In : N. Crummy (ed.), Image, Craft and the Classical World.

Essays in honour of Donald Bailey and Catherine Johns

(Monogr. Instrumentum 29), Montagnac 2005, p. 285-289.

28
Now we see through a glass, darkly
by David Buckton1

One of the many enigmatic objects in the British The pendant is important in that it could constitute the
Museum is a gold and enamel pendant with a complicated only proof of the survival of filigree enamelling into the
image and the inscription ‘seal of God’ in Greek (Fig. 1). Late Antique or Early Byzantine period. Although the
It has been in the museum since 19112, and this its first continuation of the ancient Hellenistic technique has long
publication is dedicated to two greatly valued friends and been taken for granted, the only two pieces of evidence to
former colleagues. have been published, filigree enamels in Paris and
It comprises a disk of gold sheet, 34 mm in diameter, Baltimore, Maryland (Wessel 1967, nos 1-2; Haseloff
with a strip of gold hard-soldered to it as a suspension- 1990, Abb. 1-2), have recently been re-dated from the 5th
loop. Both the disk and the loop are decorated with gold century CE to between 1892 and 18973. An initial
approach to the British Museum Research Laboratory
filigree and granulation. The disk has a border containing
the inscription CFRAGIC QEOV in filigree wire letters,
brought the reassuring response that the vitreous material
on the ‘seal of God’ pendant was absolutely typical of
the words separated by a five and a six-pointed star and a
Late Antique or Early Byzantine enamel.
lunate shape, with granules inside the wire outlines, a
circle containing three evenly spaced chevrons, all in Enamel is, of course, glass. When it is heated to its
wire, and a filigree motif with granules raised on wire melting-point, glass bonds to metal, and for some three
and a half thousand years this property has been exploited
loops. The letters are separated by granules arranged in
to add polychrome decoration to precious and base metal.
threes; groups of three are also spaced inside the outer
According to ‘Theophilus’, a Benedictine monk and priest
edge of the border.
writing under a pseudonym, probably in north-west
Inside the border is a human head, with the hair or Germany at the beginning of the 12th century, the enamel-
headgear encroaching on the brow and both cheeks. There lers of his day used coloured glass tesserae from mosaics
is no mouth; short lengths of wire run into the face from found ‘in the ancient buildings of the pagans’ (Dodwell
the chin. On the top of the head, slightly to the left of 1986, 44, Liber II, cap. XII)4. This surprising assertion has
centre, is a triple aigret, and to the right is an asymmetric now been substantiated by the discovery that most early
scroll of doubled wire. Projecting from or from behind the medieval enamel – whether Byzantine or western – was
head at top left and centre left and right are animal-heads. produced using glass which had been manufactured in the
Below the human head are the head, body and tail of an Roman world before the 4th century CE. The 4th century
animal facing left. In the surrounding field granules are witnessed a change in glass constituents, which has made
grouped in threes; in one case, seven granules make up a it possible to establish that in the post-Roman period,
rosette. In places are remains of a dark reddish brown even as late as the 13th century, enamellers were using
vitreous laminate. The back of the object is plain. pre-4th-century Roman glass5. If the 4th century, after

1 Rose Cottage, Ashfield Road, Elmswell, Bury St Edmunds, Suffolk IP30 9HG.
2 P&E 1911,5-12,2. Bought from L. Blumenreich, 83 Dartmouth Road, Brondesbury, London N.
3 Bibliothèque Nationale, Cabinet des Médailles, M.1688, and Walters Art Gallery, 44.304. These filigree enamel medallions have exactly the same
dimensions, stylistic features and technical idiosyncrasies and are patently the work of one goldsmith. Enamel on the Baltimore object (acquired
in 1931) has been shown to date from no earlier than the 17th century (Henderson 1992). The Paris example, acquired by the Cabinet des Médailles
in 1897, incorporates a portrait of Licinia Eudoxia, daughter of Theodosius II (408-450) and wife of Valentinian III (425-455), which can only
have been copied from a coin in the British Museum, published, with an illustration, in 1892 (Buckton forthcoming).
4 Theophilus goes on to mention ‘various small vessels’ of coloured glass, apparently also to be found in ruined Roman buildings, which could be re-
used in glassworking.
5 The analyses and their interpretation have been published a number of times and progressively refined (See Freestone 1993a; 1993b, 37-45;
Freestone et al. forthcoming).

— 285 —
 28 - D. Buckton

Fig. 1 — Gold filigree enamel pendant, British Museum, P&E 1911,5-12,2, and detail. Diameter 34 mm. Drawing: James Thorn; © The British Museum 2005.

Christians had been granted the freedom to practise their 1895, 123-4)6. On one side is a stylized variation on the
religion throughout the Roman Empire, also saw the first iconography of the haematite intaglios; the other side,
overtly Christian art, earlier mosaics could well have been which Schlumberger designated the reverse, corresponds
supposed pagan. to the filigree enamel pendant. A comparison of one with
Its inscription relates the pendant to 4th-century the other nevertheless reveals significant discrepancies.
magical gems inscribed CFRAGIC QEOV on the reverse. Perhaps the most obvious is that there is no ‘seal of God’
On the obverse, these haematite intaglios show a inscription on the lead amulet, which has in its place a
horseman, often identified by inscription as Solomon, largely indecipherable legend7. The layout of the image
spearing and trampling a prostrate figure (Fig. 2). differs: on the BM pendant it completely fills the avail-
According to the Testamentum Solomonis, the archangel able space, whereas on the Schlumberger amulet it occu-
Michael gave Solomon a ‘seal of God’ which endowed pies the upper three-quarters only. Judging from the gaps
him with magical powers, particularly over demons
evident in the inscriptions on the lead amulet, the space at
(Michel 2001, 268).
the bottom of the image is likely to have contained, or
If the iconography on the intaglios is compared with
that on the pendant, it is just possible to interpret the
human head surmounting the head, body and tail of an
animal on the latter as a horseman. There are, though, far
more differences than there are similarities: there is no
room on the pendant for the horse’s legs, let alone a pros-
trate figure under its hooves, and the animal heads projec-
ting from the human head have no parallels on the inta-
glios.
However, several amulets from the eastern Mediter-
ranean region, most of them invoking Solomon as protec-
tion against ill-health, ill-fortune and ill-will, were publi-
shed by Gustave Schlumberger (1892), and one of these,
a lead pendant (Fig. 3), bears an image very close to that Fig. 2 — Haematite intaglio, British Museum, EA 56204. Drawing: James T.
on the British Museum object (Schlumberger 1892, 79; Farrant; © The British Museum 2005.

6 Schlumberger collection, acquired in Constantinople, n° 5.

7 Schlumberger (1892) found the inscription illegible, but subsequently (1895) identified the nouns ‘ evlafo" (deer) and, less confidently, laywv"

— 286 —
 Now we see through a glass, darkly

Fig. 3 — Lead pendant, formerly Schlumberger collection (illustration after Schlumberger).

been intended for, at least the legs of the animal, the head, dence which, although in itself unsurprising in the context
body and tail of which survived to be drawn by the illus- of Solomon amulets, inevitably raises the question of
trator. The losses could have been the result of defective whether the goldsmith could have been copying not the
casting or die-stamping, or of subsequent damage, wear or objects themselves but their illustrations, conveniently
corrosion. collected in a single publication.
At the top right of the enamelled version is an arrange- If a pendant acquired by the British Museum in 1911
ment of filigree which, while it approximates to the indi- had indeed been inspired by drawings published in 1892,
vidual lines of the analogous feature on the lead amulet, these years would obviously provide the date-bracket for
unlike the latter cannot be recognized as the head of a its manufacture. Between these very dates, it was sug-
stag. The animal-heads seem to be identified by names gested in an article published in 1988, more than 150
which can be made out from occasionally transposed enamels purporting to be Byzantine had in reality been
Greek letters on the obverse of the lead object: ONOC made by craftsmen ‘moonlighting’ from the Fabergé or
(ass), ELAFOC (deer), AMNOC (lamb), and KUNI- other St Petersburg workshops (Buckton 1988). The
KOC, interpreted as kûnivsko" (young dog) by terminus post quem, 1892, was provided by the publica-
Schlumberger (1895, 124). This last could also be tion of a work on Byzantine enamel which did not
emended to kuvniklo" (rabbit), but any corresponding mention the enamels in question and contained colour-
representation certainly looks more kûnikov" (dog-like). plates apparently serving as models for them (Kondakov
On the gold pendant the animal-heads are not identi- 18929). The terminus ante, 1911, was the year the proble-
fied, and, since any clues to the meaning or function of the matic enamels were catalogued.
two pendants are supplied by the lead one, it seems likely In 1995 it was revealed that a confession written in
that this, or a uniformly flawed duplicate, provided the 1916 had come to light in the Fabergé archives (Norman
model for the gold version. If this were indeed the case, 1995). A craftsman employed by the firm had admitted
the ‘seal of God’ inscription might well have been taken being involved with others in the manufacture of over a
from another object. It does, in fact, appear in one of the hundred of the enamels. They had indeed been made in St
other line-drawings in Schlumberger’s article8, a coinci- Petersburg – actually between 1892 and 190910.

(hare).
8 A haematite intaglio acquired in Beirut, n° 13: Schlumberger 1892, 84; Schlumberger 1895, 129.
9 Kondakov’s book was also published as Geschichte und Denkmäler des byzantinischen Emails (Byzantinische Zellen-Emails: Sammlung A. W.
Swenigorodskoi) and as Histoire et monuments des émaux byzantins (Émaux byzantins: collection Zvénigorodskoï), both Frankfurt am Main, 1892.

— 287 —
 28 - D. Buckton

While the St Petersburg forgeries are cloisonné The re-use of glass manufactured in the Roman world
enamels and have little in common with the filigree is the pendant’s sole saving grace. The only way a late
enamel of the pendant, they do show that Byzantine 19th or early 20th-century goldsmith could have known
enamels were being counterfeited in the last decade of the that early medieval enamellers used Roman glass was,
19th century and the first decade of the 20th, and that the however indirectly, from Theophilus13. His treatise is
fakes were based on published illustrations. The re-dating divided into three books, the first largely concerned with
of the medallions in Paris and Baltimore from the 5th painting, the second with glass, and the third with metal-
century to the last decade of the 19th suggests that the work. The reference to mosaic tesserae comes in Book II,
technique of filigree enamel was also being employed by while Book III contains step-by-step instructions for
forgers. Back in the BM Research Laboratory, the gold
making cloisonné enamel. Before the 19th century,
pendant was subjected to further scientific tests, and,
editions and commentaries had concentrated on Book I 14.
whereas the glasses used for the enamelling were
Increasingly complete editions followed in 1843 (in
confirmed as compatible with a Late Antique or Early
Byzantine date, the gold turned out to be suspiciously French), in 1847 (in English) and in 1874 (in German),
pure, around 99.1%, with 0.8% copper and 0.1% silver11. which proved fruitful sources for writers specifically on
This contrasts with the alloys found in Early Byzantine the subject of enamel (Dodwell 1986, liv–lvii,
jewellery, which are typically between 87% and 97% gold lxxiv–lxxvii). Notable among these, in the early years of
(with a mean of 92%) and with a silver content at least the 20th century, was H. H. Cunynghame, who explicitly
twice that of the copper (Oddy & La Niece 1986). promoted the treatise as the manual of a practising
The St Petersburg forgeries are also notable for their craftsman, in contrast to the usual medieval compilations
extremely high-carat gold, but the glasses are those which of untried recipes and other hand-me-downs15.
were readily available in the Fabergé and other work- However, the first to bring Theophilus directly to the
shops. In contrast, the goldsmith responsible for making attention of latter-day jewellers and metalsmiths was
the pendant seems to have used Roman glass to enamel it. almost certainly Charles de Linas. In a review ‘pour le
If this occurred in the early middle ages, the re-use of the commerce’ of the historical sections of trade exhibitions
glass was absolutely characteristic of the period but it is mounted in Europe in 1880, he supplied a French transla-
hard to account for the unusual composition of the gold tion of the entire chapter about the re-use of ancient glass,
alloy employed. If, on the other hand, the pendant was even helpfully glossing the word ‘pagans’ as ‘Romans’16.
made between 1892 and 1911, it is equally difficult to In the context of the colours of enamel on medieval works
explain the use of glass manufactured one and a half in the exhibitions, he went on to emphasize how at the
millennia earlier.
time of ‘le moine artiste’ Theophilus, ‘and doubtless
It has to be said that misgivings far outnumber any before that’, enamelling was done with glass taken from
positive feelings. The pendant imperfectly reproduces an Roman mosaics (de Linas 1881, 117).
image on a published object, even apparently compensa-
ting for losses from that object, its inscription appears on This information was therefore accessible by the time
a different object illustrated in the same publication, it the filigree enamel medallions in Paris and Baltimore and
made its appearance at the end of a period notorious for the St Petersburg cloisonné enamel forgeries were made.
enamel forgeries, the composition of its gold is unlike the Uniquely, however, the goldsmith responsible for the fili-
alloys normally found in Late Antique and Early Byzan- gree enamel pendant in the British Museum not only had
tine jewellery, and no other filigree enamel is now attri- the information – whether from de Linas or another
buted to Late Antiquity or Early Byzantium12. Since pen- source – but acted on it. A significant proportion of the 90-
dants tend to swivel on their cords or chains, it is, more- odd-year delay in publishing the object is down to his or
over, usual for their backs to have some sort of decoration. her enterprise.

10 The confession was published in Faberzhe et al. 1997, 341-9. For a summary in English, see Buckton 2001.
11 Department of Scientific Research, British Museum, report dated 7 October 1996: the glass is a low-potash, low-magnesia soda-lime-silica variety.
Its high chlorine content (1.2 wt.% Cl) endorses the view that it was made using soda from a natural mineral or plant-ash source rather than a
synthetic alkali.
12 See note 3. Interestingly, the Fabergé craftsman who confessed to forging cloisonné enamels (note 10) was a filigranshchik by trade.
13 No other medieval treatise includes the information.
14 Primarily because it contains references to oil as a medium for painting, refuting Vasari’s assertion that painting in oil had been pioneered by the
van Eyck brothers.
15 While Cunynghame (On the Theory and Practice of Art-Enamelling upon Metals, London, 1901, and European Enamels, London, 1906) repeats
Theophilus’ detailed instructions for making a cloisonné enamel, he does not identify the source of the glass. (For the view that Theophilus had
himself never practised enamelling, see Buckton 1994.)

— 288 —
 Now we see through a glass, darkly

Bibliography Freestone, I.C., Bowman, S.G.E. & Stapleton, C.P., forth-

coming. ‘The composition and origins of Byzantine and
early medieval enamel glasses’, in D. Buckton,
Botkin, M.P. 1911. Collection Botkine (St Petersburg). Catalogue of medieval enamels in the British Museum,
Buckton, D. 1988. ‘Bogus Byzantine enamels in I: early medieval and Byzantine enamel (London).
Baltimore and Washington, DC’, Journal of the Walters Haseloff, G. 1990. Email im frühen Mittelalter.
Art Gallery 46, 11–24. Frühchristliche Kunst von der Spätantike bis zu den
Buckton, D. 1994. ‘Theophilus and enamel’, in D. Karolingern (Marburg).
Buckton & T.A. Heslop (eds), Studies in medieval art Henderson, J. 1992. ‘A scientific analysis of the enamel
and architecture, presented to Peter Lasko, 1–13 decorating a gold medallion in the Walters Art Gallery’,
(Stroud/London/Dover, New Hampshire). Journal of the Walters Art Gallery 49/50, 27–31.
Buckton, D. 2001. ‘Stalin and Georgian enamels’, in A. Kondakov, N.P. 1892. Istoriia i pamiatniki vizantiiskoi
Eastmond (ed.), Eastern approaches to Byzantium, emali (Vizantiiskie emali: sobranie A. V.
211–218 (Aldershot). Zvenigorodskogo) (St Petersburg).
Buckton, D. forthcoming. ‘“Early Byzantine” enamel in de Linas, C. 1881. Émaillerie, métallurgie, toreutique,
France’, Festschrift for Christopher Walter. céramique: les expositions rétrospectives à Bruxelles,
Dodwell, C.R. 1986. Theophilus, the various arts, De Düsseldorf et Paris en 1880 (Paris–Arras).
diversis artibus (Oxford). Michel, S., 2001. Die magischen Gemmen im Britischen
Faberzhe, T.F., Goryna A.S. & Skurlov, V.V. 1997. Museum, I (London).
Faberzhe i peterburgskie iuveliry: sbornik memuarov, Norman, G. 1995. ‘Master forger’s legacy goes on sale’,
statei, arkhivnykh dokumentov po istorii russkogo iuve- The Independent, 15 December 1995, 7.
lirnogo iskusstva, posviashchaetsia 150-letiiu so dnia Oddy, A. & La Niece, S. 1986. ‘Byzantine gold coins and
rozhdeniia Karla Faberzhe 1846–1996 (St Petersburg). jewellery: a study of gold contents’, Gold Bulletin 19/1,
Freestone, I.C. 1993a. ‘Theophilus and the composition of 19–27.
medieval glass’ in P.B. Vandiver, J.R. Druzik, G.S. Schlumberger, G. 1892. ‘Amulettes byzantins anciens,
Wheeler & I.C. Freestone (eds), Materials Research destinés à combattre les maléfices et maladies’, Revue
Society symposium proceedings (Materials issues in art des Études Grecques 5 (1892), 72-93; revised in
and archaeology III) 267, 739-46 (Pittsburgh). Schlumberger 1895, 118-140.
Freestone, I.C. 1993b. ‘Compositions and origins of Schlumberger, G. 1895. ‘Amulettes byzantines anciennes,
glasses from Romanesque champlevé enamels’, in N. destinées à combattre les maléfices et maladies’ in
Stratford, Catalogue of medieval enamels in the British Mélanges d’archéologie byzantine (Paris).
Museum, II: northern Romanesque enamel, 37-43 Wessel, K. 1967. Die byzantinische Emailkunst vom 5. bis
(London). 13. Jahrhundert (Recklinghausen).

— 289 —