Cupid or Adonis? A New Roman Relief Carving from Lincoln

NOTES HIGH-PRECISION RADIOCARBON DATING THE Twelfth International Radiocarbon Conference in Trondheim during July 1985 was given remarkable sets of high-precision data calibrating the radiocarbon (I4C) time-scale against the tree-ring sequences at 10- or 20-year intervals through the past 7150 years. This was the result of work, independently, of M. Stuiver's laboratory in Seattle (4500 years) and of G. W. Pearson and his team in Belfast (7150 years). The agreement between these long arrays of data is impressive, especially because different 14C counting procedures are used in the two laboratories, and in Seattle tree-ring material was provided from C. W. Ferguson's Douglas Fir and Californian Bristlecone Pine sequences, whereas the Belfast work used the long Irish Oak sequences worked out by Baillie and Pilcher, supplemented by the German oak series (J. R. Pilcher, M. G. L. Baillie, B. Schmidt and B. Becker, Nature, cccxii (1984), 150-2). The fluctuations of this high-precision calibration {Radiocarbon, Calibration Volume (1986)) establish on a 10- to 20-year interval basis the reality of the 'wiggles' first propounded by Hans Suess from the 1960s onwards {Radiocarbon, xx (1978), 1-18), and show their applicability through the northern hemisphere. We can therefore be confident that in the 14C data we are measuring globally (northern hemisphere) applicable constants of Nature. This has given an internationally acceptable radiocarbon time-scale through the past 7150 years. With this high-precision 14C time-scale, archaeology (and indeed all the relevant timesciences) have been given an exact chronometric tool. It is now up to them to see that sample material becomes available that has the required stringent credentials to make optimal use of this impressive laboratory research, to yield exact calendrical dates. The purpose of this note is to show one example of how this has been done. The high-precision 14C calibration data for the Christian era (from Seattle) have in fact been available for some years (M. Stuiver, Radiocarbon, xxiv (1982), 1-26), but archaeologists have seemed reluctant to devise dating research programmes to make effective use of it. One laudable attempt to apply it to a medieval problem was perhaps addressed to too complex a site—relating the burnt layers of medieval Bergen to its recordedfires—withtoo little data (K. Krzywinski andS. Gulliicsen, The Bryggen Papers, Suppl. Ser. i (1985), 40-51). But medievalists should not be in the least dissuaded: the procedure could be invaluable (E. M. Jope, Radiocarbon, xxviii (1986), in press). It must first be emphasized that this high-precision calibration data can only be meaningfully applied if the 14C estimations are made on sample material having appropriately stringent (and definable) credentials, for both context and characteristics of sample. We must accept the basic principle that it is only possible to derive a sharp and true calendrical date from a 14C estimation if the carbon in the sample has itself a sharp 'point' date—that is, a definable short time-range (10 to 20 years, corresponding with the intervals of the calibration) during which the carbon was alive, part of the biosphere. Much material submitted for 14C estimation is too heterogeneous, and taken from too wide a time-range, to be susceptible of exact dating. No amount of statistical manipulation will yield sharp calendrical dates of meaningful accuracy out of data from samples (however numerous) that do not meet this stringent requirement. This must apply, for instance, to many powdered charcoal samples. Optimal use of the high-precision calibration data can best be made with sample material that preserves considerable arrays of definable annual-growth increments: that is, in practice, fairly substantial pieces of timber (charred or otherwise) with a good recognizable growth-ring NOTES 359 pattern. This is of prime importance around the time-ranges of complex inflexions in the calibration diagrams, because with serial 14C data on contiguous 10- or 20-year groups of treerings, wiggle-matching can often resolve ambiguities (G. W. Pearson el al., Radiocarbon, Calibration Volume (1986)). There is a further important requirement. The groups of tree-rings must be selected from as near to the final outer growth of the tree as possible (or in some definable relation to it) if the 14C date is to be relatable to the time of felling of the tree. Allowance for sapwood also still involves some uncertainty, and these matters affect the use of dendrochronology as much as of H C dating (M. Baillie, Tree Ring Dating and Archaeology (1982), 53-68). Even then, especially with medieval building timbers, the cut timber may have been left for seasoning for a few years. Other sample material not susceptible of tree-ring dating (notably bone of younger or small animals) can also sometimes be made to serve for high-precision 14C dating, at least in timeranges without inflexion complexes in the calibration diagrams. High-precision 14C determination at present needs fairly large samples (200-400 g.). It would be of very great advantage if the funding could be available to develop small-sample highprecision counting, so as to be able to work with samples of the order of 10-20 g. 2550 2600 2450 cot BP 2350 2250 2150 MAIDEN CASTLE, BICKtRTON .CHESHIRE 2500 3.3B 2400 Z O Q < 2300 20 YEAR GOBBITS TIMBERS FROM 2200 RAMPART 2100 EXCAVATOR: J.J.TAYLOR 14 C : G.W.PEARSON I 2000 600 500 I 400 300 200 col BC FIG. 1. High-precision 14C estimations on sample 2 (duplicate estimation on whole process) and sample 3. <7=±i6 years 360 THE ANTIQUARIES JOURNAL With these considerations in mind (and others, see E. M. Jope, Radiocarbon, xxviii (1986), for observations on charcoal sorting, hearths, etc.), archaeologists should therefore attempt as far as possible to use this high-precision method to apply calendrical datings to structures. This is especially important with sample material to which direct tree-ring dating is not so easily applied. The example below demonstrates that it can be done. The defences of the hillfort Maiden Castle, Bickerton, Cheshire, are here chosen to show what can be achieved with four high-precision 14C determinations on two timbers, to yield calendrical datings for such a structure. These defences were originally excavated by W. J. Varley, showing two lines of wall defences with horizontal massive oak timbers (Liverpool Annals Arch. Anthrop. xxii (1935), 97-110; xxiii (1936), 101-12). Recently our Fellow Dr J. J. Taylor of Liverpool University judged that such timbers might give an excellent opportunity for close dating of these defences, and accordingly did a supplementary excavation to get sample material for highprecision 14C estimation by Dr G. W. Pearson in Belfast. Two timbers were selected: sample 2, a section of small scantling (about 20 rings), probably alder; sample 3, a section of oak, with rings towards the outer limits of tree growth, and therefore near the date of felling of the tree. The calibrated 14C datings should therefore give estimates close to the building date of this hillfort wall. These dates are: (2) 2322 and 2285 B.P. (± i6)=395 and 390 B.C., and (3) 2411 and 2410 B.P. (±i6)=4O5 B.C. (fig. 1); the ±16 covers the whole process, including sample preparation, and on so steep a part of the calibration, the precision in terms of calendar years will be considerably sharper. The calendar dating for sample 2, 390-395 B.C., falls on this very steep part of the calibration diagram, and the two replicate determinations thus give very close agreement in calendar date. This sample (2) of small scantling probably represents virtually its final growth years, and thus c. 390 B.C. must give a close estimate for the building date of this wall. The value for the oak sample, 3, however, cuts the top of this long, steep part of the calibration diagram at 405 B.C., but it runs right back through the 100 years of inflexion complex, and a dating range of 405 to 510 B.C. is thus possible. However, bearing in mind the structural relation of this timber to sample 2, with its close dating of 405 B.C., we may conclude that the dating of sample 3, with its range 405-510 B.C., is not incompatible with that of 2, and supports the dating c. 390 B.C. for the building of this hillfort wall. This illustrates the great value of the high-precision calibration curve, and allows the conclusion that this hillfort wall at Maiden Castle, Bickerton, was built within a few years of 390 B.C. A research programme of dating on hillfort rampart timbers and other suitably preserved structural timbers of the Iron Age, which could be based, as we have seen, on a comparatively small number of high-precision 14C determinations, would undoubtedly cost moneys but the results could be of such value to Iron Age studies as to be well worth the expenditure. E. M. JOPE CUPID OR ADONIS? A NEW ROMAN RELIEF CARVING FROM LINCOLN NEW discoveries of Roman sculpture in Lincolnshire regularly seem to produce surprises. The sculptors who worked in the area appear to have had a talent for unusual treatments of Mediterranean and native subjects, even if they did not always follow the canons of GraecoRoman proportions. Recent examples include the Stragglethorpe Rider God and the relief of three seated goddesses found in Lincoln in 1980.' In 1985, in excavations at Hungate in the city, another unusual and interesting relief carving was found in the destruction debris of a late Roman building.2 It is, as might be expected, of Lincolnshire limestone, and is 33-5 cm. wide, 35 cm. high as it survives, and 95 cm. from front to back. It is carved in high relief with two figures sitting on a NOTES 361 couch: a nude male with wings, and a female wearing a tunic, a mantle and a bonnet, and holding something in front of her (pi. LXI). The top of the stone is broken at both corners and in the middle, but behind each figure's head there remains a part of the dressed surface, sloping slightly upwards towards the centre of the stone, showing that the top was in the form of a low gable. Parts of the incised groove denning the border of the pediment survive to the right of each figure's head. There is a slight upward curve at the outer ends of the dressed surfaces, indicating that the gable had a raised shoulder at each corner, now broken, representing the acroteria of a pedimental roof. Several of the tombstones from Lincoln have similar, though steeper, gabled tops with vestigial acroteria of this kind.3 To the right of the female figure there is the baluster leg, the arm-rest and part of the cushioned seat of the couch. Above them, the right-hand margin of the stone is bordered by three bands defined by V-profiled incised grooves. Since the central band is wider than the others, they probably do not represent a fluted pilaster, but the reeded upright of the back of the couch. The lower left corner of the stone is damaged, but traces remain of the couch seat and arm-rest beside the male figure. No corresponding vertical bands are visible on this side, but there is little available space for them. Both ends are similarly dressed, with a chisel, and there is no particular reason to suppose any cutting back at the time of reuse which would have removed the border. The back of the stone is dressed with a punch, and there are marks of the same tool on the underside, which is, however, flat and smooth and appears to have been exposed to weathering, presumably the result of reuse of the stone as building material. In view of its symmetry, in the presence of the arm-rests on each side and the remains of the gabled top, we may consider that the relief is complete as we have it, minor damage apart; that is, there is no reason to suppose that it formed part of a larger composition. There is no indication of date, apart perhaps from the slight resemblance of the male figure's hairstyle to that of the Emperor Trajan, but the sculpture is likely to have been carved some time between the early second and mid third centuries. At first sight, there would seem to be little doubt in identifying the nude winged malefigureas Cupid or Amor. Although his pot belly and rounded cheeks are more characteristic of the infant than of the adolescent Cupid, he is represented as being of similar size to his apparently adult companion, towards whom he turns. His head is tilted back, and his hair is marked by incised grooves which run back from the forehead, curving over the ears and diverging from the crown as they approach the back of the head. Although the right leg is damaged, the outline of the shin and the foot is visible; the leg is bent back, implying the motion of his turn towards his companion. The female figure, however, sits straight and upright on the couch, looking directly ahead. Her right arm is outstretched towards her companion, and she cups his chin between her thumb and fingers in a gesture suggesting erotic tenderness.4 He responds by touching her wrist with the fingers of his right hand, while in his left hand he clutches her elbow. Her rather commanding intimacy shows that she is no mere mortal, but a goddess or, at least, a mythological character. She is dressed in a tunic with two other garments over it. The tunic is sleeveless, but it covers her right shoulder and part of the upper arm. Its folds are rendered as closely set, rather angular, channels which are splayed at the bottom where her toes peep out from beneath the hem. A mantle is draped over her left shoulder and across her breast, down to the waist on her right side, where the folds curve up again to pass behind her right shoulder. The lower part of the mantle appears to cover her lap, falling between her knees to a tassel at its pointed end. Her knees are set wide apart, and are covered by drapery of apparently heavier cloth than the tunic beneath, extending to just above the ankles. Alternatively, this may be understood as the lower part of the mantle, and the tasselled pendent triangle as a towel or apron laid over it.s On her head she wears 362 THE ANTIQUARIES JOURNAL a bonnet or diadem with a vertical front, broader above the forehead than at the sides, the top of which is rendered as a smooth continuous curve, cut with a claw chisel. With her left hand she holds a large object on her lap. Its top is carved as a series of protuberances, arranged in the form of a rosette. The front is slightly convex, cut with several vertical grooves, and curved at the bottom. The sides are plain. It appears to be some form of container. Who is she? The costume is that of the north-western provinces of the Empire, apart from the headdress, which does not seem to belong to the everyday dress of women in the region.6 Mother Goddesses, however, are dressed in this fashion, and often hold a basket or tray of fruit; but as this Lincoln deity is not one of a trio, she cannot be understood simply as a Dea Mater.7 Nor is she a single Dea Nutrix, of the type best known from pipeclay figurines, since she suckles no infant. The presence of Cupid beside her suggests that she is a Roman deity, albeit in provincial dress. Cupid is frequently represented in amorous association with Psyche. According to Apuleius' version of the myth,8 Venus, jealously trying to protect her divine profession from Psyche's rival, though unintended, attractions (which had already included an illicit liaison with Cupid), set her various tasks, the last of which was the retrieval from Hades of a box which contained Beauty. Contrary to instructions, Psyche opened the box on her way back, and it released, not Beauty, but Sleep. It was only through Cupid's intervention at a high Olympian level that she was saved from the narcotic effect, and reunited with him. The female figure on the Lincoln relief might seem to stare ahead with the appearance of stupefaction; but the normal representation of the myth was to show Psyche and Cupid in mutual embrace, the best-known example being the statue of the pair from the house at Ostia which is identified by their names.9 Moreover, the Lincoln deity does not have the butterfly wings with which Psyche is often, though not invariably, represented,10 nor does the object which she holds look much like a beauty casket. Nor is Psyche shown wearing matronal dress or such headgear as the Lincoln goddess wears. So there is little reason to suppose that it is Psyche who is represented here. The personage with whom Cupid is most commonly associated is Venus, his mother. If, however, it is she who is to be identified here, there are some iconographic problems in relation to the manner of their representation together. The natural affection between them is the subject of a number of Hellenistic renderings, in marble, in metalwork and on vase paintings, which show them in fond embrace.11 In all of them, Venus/Aphrodite is sitting and bending forwards, or crouching, and Cupid, shown as a young boy, has his arms around her neck and is usually standing between her knees. The Lincoln relief, however, is not at all like this. Its best parallels, in fact, are in representations of Venus, not with Cupid, but with her beloved Adonis, where the two are shown seated side by side on a couch or double throne, or in adjoining chairs. The context is that of Adonis' death and annual resurrection from Hades, to be reunited with Venus for six months before returning to the realm of Proserpina for the remainder of the year.12 On the wall-paintings from the villa at Boscoreale, Adonis' posture is very similar to that of the Lincoln Cupid, but in reverse, since he sits on Venus' left side; they do not embrace, however.13 They are shown, sitting together and embracing, on Hellenistic and Etruscan bronze mirrors, and on a sardonyx cameo now in Paris.14 The closest comparisons for the Lincoln relief, though, are to be found on some late second- and early third-century Roman sarcophagi which have as their main subject the death of Adonis, attacked by a wild boar. Several of them have, at the left-hand end, a representation of Venus embracing Adonis. They are seated, half-turned towards one another, and Venus clasps Adonis' cheek or chin in her right hand.15 She sits to the left, he to the right, so the image is the reverse of the Lincoln relief, as in the Boscoreale painting; and she does not look straight ahead, but is turned towards him. Nevertheless, these versions of Venus and Adonis seem to be the closest for comparison with the Lincoln relief, particularly in the gesture of her clasping his chin. PLATE LXI T H E ANTIQUARIES JOURNAL Photograph: Trust for Lincolnshire Archaeology Roman relief carving from Lincoln (height 35 cm.) NOTES 363 There are some obstacles to accepting this as the true interpretation, but also some explanations which may remove them. First, the female on the Lincoln relief holds an object which seems to be a container, and Venus, in the examples just mentioned, does not. A possible solution is suggested by other representations associated with the cult of Adonis, which show the Adonis garden. This took the form of a basket or a vase, often the bottom or inverted top of a broken vessel, in which such fast-growing green plants as lettuce or fennel were planted. They were placed on the roof-tops of Greek houses for the annual festival of Aphrodite and Adonis, the sprouting greenery being the spring-time growth which Adonis' resurrection symbolized.16 Eros might appropriately lend a hand, as he does on a Red Figure lekythos from Ruvo, where he is shown passing the house-plant up to a woman standing on a ladder.17 It may be suggested that it was an Adonis garden which the Lincoln sculptor showed Venus to be holding in her lap. The male figure has so far been described as Cupid, and Adonis is not conventionally represented with wings. This anomaly may be resolved by suggesting that the sculptor has made a conflation between representations of Venus with the chubby infant Cupid, and those of Venus with the slim, youthful but fully grown Adonis,18 and that he has portrayed the malefigureas adult in size, but has retained the rounded belly and wings of the boy Cupid. A third problem of interpretation is that of the Lincoln female's frontal pose. In the comparanda which have been mentioned above, Venus may appear rather more restrained than Psyche, but she is not so aloof from the object of her love as the Lincoln goddess seems to be. To this, two answers may be suggested. One is a deduction from an objective observation: Celtic Mother Goddesses are normally represented in this manner, seated and looking straight ahead, so the sculptor may merely have followed that conventional model. The other is a subjective interpretation: that Venus, as Mother or Mistress, is looking out to the world, and proclaiming her pride in the devotion to the power and immortality of love which is personified next to her. The myths, both of Cupid and Psyche, and of Venus and Adonis, stood as symbols in Roman funerary art of the afterlife of the soul.19 Here, we have a carving which could be interpreted as representing either myth, though the discussion above favours the view of it as being Venus and a composite Cupid/Adonis. To its significance in terms of classical mythology must be added that of fertility, which the Celtic Mother Goddesses represented: the fruitfulness of animals and plants and here also, perhaps, that of men and women, achieved through the Love which binds them. There is no knowing where in Lincoln this carving was originally set up. Despite the possibility of a direct funerary significance, it could equally well have graced a shrine within the city, whether public or domestic. Thefindspotis close to the centre of the lower colonia, where a temple might well have stood, and the size of the carving is very similar to that of the two reliefs from Lincoln which show three seated goddesses,20 suggesting that its original purpose was of a similar nature. Since the relief carries no inscription, however, there is no evident reason for assuming that it had such a specific religious purpose. Indeed, there are many parallels elsewhere in the Empire for the more generalized use of such scenes in the wall-paintings and reliefs which decorated private houses,21 and also their gardens, of which Venus was protector.22 To us, this piece of sculpture may seem to have several potential layers of meaning. Given the pleasure in ambiguity which is characteristic of the art of pre-Roman Britain,23 and the astonishingly mature approach to myth which can be seen in quite a number of Romano-British works of art,24 it would be a mistake to assume that there was no subtlety in the sculptor's intentions. In his portrayal of Venus as Mother Goddess with one who might be seen as both her son Cupid and her lover Adonis, we may feel that he was condensing the rich mythological repertoire at his disposal, and successfully unifying the great themes of love, fertility and immortality. THOMAS BLAGG and MARTIN HENIG PLATE LXII THE ANTIQUARIES JOURNAL hxcmpkr tjt_£rimoredditHm dextrorfim. iienmiritifpmani pviri^gnt prrnij m The Glastonbury Abbey memorial plate (from H. Spelman, Concilia, decreta, leges..., i (1639)) 364 THE ANTIQUARIES JOURNAL THE GLASTONBURY ABBEY MEMORIAL PLATE RECONSIDERED THIS relatively little-known relic of Glastonbury's medieval antiquarian tradition has fortunately been preserved, not merely by printed transcripts, but by prints taken from it and then off-set to reverse the mirror image of the original print. No explicit reference to the plate or its inscription is made by either Adam of Domerham (late thirteenth century), John of Glastonbury (after 1342 and before 1375-90, when the oldest manuscript was written) or William of Wyche (1497), nor by the earliest antiquaries, William Worcestre and John Leland. The text seems first to have been published by Francis Godwin in his Catalogue of the Bishops of England (1615) (p. 11) and, with the impressions referred to above, by Henry Spelman in his Concilia, decreta, leges, constitutiones in re ecclesiarum orbis Britannicis, i (1639), pp. 7-10 (see pi. LXII here). The plate was made available to him by Sir Thomas Hughes and the prints referred to made from it. In his description he comments on the mode of fastening, that it was not ancient, but describes the column from which it had been taken as 'columna vetustissima'. So far as is known this is the first example of the use of a brass inscription as a printing plate to multiply copies for study or publication. This second publication, is, therefore, of no little interest to students of monumental brasses. Later in the century a very free translation was published by Aylett Sammes in his Britannia antiqua illustrata (1676), pp. 212-13. It appears to have been ignored in the next century. Hearne, for example, in his editions of Adam of Domerham and John of Glastonbury's chronicles, did not include it among the other odds and ends of Glastonburialia used to make up the books. During the last hundred years or so it has been noted again, partially translated and discussed by Robert Willis in The Architectural History of Glastonbury Abbey (1866), pp. 19-20; and, more seriously, by Dean Armitage Robinson in Somerset and Dorset Notes and Queries, xvii (1921-3), pp. 58-62, and in his Two Glastonbury Legends (1926), pp. 56, 59-60. In the latter he included a greatly reduced illustration of the plate. The site From the information given by the dean, F. Bligh Bond excavated the area indicated for the site and discovered a substantial foundation 7 ft. 6 in. (2-28 m.) in diameter and over 5 ft. (1-5 m.) deep and published his work in Proceedings of the Somerset Archaeological and Natural History Society, lxxii (1926), pp. 16 ff. and pi. v. His interpretation, however, was vitiated by his uncritical acceptance of the plate and legends as his starting point. Dr Ralegh Radford, F.S.A., who visited the site at the time, tells me that his foundation: 'consisted of a solid perimeter with a soft centre. Most of the architectural fragments came from the soft centre, perhaps all.' The presence of the fourteenth-century mouldings was noted by Bond with some perplexity, since he believed he was excavating a sixth-century foundation. Similar fragments have been found elsewhere on the site and may well have been buried when the column was overthrown and removed at the Reformation. The plate Spelman's prints show the inscription to have been cut on two pieces of brass, an octagonal upper part with a separate base. The inscribed area was 237 mm. high, the octagon 158 mm. and the base 138 mm. wide. It was fixed by rivets through eleven projecting bosses set round the edge, a style of fastening not observed on any other monumental brass either in this country or abroad. The style of lettering, a mixed script with rather poorly formed characters using NOTES 365 Lombardic uncials and Gothic minuscules, cannot be paralleled on any surviving English brasses. From these pointers we may conclude that it was produced locally, perhaps in one of the abbey's workshops, and that the use of two pieces in such a small plate suggests that it may have been made using scrap or older fragments from monuments. The text Most of the inscription follows fairly closely the wording of a lection for the Memorial of St Joseph of Arimathea prefixed to some copies of John of Glastonbury's chronicle. This was based on the traditions of the abbey about the early history of the 'Old Church', St David's visit and subsequent gift of the 'Great Sapphire' altar as recounted in the thirteenth-century recension of William of Malmesbury's De antiquitate Glastonie ecclesie and by John himself. The text of the De antiquitate presents problems, since it includes information not given by him in other books surviving in older copies than those of his De antiquitate. The main features of the foundation legend, the building of the church by disciples of the Lord and its first dominical consecration, are all referred to in a charter of Henry II attested between 2nd and 16th December 1184 (Dom Aelred Watkin, The Great Cartulary of Glastonbury Abbey, i (1947), pp. 189-8 (Somerset Record Society, 59). The chronicle entries are, however, very differently worded, give more detail and are silent about the building of either 'Pyramid' or 'Column'. In the following analysis these abbreviations have been used: JG James P. Carley, The Chronicle of Glastonbury Abbey. AnEdition, Translation and Study of John of Glastonbury's Cronica sive antiquitates Glastoniensis ecclesie (Woodbridge, 1985). L Lection, as in JG. P Plate from the impressions in Spelman, op. cit. WM John Scott, The Early History of Glastonbury. An Edition, Translation and Study of William of Malmesbury's De antiquitate ecclesie (Woodbridge, 1981). L: ANNO POST PASSIONEM DOMINI TRICESIMO PRIMO DUODECIM ex discipulis Sancti Philippi apostoli EX QUIBUS JOSEPH AB ARIMATHIA PRIMUS ERAT in terram istam UENERUNT, qui regi Aruirago renuenti Christianitatem optulerent. This follows very closely the account given, with discreet qualifications, by WM, p. 145. P: Anno post passionem do/mini .xxxi0. duodecim sancti ex quibus / Joseph ab arimathia primus erat hue / uenerunt, L: Tamen locum istam cum duodecim hidis terre ab eo impetrauerunt, in quo virgis torquatis muros perficientes PRIMAM HUIS REGNICONSTRUXERUNT ecclesiam, QUAM CHRISTUS IN HONORE SUE MATRIS ET LOCUM ad sepulturam seruorum PRESENCIALITER DEDICAUIT. This seems to follow the version in JG, especially in the reference to the manner of the construction, rather than WM. P: Qui ecclesiam huius regni primam in hoc / loco construxerunt, quam Christus in honore sue matris & locum pro / eorum sepultura presencialiter dedicauit. N.B. The lection omits some of the information in the chronicle at this point. 366 THE ANTIQUARIES JOURNAL L: Postea venit SANCTUS DAVID MENEUENCIUM ARCHIEPISCOPUS predicatam ecclesiam adhuc sub forma priori reparatam dedicare disponens, CUI DOMINUS IN SOMPNIS APPARUIT ET EUM A PROPOSITIO REUOCAUIT; NECNON IN SIGNUM QUOD DOMINUS IPSE ECCLESIAM IPSAM PRIUS CUM eius CIMITERIO DEDICAUERAT, MANUM EPISCOPI DIGITO PERFORAUIT ET SIC PERFORATA MULTIS UIDENTIBUS IN CRASTINO APPARUIT. This follows the accounts in WM and JG fairly closely, omitting some details. A pilgrim badge, probably commemorating this story, has been published, but can no longer be found (Proc. Som. Arch. Soc. liv (1909), pp. 125-6). The abbreviated legend should be read, as Dr Radford has suggested in communicating this to me: 'M(anus) [a] C(hristo) V(ulnerata)'. P: Sancto dauid mene/uencium archiepiscopo hoc restante. Cui dominus ecclesiam illam dedica/re disponente in sompnis apparuit & eum a proposito reuo/cauit, necnon in signum quod ipse dominus ecclesiam ipsam prius cum / cimiterio dedicarat: manum episcopi digito perforauit, & sic per/forata multis uidentibus in crastino apparuit. L: Mox idem episcopus, diuina reuelacione ammonitus, aliam minoremque capellam in modum CANCELLI IN PARTE ORIENTALI HUIC ECCLESIE ADIECIT ET IN HONORE VIRGINIS Marie CONSECRAUIT. P: Postea vero idem / episcopus domino revelante ac sanctorum numero in eadem crescente: quemdam / cancellum in orientali parte huic ecclesie adiecit & in honore / beate Virginis consecravit. This is similar to JG, but he does not specify the position of the church so precisely. From this point the two texts begin to diverge quite considerably in the choice of the wording. L: In CUIUS rei memoriam preciosissimum ALTARE suum huic loco contulit. In both WM and JG the history of this relic, the Great Sapphire, is given at greater length, although William does not mention it in his account of St David in the Gesta Regum, Rolls Series 90, cap. 25. It has been studied by Dr Barb in the Journal of the Warburg and Courtauld Institutes, xix (1956), pp. 40-67, where he suggested that it might be one of the sources of the legend of the Holy Grail. P: Cuius altare inestimabili / saphiro in perpetuam huius rei memoriam insignivit. / L: Et ut semper nosceretur ubi capelle iste coniungebantur, quedam piramis in parte septemtrionali exterius et quidam gradus interius et meridies linialiter eas abscindit. P: Et ne locus aut quantitas prioris ecclesie / per tales augmentations obli/uioni tradetur: erigitur / hec columpna in linea per / duos orientales angulos / eiusdem ecclesie versus meridiem pro/tracta & predictum cancellum ab ea abstin/dente. Et erat eius longitudo ab ilia linea versus oc/cidentem .lx. pedum, latitudo vero eius .xxvi. pedum / distancia centri istius columpne a puncto me/dio inter predictos angulos .xlviii. pedum. The text of the plate can be translated thus: The 31st year after the Passion of the Lord twelve saints, among whom Joseph of Arimathea was the first, came here. They built in this place that church, the first in this realm, which Christ in honour of his Mother, and the place for their burial, presently dedicated. St David, archbishop of Menevia [i.e., Wales], rested here. To whom the Lord (when he was disposed to dedicate that church), appeared in sleep and recalled him from his purpose, also in token that the same Lord hadfirstdedicated that church with the cemetery: He pierced the bishop's hand with his finger, and thus pierced it appeared in the sight of many on the morrow. NOTES 367 Afterwards indeed the same bishop as the Lord revealed, and the number of the saints in the same grew: added a chancel to the eastern part of this church and consecrated it in honour of the Blessed Virgin. The altar whereof, of priceless sapphire, he marked the perpetual memory of these things. And lest the site or size of the earlier church should come to be forgotten because of such additions, he erected this column on a line drawn southwards through the two eastern angles of the same church, and cutting it off from the aforesaid chancel. And its length was 60 feet westward from that line, its breadth was truly 26 feet; the distance from the centre of this pillar from the midpoint between the aforesaid angles, 48 feet. Conclusions In the mass of legends which grew up at Glastonbury there seem to be a few certainties: that there was an ancient wooden church whose dimensions correspond with those of other early Celtic churches (C. A. R. Radford, 'Glastonbury Abbey before 1184: interim report of the excavations, 1908-64' in Medieval Art and Architecture at Wells and Glastonbury, British Arch. Assoc. Conference Transactions (1981), pp. 112-13); and that new details were being added to the foundation legend throughout the Middle Ages. The lection, with its elaboration in the memorial plate, seems to represent the position about the end of the fourteenth century. The main elements are all present in John of Glastonbury, who wrote towards the mid fourteenth century, and in the thirteenth-century recension of William of Malmesbury, but there is no mention of the later accretions. Clearly patterned on the lection for the Memorial of St Joseph of Arimathea, it is possible that this was used in a similar way for a Memorial of the Oldest Church. It is possible, I think, to define the date when the text was written and the plate set up more closely. Carley has shown that John of Glastonbury wrote under abbot Walter de Monnington (1342-75) and that the earliest copy was written under his successor John Chinnock (1375— 1420). During his long abbacy Chinnock actively promoted the cult of St Joseph at Glastonbury. The Magna Tabula, the original manuscript of which is in the Bodleian Library (MS Lat. hist, a 2), was set up by him in the church after 1382 to provide a potted history for visitors. He was also instrumental in getting the sub-apostolic foundation of his church recognized, not only by other Benedictine houses in England, which gave the primacy to him and his successors, but also by the Councils of Pisa, Constance and Basle for the precedency of the English bishops; although this was more hotly contested. Since there is no real evidence for dating the foundation on the basis of the mouldings used by Bond as evidence of a 'later' remodelling, we cannot conclude that the plaque and column were contemporary. The mixed script indicates that the plate cannot be earlier than Chinnock's reign, while the style suggests a date after, rather than before, 1400. JOHN A. GOODALL NOTES 1 T. Ambrose and M. Henig, 'A new Roman rider-relief from Stragglethorpe, Lincolnshire', Britannia,,xi (1980), 135-8; T. F. C. Blagg, 'A Roman relief-carving of three female figures, found at Lincoln', Antiq.J. lxii (1982), 125-6. See also M. Henig, 'Graeco-Roman art and Romano-British imagination', J.B.A.A. cxxxviii(1985), 1-22, fora general discussion of originality in both style and content in Romano-British art. 2 Archaeology in Lincolnshire 1984-1985 (First Annual Report of the Trust for Lincolnshire Archaeology, 1985), 44-6 and 49-50; we are grateful to M. J. Jones, F.S.A., Director (City of Lincoin), Trust for Lincolnshire Archaeology, for 368 THE ANTIQUARIES JOURNAL inviting us to exhibit the carving at a Ballot of the Society and for permission to publish it here. 3 R.I.B. 255, 256, 257, 262, 268. 4 cf. the gesture of Venus towards Mars on a wallpainting in Naples: R. Stuveras, Le Putto dans I'art romain, Coll. Latomus, xcix (Brussels, 1968), 200 p. 352; illustrated in Ferri, op. cit. (note 10), 114, fig. 101. 16 Lehmann, op. cit. (note 12), 47 and 125-8, for discussion and references to the ancient sources and modern literature on the subject. 17 Furtwangler-Reichhold, Griechische Vasenand pi. LXXIV, a. malerei, ii (Munich, 1909), 98-9 and pi. LXXVIII; 5 We are grateful to Dr J. P. Wild, F.S.A., for his Lehmann, op. cit. (note 12), 126-7 and fig- 67. 18 Adonis might appear as a slim youth or, in comments on the costume; cf. p. 405 in his 'The clothing of Britannia, Gallia Belgica and Germania Roman representations, as a 'strapping, fullyInferior', Aufstieg und Niedergang der'romischen developed if beardless figure': Lehmann, op. cit. (note 12), 39; cf. Ovid, Metamorphoses, x.522-3: Welt, ii.12.3 (1985), 362-422. 6 The headgear is not like the close-fitting bonnet 'nuper erat genitus, modo formosissimus infans, / worn by some Treveran women, for example: J. P. iam iuvenis, iam vir, iam se formosior ipso est.' 19 Most of the representations of Greek myths Wild, 'Clothing in the north-western provinces of the Roman Empire', Banner Jahrb. clxviii (1968), discussed by Toynbee (op. cit. in note 15) are on tombs and gravestones. From Britain, note the 166-240, pp. 198-9 and fig. 20. 7 J. M. C. Toynbee, Art in Britain under the fragmentary relief from Chester, showing the death of Adonis: R. P. Wright and I. A. Richmond, Romans (Oxford, 1964), 171-4. 8 Catalogue of the Roman Inscribed and Sculptured Apuleius, Metamorphoses, iv.28-vi.24. 9 R. Calza and E. Nash, Ostia (Florence, 1959), Stones in the Grosvenor Museum, Chester (Chester, fig. 44. There is part of the torso of a similar group 1955), no. 140. 20 We are grateful to Hugh Thompson, F.S.A., from Woodchester: G. Clarke, 'The Roman villa at and Catherine Johns, F.S.A., respectively, for Woodchester', Britannia, xiii (1982), 197-228. 10 e.g. S. Ferri, Arte romana sul Danubio (Milan, these observations. It was also noted, on exhibit at 1933)) 235 and 245, fig. 298; S. Reinach, Repertoire ballot, that the carving appeared to best effect at a des reliefs grecs et romains (Paris, 1912), ii, 471, and distance of 20-30 ft. (6-9 m.). For the other Lincoln reliefs: Guide to the Antiquities ofRoman Britain iii, 294-5. 11 Lexicon Iconographicum Mythologiae Classicae, (British Museum, 1964), 55 and pi. xix; T. F. C. ii (Zurich and Munich, 1984), nos. 1241, 1242, Blagg. op. cit. in note 1. 21 Examples from the Empire as a whole are too 1244, 1245. 12 For the sources, see P. W. Lehmann, Roman numerous to require citation. Among Romano-BriWall-Paintings from Boscoreale in the Metropolitan tish wall-paintings, Cupid is figured in the baths of Museum of Art (Cambridge, Mass., 1953), 46, nn. the villa at Southwell, Nottinghamshire, and, possibly with Venus, in a house at Kingscote, 68 and 69. 13 Gloucestershire: N. Davey and R. Ling, WallIbid., 38-49 and pis. IV and v. 14 Lexicon Iconographicum Mythologiae Classicae, Painting in Roman Britain, Britannia monograph 3 (London, 1982), 155-7 and 79, and 121. M.H. i (Zurich and Munich, 1981), 224-5, n o s - I2> J5> believes that the subject of this latter fresco might 19a. 15 C. Robert, DieantikenSarkophag-Reliefs, iii. 1: be Venus and Adonis with Cupid as an Einzelmythen(Rome, i969),Tafn.nno. 9,111 no. 12 intermediary. 22 Varro, De re rustica, i.i.6. and v no. 20. See also Lehmann, op. cit. (note 12), 23 N . K . Sandars, Prehistoric Art in Europe (Har43-5. They are represented in a similar pose on the pediment of a funerary monument from Flavia mondsworth, 1968), 275-84. 24 Henig, op. cit. in note 1. Solva in Noricum: J. M. C. Toynbee, 'Greek myth in Roman stone', Latomus, xxxvi (1977), 343-412,