence points to a nomadic appreciation for trick vessels. The Mala Pereshchepina hoard, for example, associated with the seventh-century Bulgarian Khan Kubrat, contained silver and gold goblets with rattles inserted in their feet.^[111] From the thirteenth century, the so-called Coupe de St. Sigismond at St. Maurice d’Agaune in Switzerland, attributed to a workshop operating in the Mongol Empire, has a “singing” device concealed in the spherical knob of its lid.^[112] In the end, however, the individuals who placed the potion of charred herbs in the Chungul cup at the time of the leader’s burial may not have been at all aware of the siphon mechanism, and the vegetal matter would likely have stopped it up and prevented it from working as intended.

Returning to the unfortunate Baldwin of Flanders, the thirteenth-century Byzantine historian George Akropolites gives a succinct account of his ultimate fate: “They say that after [Kalojan] killed Baldwin, his head served as a goblet for the barbarian, after it had been cleaned of all its contents and decorated all round with ornament.”^[113] Whether or not this account is literally true— one can be forgiven for hoping that it is not—it doubtless was written to echo the fate of the Byzantine emperor Nikephoros I after his defeat by an earlier Bulgarian leader, Khan Krum, at the Battle of Pliska in the year 811.^[114] However gruesome, it would be a poetic twist if Baldwin’s ultimate end was as a luxurious drinking cup, while his own cup wound up in the grave of a nomadic leader responsible for his disastrous defeat at Adrianople.^[115] All conjecture aside, the Chungul Kurgan has now yielded up new information not just for the history of luxury metalwork in the late twelfth century, but for the history of technology and its use in courtly settings of the medieval West.