An Anglo-Saxon cruciform Brooch from Lyminge

AN ANGLO-SAXON CRUCIFORM BROOCH FROM L YMINGE B.M. AGER An Anglo-Saxon cruciform brooch was found by Mr W. Laing of Lyminge while digging in his garden in 1982 (N.G.R. TR 162403). The brooch was taken by him to the Royal Museum, Canterbury, for identification. The find-spot is some 400 yds. south from the site of the discovery during railway excavations in 1885 of Anglo-Saxon burial remains and is almost three-quarters of a mile from the main Anglo-Saxon cemetery to the north of the village (Meaney 1964, 127). It has not been ascertained whether the brooch was lost or was part of a grave group. The brooch, of which only the head-plate and bow survive (Fig. la) is of copper alloy and measures 5.3 cm. long by 2. 78 cm. at its widest. It is cast all in one piece with the three half-round, moulded knobs of the head-plate. On the flat back of this plate are two integral, pierced lugs holding the iron spindle for the now corroded spring of the pin. The bow, which is long, straight-sided and highly arched, is cast slightly off-centre to the head-plate, which is remarkably thick (0.4 cm. maximum). The brooch has, unfortunately, been broken in the past across the junction of the bow and the foot-plate, on the back of which would have been the catch-plate for the pin. Since no exact parallel for the extant portion of the brooch is known, the form of the foot remains conjectural, although the parallels noted below suggest the possibility that it had a triangular terminal rather than the more usual moulded, horse-like head. The ornament consists of moulding of the knobs (which are non-functional), of triangular notches in either side of the ends of the bow, enclosed between a pair of transverse grooves, and of a broad, median furrow, bounded by a groove either side, along the ridge of the sloping-sided bow. After casting, it appears that the brooch was finished with a file, which has left fine, close, transverse striations 59 B.M. AGER a b Fig. 1. a, Anglo-Saxon cruciform Brooch from Lyminge; b, Bow Brooch from Faversham; c, cruciform Brooch from Brooke, Norfolk. (Scale: ½) 60 ANGLO-SAXON BROOCH FROM LYM INGE under much of the bow and, possibly, the lengthwise grooves on the left-hand side of the top of it. This is only the second brooch of the cruciform class to be found at Lyminge. The first, discovered during the railway works already mentioned, also lacks any close context.' Until the new find, a total of twenty-three cruciform brooches was known from Kent east of the Medway ,2 all but one of early types assignable to groups I and II of Aberg's classification.' They form a somewhat mixed collection, only one identical pair being known (from Bifrons, grave 15), and the new Lyminge brooch has no parallel within the County. The brooch is difficult to classify satisfactorily, because the footplate is missing. However, the integral, flat-backed knobs round the head-plate would place it either in Aberg's group II or III. Among the English examples of these groups, the small, flat, square headplate is very unusual and is then generally associated with plain, flat knobs (more typical of the small long brooch class), as on the brooch from Little Wilbraham, Cambs. ,4 and on another from Brooke, Norfolk (Fig. lc).5 Only five examples of related form with knobs of half-round section are known to me, including the other cruciform from Lyminge, but the head-plate of the latter, compared with that of the new find, is flatter, more oblong and decorated with six ring-anddot stamps, while the bow is short, low and thin. A slightly closer parallel, though still with a more oblong head-plate and shorter, broad bow is the brooch from Kempston, Beds.6 But the closest formal similarity as regards the 'head' is with a matching pair from Empingham, Rutland,7 although even here there are obvious differences in respect of the relative length of the bow and size of the knobs. Foreign parallels are also rather scarce, but the distinctive small, flat, square head-plate is quite well represented among the typologically earlier brooches with full-round knobs from Scandinavia and Schleswig, e.g. the brooch with zoomorphic terminal from a grave at 1 J. Reichstein, Die kreuzformige Fibel (Offa-Biicher 34, Neumiinster, 1975), no. 816, Taf. 101,5. ' S.C. Hawkes and M. Pollard, 'The gold Bracteates from sixth-century AngloSaxon Graves in Kent, in the Light of a new Find from Finglesham', Fruhmittelalterliche Studien, 15 (Berlin, 1981), 322-324. 3 N. Aberg, The Anglo-Saxons in England (Uppsala, 1926), 36--42. • Op. cit. in note 1, no. 790, Taf. 98,4. 5 British Museum, reg. no. 70, 11-5, 6. 6 Op. cit. in note 1, no. 763, Taf. 68,10. 1 Ibid., no. 848, Taf. 69, 1-2. 61 B.M. AGER Maglarp, Skane, Sweden/ associated with a handled urn of the early Migration Period. A square plate with long, narrow bow and integral, though flat, knobs (as on the Little Wilbraham and Brooke brooches) appears on examples from Nordre Fevang, Vestfold, Norway, and from Romdrup, Alborg Arnt, Denmark.9 Both the former example, together with the Kempston brooch and the pair from Empingham, are included in Reichstein's Foldvik-Empingham type-group (though perhaps Nordre Fevang - Empingham would be more apt, since the head-plates of the Foldvik brooches have a distinctive, raised central panel and lateral wings which set them apart from the rest). A final comparison can be made with the brooch from Blomsgarden, Vastergotland, Sweden,10 which is very like the Lyminge brooch in its head-plate and knobs, but has a centrally expanded bow. Both this and the Little Wilbraham brooch mentioned above are included by Reichstein in his Bradwell-on-Sea type-group, though they lack the lateral wings to the head-plate of the Bradwell brooch. In all the above cases where the knobs are half-round and integral, except for the other Lyminge brooch and one more from Barrington B, Cambs., 11 which appears to have a flat head-plate, the foot-plates have some form of triangular or shovel-like terminal. It may be, therefore, that the new Lyminge brooch was similarly made, although in view of the great variety of form within the cruciform class, it would be unwise to draw too firm a conclusion. Some general comments can be made on the rather limited decoration of our brooch. The notch-and-groove motif on either end of the bow is quite common on cruciform brooches in England as well as northern Germany and Scandinavia. It is ultimately derived from northern Gallo-Roman metalwork of the latter fourth and earlier fifth centuries, on which it is often found across the base of buckle-tongues and on strap-attachments and belt-stiffeners, e.g. on buckles from Bremen-Mahndorf and Aalden, or on various fittings from Liebenau, Krefeld-Gellep and Perlberg.12 The adoption of the pattern by Germanic craftsmen underlines the extraordinary degree to which they were dependent on provincial-Roman ornament for a 8 B. Stjernquist, 'Chronologische Probleme der Volkerwanderungszeit in Sildschweden', in G. Kossack and J. Reichstein (eds.), Archiiologische Beitriige zur Chronologie der Volkerwanderungszeit (Antiquitas, Reihe 3, Serie in 4to, Bonn, 1977), Abb. 2, I. 9 Op. cil. in note 1, nos. 28 and 495, Taf. 24, 2 and 125;15. 10 Ibid., Taf. 98, 6. 11 Ibid., Taf. 89, 8. 12 H. W. Bohme, Germanische Grabfunde des 4. bis 5. Jahrhunderts zwischen unterer Elbe und Loire (Munich, 1974), Taf. 15, 10; 28, 6; 33, 15; 58, 5; and 79, 26-8. 62 ANGLO-SAXON BROOCH FROM L YMINGE wide range of ideas and techniques. It went on to .enjoy a prolonged existence, occurring for example on a sixth-century brooch of Aberg's group IV from Barton Seagrave, Northants. 13 The broad furrow-between-grooves ornament along the ridge of the bow is much rarer, but is still a feature that links the English brooches with their foreign counterparts, though this time apparently over a more restricted area from Schleswig to Norway, e.g. the brooches from Lyminge, Kent; Trumpington, Cambs.; Nebstrup, Randers Arnt, Denmark; Skogen, Vestfold, Norway, 14 and Borgstedt, Schleswig. 15 In all cases, except possibly the last, this ornament is combined with the notch-and-groove motif. As regards features of construction, the double support for the spindle (more usually only single) occurs on several of the Kentish cruciforms, e.g. from Howletts, grave 2, Lyminge and Faversham. But it is by no means confined to Kent, nor to this particular class. One method of winding the spring round the spindle is illustrated by Salin. 16 It is important, when considering dating, to note that double lugs are associated with short catch-plates reaching from the end of the bow to the upper end of the terminal, since catches extending the full length of the foot are a sign of very early date." At present, the chronology of English cruciforms is dependent on the North German/Norwegian typology presented by Dr. Reichstein,18 since their earliest development can be traced there and since only the latest of the English forms are associated with stylistically datable material. Connections have been noted above between the Lyminge brooch and others belonging to Reichstein's types Bradwell-on-Sea and Foldvik-Empingham, the former having a terminal date in the late fifth century. The latter type belongs to a group (no. 5), some examples of which extend into the early sixth century, 19 though as remarked above, there must be doubts as to whether the type is homogeneous. 20 Certainly, on typological 13 British Museum, reg. no. 91, 3-19, 16. See op. cit. in note 3, fig. 77, though the detail is insufficient to show the motif. 14 Op. cit. in note 1, Taf. 101, 5; 116, 9; 144, 8; 116, 6-7. 15 H. Kuhn, Die germanischen Bugelfibeln der Volkerwanderungszeit (Graz, 1981), III, Taf. 11, 58. 16 B. Salin, Die altgermanische Thierornamentik (Stockholm, 1904), Abb. 39. 17 H. Shetelig, 'The cruciform Brooches of Norway', Bergens Museums Arbog, 8 (Bergen, 1906), 57. 18 Op. cit. in note 1. 19 Ibid., 67-68. 20 Similar doubts about the coherence of Reichstein's classification, which make it difficult to apply to new finds particularly, have been expressed in reviews by T.M. Dickinson, 'Post-Roman and pagan Anglo-Saxon', Arch. Joum., cxxxv (1978), 336-7; 63 B.M. AGER grounds, the Lyminge brooch must relate to a very early stage in the development of either of these types. Both the flat, thick square head-plate, only slightly wider than the bow, and the long, narrow, paralleled-sided, high bow of thick, flat-based section are features of early cruciforms, though not the earliest, in Norway and other regions.21 On the other hand, the side-knobs are cast in one piece with the brooch. The chronological significance of this development from brooches with detachable side-knobs is difficult to pin down, and it is as well to emphasise Shetelig's crucial observation that many different forms could be in use contemporaneously. In Norway, the change was probably already under way in the earlier half of the fifth century and half-round knobs were quite common in the latter half, though head-plates were becoming extended at this time too.n In England, brooches of the Midlum type from Spong Hill, Norfolk, with separately cast side-knobs, can justifiably be dated to the middle years of this century ,n from which it can be inferred that brooches of the same type with integral knobs probably belong more to the later fifth. In respect of the wide head-plate, however, the Midlum form is typologically later than Lyminge. Attention has been drawn to a brooch from Mucking, Essex, grave 987,2' which is related in form to the flat-knobbed Little Wilbraham parallel noted above, but, as observed by Dr Welch,2i it fits uneasily the suggested date of deposition 'probably to the earliest decades' of the fifth century. In conclusion, the Lyminge brooch represents a very early variety of Aberg's group II and seems unlikely to be much later, if at all, than the composite version of the Midlum type, which might be assigned provisionally to the middle years of the fifth century. Connections between the Anglo-Saxon cruciforms and Scandinavian brooches, particularly of Jutland and neighbouring areas and the probability of the development of the former from the latter, have been remarked on by several scholars.2 • I am indebted to my and C. Hills, 'Barred zoomorphic combs of the migration period', in V.I. Evison (Ed.), Angles, Saxons and Jutes: Essays presented to J.N.L. Myres (Oxford, 1981), 106-8. 21 Op. cit. in note 1, 70; and in note 17, 21, 54-5. 22 Op. cit. in note 17, 124, 153. lJ C. Hills, op. cit. in note 20, 108. "V.I. Evison, 'Distribution Maps and England in the first two Phases', in V.I. Evison (Ed.) op. cit. in note 20, Fig. 4b. Zl M. Welch, review of ibid. in Med. Arch., xxvi (1982), 228-9. 26 Op. cit. in note 17, 27, 99-102, 110-112. Also E.T. Leeds, 'Denmark and early England', Antiq. Journ., xxvi (1946), 35; op. cit. in note 2, 322-4; and S.C. Hawkes, 'Anglo-Saxon Kent c. 425-725', in P.E. Leach (Ed.), Archaeology in Kent to AD 1500 (CBA Research Report 48, 1982), 70. The reservations expressed by Prof. V.I. Evison in op. cit. in note 24, 138, were written before publication of the last but one reference. 64 ANGLO-SAXON BROOCH FROM L YMINGE colleague Mr D. Kidd for pointing out to me a further link between early Kentish bow-brooches and Scandinavian types. Reichstein has observed a close parallel for a cruciform brooch from Faversham with a similar brooch from Oxb0l, south-west Jutland/' the former having a flat, discoid projection on top of the bow and the latter a more ovoid flattening. The disc can be seen, too, on another Faversham brooch (Fig. lb ),2'' with openwork, zoomorphic head-plate, and is a common feature on fifth-century bow-brooches from Gotland. 2'' It is therefore particularly interesting to note that the treatment of the 'horse-head' terminal of the other cruciform from Lyminge has a Jutlandic connection:)(' and that the brooch as a whole, as noted by Mrs. S. Hawkes is remarkably similar to one from Nebstrup, Randers Arnt, J utland.-11 Regarding the new find specifically, it has already been noted that the broad furrow-between-grooves of the bow connects it with the Schleswig/south Scandinavia region. Also, the long bow perhaps relates it rather to Danish forms, on which it is a regular feature, than to those of other areas,32 though this does not rule out northern Germany. On the other hand, the distributions abroad of Reichstein's types Foldvik-Empingham and Bradwell-onSea, to which the brooch appears to have the closest affinities, are both wholly Scandinavian.33 On balance, therefore, it seems safe to conclude that the brooch was made by a craftsman familiar with the styles of jewellery current in southern Scandinavia during the early Migration Period and, in view of its date, it may well have been worn by a woman immigrant from that region." 21 Op. cir. in note I. 106 (nos. 472 and 811), Taf. 114, 4-5. 2􀂫 British Museum, reg. no. 1084'70. 29 B. Nerman, Die Volkerwanderungszeit Gotlands (Stockholm, 1935), Taf. 2, 11; 4, 24; 5, 29-34. 30 Op. cit. in note 17, 103. 31 Op. cit. in note 2, 322-3. n Op. cit. in note 17, 100. " Op. cit. in note 1, Abb. 4 and 12. l-1 I am most grateful to Mrs. L. Webster and to Mr D. Kidd for reading the original draft of this note and for their helpful criticism and advice; also to Mr K. Reedie for asking me to publish this brooch. Too late for proper inclusion in this paper, Mr Kidd has kindly drawn my attention to a group of Dutch and German bird-headed brooches related in form to the Faversham one (Fig. lb) in an article by J. Werner, 'Biigelfibeln des 6, Jahrhunderts aus Domburg, Zeeland', Berichten van de Rijksdienst voor het Oudhtidkundig Bodemonderzoek, vi (1955), 7S-7. 65